diff options
9852 files changed, 1295995 insertions, 420064 deletions
diff --git a/.appveyor.yml b/.appveyor.yml new file mode 100644 index 0000000000..05ff204541 --- /dev/null +++ b/.appveyor.yml @@ -0,0 +1,134 @@ +--- +version: '{build}' +init: + - git config --global user.name git + - git config --global user.email svn-admin@ruby-lang.org + - git config --global core.autocrlf false + - git config --global core.eol lf + - git config --global advice.detachedHead 0 +shallow_clone: true +clone_depth: 10 +platform: + - x64 +skip_commits: + message: /\[DOC\]/ + files: + - doc/* + - '**/*.md' + - '**/*.rdoc' + - '**/.document' + - '**/*.[1-8]' + - '**/*.ronn' +environment: + ruby_version: "24-%Platform%" + matrix: + # Test only the oldest supported version because AppVeyor is unstable, its concurrency + # is limited, and compatibility issues that happen only in newer versions are rare. + # You may test some other stuff on GitHub Actions instead. + - build: vs + vs: 120 # Visual Studio 2013 + ssl: OpenSSL-v111 + # The worker image name. This is NOT the Visual Studio version we're using here. + APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015 + GEMS_FOR_TEST: "" + RELINE_TEST_ENCODING: "UTF-8" +cache: + - c:\Tools\vcpkg\installed\ +for: +- + matrix: + only: + - build: vs + install: + - ver + - chcp + - SET BITS=%Platform:x86=32% + - SET BITS=%BITS:x=% + - SET OPENSSL_DIR=C:\%ssl%-Win%BITS% + - cd C:\Tools\vcpkg + - git pull -q + - .\bootstrap-vcpkg.bat + - ps: Start-FileDownload 'https://github.com/microsoft/vcpkg-tool/releases/download/2023-08-09/vcpkg.exe' -FileName 'C:\Tools\vcpkg\vcpkg.exe' + - cd %APPVEYOR_BUILD_FOLDER% + - vcpkg --triplet %Platform%-windows install --x-use-aria2 libffi libyaml readline zlib + - CALL SET vcvars=%%^VS%VS%COMNTOOLS^%%..\..\VC\vcvarsall.bat + - SET vcvars + - '"%vcvars%" %Platform:x64=amd64%' + - SET ruby_path=C:\Ruby%ruby_version:-x86=% + - SET PATH=\usr\local\bin;%ruby_path%\bin;%PATH%;C:\msys64\mingw64\bin;C:\msys64\usr\bin + - ruby --version + - 'cl' + - echo> Makefile srcdir=. + - echo>> Makefile MSC_VER=0 + - echo>> Makefile RT=none + - echo>> Makefile RT_VER=0 + - echo>> Makefile BUILTIN_ENCOBJS=nul + - type win32\Makefile.sub >> Makefile + - nmake %mflags% up VCSUP="echo Update OK" + - nmake %mflags% extract-extlibs + - del Makefile + - mkdir \usr\local\bin + - mkdir \usr\local\include + - mkdir \usr\local\lib + - for %%I in (%OPENSSL_DIR%\*.dll) do mklink /h \usr\local\bin\%%~nxI %%I + - for %%I in (c:\Tools\vcpkg\installed\%Platform%-windows\bin\*.dll) do ( + if not %%~nI == readline mklink \usr\local\bin\%%~nxI %%I + ) + - attrib +r /s /d + - mkdir %Platform%-mswin_%vs% + build_script: + - set HAVE_GIT=no + - cd %APPVEYOR_BUILD_FOLDER% + - cd %Platform%-mswin_%vs% + - >- + ..\win32\configure.bat + --with-opt-dir="/usr/local;c:/Tools/vcpkg/installed/%Platform%-windows" + --with-openssl-dir=%OPENSSL_DIR:\=/% + - nmake -l + - nmake install-nodoc + - \usr\bin\ruby -v -e "p :locale => Encoding.find('locale'), :filesystem => Encoding.find('filesystem')" + - if not "%GEMS_FOR_TEST%" == "" \usr\bin\gem install --no-document %GEMS_FOR_TEST% + - \usr\bin\ruby -ropenssl -e "puts 'Build ' + OpenSSL::OPENSSL_VERSION, 'Runtime ' + OpenSSL::OPENSSL_LIBRARY_VERSION" + test_script: + - set /a JOBS=%NUMBER_OF_PROCESSORS% + - nmake -l "TESTOPTS=-v -q" btest + - nmake -l "TESTOPTS=-v -q" test-basic + - >- + nmake -l "TESTOPTS=--timeout-scale=3.0 + --excludes=../test/excludes/_appveyor -j%JOBS% + --exclude win32ole + --exclude test_bignum + --exclude test_syntax + --exclude test_open-uri + --exclude test_bundled_ca + " test-all + # separately execute tests without -j which may crash worker with -j. + - >- + nmake -l + "TESTOPTS=--timeout-scale=3.0 --excludes=../test/excludes/_appveyor" + TESTS=" + ../test/win32ole + ../test/ruby/test_bignum.rb + ../test/ruby/test_syntax.rb + ../test/open-uri/test_open-uri.rb + ../test/rubygems/test_bundled_ca.rb + " test-all + - nmake -l test-spec MSPECOPT=-fs # not using `-j` because sometimes `mspec -j` silently dies on Windows +notifications: + - provider: Webhook + method: POST + url: + secure: CcFlJNDJ/a6to7u3Z4Fnz6dScEPNx7hTha2GkSRlV+1U6dqmxY/7uBcLXYb9gR3jfQk6w+2o/HrjNAyXMNGU/JOka3s2WRI4VKitzM+lQ08owvJIh0R7LxrGH0J2e81U # ruby-lang slack: ruby/simpler-alerts-bot + body: >- + {{^isPullRequest}} + { + "ci": "AppVeyor CI", + "env": "Visual Studio 2013", + "url": "{{buildUrl}}", + "commit": "{{commitId}}", + "branch": "{{branch}}" + } + {{/isPullRequest}} + on_build_success: false + on_build_failure: true + on_build_status_changed: false diff --git a/.dir-locals.el b/.dir-locals.el new file mode 100644 index 0000000000..8f4b96445c --- /dev/null +++ b/.dir-locals.el @@ -0,0 +1,39 @@ +;; Copyright (c) 2018 Urabe, Shyouhei. All rights reserved. +;; +;; This file is a part of the programming language Ruby. Permission is hereby +;; granted, to either redistribute and/or modify this file, provided that the +;; conditions mentioned in the file COPYING are met. Consult the file for +;; details. + +((nil . + ((indent-tabs-mode . nil) + (require-final-newline . t) + (tab-width . 8) + (show-trailing-whitespace . t) + (whitespace-line-column . 80))) ;; See also [Misc #12277] + + ;; (bat-mode . ((buffer-file-coding-system . utf-8-dos))) + + (ruby-mode . ((ruby-indent-level . 2))) + + (rdoc-mode . ((fill-column . 74))) + + (yaml-mode . ((yaml-indent-offset . 2))) + + (makefile-mode . ((indent-tabs-mode . t))) + + (c-mode . ((c-file-style . "ruby"))) + + (c++-mode . ((c-file-style . "ruby"))) + + (change-log-mode . + ((buffer-file-coding-system . us-ascii) + (indent-tabs-mode . t) + (change-log-indent-text . 2) + (add-log-time-format . (lambda (&optional x y) + (let* ((time (or x (current-time))) + (system-time-locale "C") + (diff (+ (cadr time) 32400)) + (lo (% diff 65536)) + (hi (+ (car time) (/ diff 65536)))) + (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t))))))) @@ -9,9 +9,28 @@ # prelude prelude.rb - rbconfig.rb +array.rb +ast.rb +dir.rb +gc.rb +io.rb +kernel.rb +marshal.rb +mjit.rb +numeric.rb +nilclass.rb +pack.rb +ractor.rb +string.rb +symbol.rb +timev.rb +thread_sync.rb +trace_point.rb +warning.rb +yjit.rb + # the lib/ directory (which has its own .document file) lib @@ -19,9 +38,17 @@ lib ext # rdoc files -NEWS +NEWS.md README.md README.ja.md +COPYING +COPYING.ja + +LEGAL + +# win32/README.win32 linked from README.md +win32 + doc diff --git a/.editorconfig b/.editorconfig index 49cc692091..4633a7acae 100644 --- a/.editorconfig +++ b/.editorconfig @@ -3,7 +3,7 @@ root = true [*] end_of_line = lf indent_size = 4 -indent_style = tab +indent_style = space insert_final_newline = true tab_width = 8 trim_trailing_whitespace = true @@ -13,12 +13,18 @@ end_of_line = crlf [*.gemspec] indent_size = 2 -indent_style = space [*.rb] indent_size = 2 -indent_style = space [*.yml] indent_size = 2 -indent_style = space + +[{*[Mm]akefile*,*.mak,*.mk,depend}] +indent_style = tab + +[enc/*] +indent_size = 2 + +[reg*.[ch]] +indent_size = 2 @@ -1,3 +1,5 @@ +set startup-with-shell off + define hook-run set $color_type = 0 set $color_highlite = 0 @@ -134,7 +136,7 @@ define rp printf "%sT_ARRAY%s: len=%ld ", $color_type, $color_end, $len if ($flags & RUBY_FL_USER2) printf "(shared) shared=" - output/x ((struct RArray*)($arg0))->as.heap.aux.shared + output/x ((struct RArray*)($arg0))->as.heap.aux.shared_root printf " " else printf "(ownership) capa=%ld ", ((struct RArray*)($arg0))->as.heap.aux.capa @@ -154,8 +156,12 @@ define rp else if ($flags & RUBY_T_MASK) == RUBY_T_HASH printf "%sT_HASH%s: ", $color_type, $color_end, - if ((struct RHash *)($arg0))->ntbl - printf "len=%ld ", ((struct RHash *)($arg0))->ntbl->num_entries + if (((struct RHash *)($arg0))->basic.flags & RHASH_ST_TABLE_FLAG) + printf "st len=%ld ", ((struct RHash *)($arg0))->as.st->num_entries + else + printf "li len=%ld bound=%ld ", \ + ((((struct RHash *)($arg0))->basic.flags & RHASH_AR_TABLE_SIZE_MASK) >> RHASH_AR_TABLE_SIZE_SHIFT), \ + ((((struct RHash *)($arg0))->basic.flags & RHASH_AR_TABLE_BOUND_MASK) >> RHASH_AR_TABLE_BOUND_SHIFT) end print (struct RHash *)($arg0) else @@ -259,6 +265,10 @@ define rp printf "%sT_ZOMBIE%s: ", $color_type, $color_end print (struct RData *)($arg0) else + if ($flags & RUBY_T_MASK) == RUBY_T_MOVED + printf "%sT_MOVED%s: ", $color_type, $color_end + print *(struct RMoved *)$arg0 + else printf "%sunknown%s: ", $color_type, $color_end print (struct RBasic *)($arg0) end @@ -294,6 +304,7 @@ define rp end end end + end end document rp Print a Ruby's VALUE. @@ -533,13 +544,13 @@ end define rp_class printf "(struct RClass *) %p", (void*)$arg0 - if ((struct RClass *)($arg0))->ptr.origin_ != $arg0 - printf " -> %p", ((struct RClass *)($arg0))->ptr.origin_ + if RCLASS_ORIGIN((struct RClass *)($arg0)) != $arg0 + printf " -> %p", RCLASS_ORIGIN((struct RClass *)($arg0)) end printf "\n" rb_classname $arg0 print/x *(struct RClass *)($arg0) - print *((struct RClass *)($arg0))->ptr + print *RCLASS_EXT((struct RClass *)($arg0)) end document rp_class Print the content of a Class/Module. @@ -698,12 +709,6 @@ define nd_cval rp ($arg0).u3.value end - -define nd_cnt - printf "%su3.cnt%s: ", $color_highlite, $color_end - p ($arg0).u3.cnt -end - define nd_tbl printf "%su1.tbl%s: ", $color_highlite, $color_end p ($arg0).u1.tbl @@ -742,12 +747,6 @@ define nd_lit rp ($arg0).u1.value end - -define nd_frml - printf "%su1.node%s: ", $color_highlite, $color_end - rp ($arg0).u1.node -end - define nd_rest printf "%su2.argc%s: ", $color_highlite, $color_end p ($arg0).u2.argc @@ -774,12 +773,6 @@ define nd_args rp ($arg0).u3.node end - -define nd_noex - printf "%su1.id%s: ", $color_highlite, $color_end - p ($arg0).u1.id -end - define nd_defn printf "%su3.node%s: ", $color_highlite, $color_end rp ($arg0).u3.node @@ -797,17 +790,6 @@ define nd_new end -define nd_cfnc - printf "%su1.cfunc%s: ", $color_highlite, $color_end - p ($arg0).u1.cfunc -end - -define nd_argc - printf "%su2.argc%s: ", $color_highlite, $color_end - p ($arg0).u2.argc -end - - define nd_cname printf "%su1.id%s: ", $color_highlite, $color_end p ($arg0).u1.id @@ -997,8 +979,8 @@ end define rb_ps_vm print $ps_vm = (rb_vm_t*)$arg0 - set $ps_thread_ln = $ps_vm->living_threads.n.next - set $ps_thread_ln_last = $ps_vm->living_threads.n.prev + set $ps_thread_ln = $ps_vm->ractor.main_ractor.threads.set.n.next + set $ps_thread_ln_last = $ps_vm->ractor.main_ractor.threads.set.n.prev while 1 set $ps_thread_th = (rb_thread_t *)$ps_thread_ln set $ps_thread = (VALUE)($ps_thread_th->self) @@ -1021,25 +1003,55 @@ define print_lineno set $pos = $pos - 1 end - set $i = 0 + set $index = 0 set $size = $iseq->body->insns_info.size set $table = $iseq->body->insns_info.body set $positions = $iseq->body->insns_info.positions #printf "size: %d\n", $size if $size == 0 else - set $i = 1 - while $i < $size - #printf "table[%d]: position: %d, line: %d, pos: %d\n", $i, $positions[$i], $table[$i].line_no, $pos - if $positions[$i] > $pos - loop_break + if $size == 1 + printf "%d", $table[0].line_no + else + if $positions + # get_insn_info_linear_search + set $index = 1 + while $index < $size + #printf "table[%d]: position: %d, line: %d, pos: %d\n", $i, $positions[$i], $table[$i].line_no, $pos + if $positions[$index] > $pos + loop_break + end + set $index = $index + 1 + if $positions[$index] == $pos + loop_break + end end - set $i = $i + 1 - if $positions[$i] == $pos - loop_break + else + # get_insn_info_succinct_bitvector + set $sd = $iseq->body->insns_info.succ_index_table + set $immediate_table_size = sizeof($sd->imm_part) / sizeof(uint64_t) * 9 + if $pos < $immediate_table_size + set $i = $pos / 9 + set $j = $pos % 9 + set $index = ((int)($sd->imm_part[$i] >> ($j * 7))) & 0x7f + else + set $block_index = ($pos - $immediate_table_size) / 512 + set $block = &$sd->succ_part[$block_index] + set $block_bit_index = ($pos - $immediate_table_size) % 512 + set $small_block_index = $block_bit_index / 64 + set $small_block_popcount = $small_block_index == 0 ? 0 : (((int)($block->small_block_ranks >> (($small_block_index - 1) * 9))) & 0x1ff) + set $x = $block->bits[$small_block_index] << (63 - $block_bit_index % 64) + set $x = ($x & 0x5555555555555555) + ($x >> 1 & 0x5555555555555555) + set $x = ($x & 0x3333333333333333) + ($x >> 2 & 0x3333333333333333) + set $x = ($x & 0x0707070707070707) + ($x >> 4 & 0x0707070707070707) + set $x = ($x & 0x001f001f001f001f) + ($x >> 8 & 0x001f001f001f001f) + set $x = ($x & 0x0000003f0000003f) + ($x >>16 & 0x0000003f0000003f) + set $popcnt = ($x & 0x7f) + ($x >>32 & 0x7f) + set $index = $block->rank + $small_block_popcount + $popcnt end end - printf "%d", $table[$i-1].line_no + printf "%d", $table[$index-1].line_no + end end end @@ -1066,9 +1078,9 @@ define print_id else set $serial = (rb_id_serial_t)$id end - if $serial && $serial <= global_symbols.last_id + if $serial && $serial <= ruby_global_symbols.last_id set $idx = $serial / ID_ENTRY_UNIT - set $ids = (struct RArray *)global_symbols.ids + set $ids = (struct RArray *)ruby_global_symbols.ids set $flags = $ids->basic.flags if ($flags & RUBY_FL_USER1) set $idsptr = $ids->as.ary @@ -1090,11 +1102,11 @@ define print_id set $arylen = $ary->as.heap.len end set $result = $aryptr[($serial % ID_ENTRY_UNIT) * ID_ENTRY_SIZE + $t] - if $result != RUBY_Qnil + if $result != RUBY_Qnil print_string $result - else - echo undef - end + else + echo undef + end end end end @@ -1267,7 +1279,7 @@ document rb_count_objects Counts all objects grouped by type. end -# Details: https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/MachineInstructionsTraceWithGDB +# Details: https://bugs.ruby-lang.org/projects/ruby-master/wiki/MachineInstructionsTraceWithGDB define trace_machine_instructions set logging on set height 0 @@ -1306,3 +1318,33 @@ define dump_node ((struct RString*)$str)->as.heap.ptr : \ ((struct RString*)$str)->as.ary) end + +define print_flags + printf "RUBY_FL_WB_PROTECTED: %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_WB_PROTECTED ? "1" : "0" + printf "RUBY_FL_PROMOTED0 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_PROMOTED0 ? "1" : "0" + printf "RUBY_FL_PROMOTED1 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_PROMOTED1 ? "1" : "0" + printf "RUBY_FL_FINALIZE : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_FINALIZE ? "1" : "0" + printf "RUBY_FL_SHAREABLE : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_SHAREABLE ? "1" : "0" + printf "RUBY_FL_EXIVAR : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_EXIVAR ? "1" : "0" + printf "RUBY_FL_FREEZE : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_FREEZE ? "1" : "0" + + printf "RUBY_FL_USER0 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER0 ? "1" : "0" + printf "RUBY_FL_USER1 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER1 ? "1" : "0" + printf "RUBY_FL_USER2 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER2 ? "1" : "0" + printf "RUBY_FL_USER3 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER3 ? "1" : "0" + printf "RUBY_FL_USER4 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER4 ? "1" : "0" + printf "RUBY_FL_USER5 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER5 ? "1" : "0" + printf "RUBY_FL_USER6 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER6 ? "1" : "0" + printf "RUBY_FL_USER7 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER7 ? "1" : "0" + printf "RUBY_FL_USER8 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER8 ? "1" : "0" + printf "RUBY_FL_USER9 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER9 ? "1" : "0" + printf "RUBY_FL_USER10 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER10 ? "1" : "0" + printf "RUBY_FL_USER11 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER11 ? "1" : "0" + printf "RUBY_FL_USER12 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER12 ? "1" : "0" + printf "RUBY_FL_USER13 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER13 ? "1" : "0" + printf "RUBY_FL_USER14 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER14 ? "1" : "0" + printf "RUBY_FL_USER15 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER15 ? "1" : "0" + printf "RUBY_FL_USER16 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER16 ? "1" : "0" + printf "RUBY_FL_USER17 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER17 ? "1" : "0" + printf "RUBY_FL_USER18 : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER18 ? "1" : "0" +end diff --git a/.git-blame-ignore-revs b/.git-blame-ignore-revs new file mode 100644 index 0000000000..6c5eac5a0f --- /dev/null +++ b/.git-blame-ignore-revs @@ -0,0 +1,23 @@ +# This is a file used by GitHub to ignore the following commits on `git blame`. +# +# You can also do the same thing in your local repository with: +# $ git config --local blame.ignoreRevsFile .git-blame-ignore-revs + +# Expand tabs +5b21e94bebed90180d8ff63dad03b8b948361089 + +# Enable Style/StringLiterals cop for RubyGems/Bundler +d7ffd3fea402239b16833cc434404a7af82d44f3 + +# [ruby/digest] Revert tab-expansion in external files +48b09aae7ec5632209229dcc294dd0d75a93a17f +8a65cf3b61c60e4cb886f59a73ff6db44364bfa9 +39dc9f9093901d40d2998653948d5da38b18ee2c + +# [ruby/io-nonblock] Revert tab expansion +f28287d34c03f472ffe90ea262bdde9affd4b965 +0d842fecb4f75ab3b1d4097ebdb8e88f51558041 +4ba2c66761d6a293abdfba409241d31063cefd62 + +# Make benchmark indentation consistent +fc4acf8cae82e5196186d3278d831f2438479d91 diff --git a/.gitattributes b/.gitattributes index 6ca2f89462..d0c2d266b4 100644 --- a/.gitattributes +++ b/.gitattributes @@ -4,3 +4,5 @@ bin svn-properties=svn:ignore=ruby bin/* diff=ruby tool/update-deps diff=ruby tool/make-snapshot diff=ruby +tool/format-release diff=ruby +tool/leaked-globals diff=ruby diff --git a/.github/SECURITY.md b/.github/SECURITY.md new file mode 100644 index 0000000000..56baa29942 --- /dev/null +++ b/.github/SECURITY.md @@ -0,0 +1,9 @@ +# Security Policy + +## Supported Versions + +See <https://www.ruby-lang.org/en/downloads/branches/>. + +## Reporting a Vulnerability + +See <https://www.ruby-lang.org/en/security/>. diff --git a/.github/codeql/codeql-config.yml b/.github/codeql/codeql-config.yml new file mode 100644 index 0000000000..91f82b842b --- /dev/null +++ b/.github/codeql/codeql-config.yml @@ -0,0 +1,3 @@ +name: "CodeQL config for the Ruby language" + +languages: cpp diff --git a/.github/dependabot.yml b/.github/dependabot.yml new file mode 100644 index 0000000000..bc63aca35b --- /dev/null +++ b/.github/dependabot.yml @@ -0,0 +1,6 @@ +version: 2 +updates: + - package-ecosystem: 'github-actions' + directory: '/' + schedule: + interval: 'monthly' diff --git a/.github/workflows/baseruby.yml b/.github/workflows/baseruby.yml new file mode 100644 index 0000000000..ebaafe3bf0 --- /dev/null +++ b/.github/workflows/baseruby.yml @@ -0,0 +1,80 @@ +name: BASERUBY Check + +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + baseruby: + name: BASERUBY + runs-on: ubuntu-22.04 + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + strategy: + matrix: + ruby: + - ruby-2.2 +# - ruby-2.3 +# - ruby-2.4 +# - ruby-2.5 +# - ruby-2.6 +# - ruby-2.7 + - ruby-3.0 + - ruby-3.1 + + steps: + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: .downloaded-cache + key: downloaded-cache + - uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0 + with: + ruby-version: ${{ matrix.ruby }} + bundler: none + - run: echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + - run: sudo apt-get install build-essential autoconf bison libyaml-dev + - run: ./autogen.sh + - run: ./configure --disable-install-doc + - run: make common-srcs + - run: make incs + - run: make all + - run: make test + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / BASERUBY @ ${{ matrix.ruby }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} diff --git a/.github/workflows/bundled_gems.yml b/.github/workflows/bundled_gems.yml new file mode 100644 index 0000000000..070c0fa1dd --- /dev/null +++ b/.github/workflows/bundled_gems.yml @@ -0,0 +1,166 @@ +name: bundled_gems + +on: + push: + branches: [ "master" ] + paths: + - '.github/workflows/bundled_gems.yml' + - 'gems/bundled_gems' + pull_request: + branches: [ "master" ] + paths: + - '.github/workflows/bundled_gems.yml' + - 'gems/bundled_gems' + merge_group: + branches: [ "master" ] + paths: + - '.github/workflows/bundled_gems.yml' + - 'gems/bundled_gems' + schedule: + - cron: '45 6 * * *' + workflow_dispatch: + +permissions: # added using https://github.com/step-security/secure-workflows + contents: read + +jobs: + update: + permissions: + contents: write # for Git to git push + if: ${{ github.event_name != 'schedule' || github.repository == 'ruby/ruby' }} + name: update ${{ github.workflow }} + runs-on: ubuntu-latest + steps: + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + + - name: Set ENV + run: | + echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + echo "TODAY=$(date +%F)" >> $GITHUB_ENV + + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: .downloaded-cache + key: downloaded-cache-${{ github.sha }} + restore-keys: | + downloaded-cache + + - name: Download previous gems list + run: | + data=bundled_gems.json + mkdir -p .downloaded-cache + ln -s .downloaded-cache/$data . + curl -O -R -z ./$data https://stdgems.org/$data + + - name: Update bundled gems list + run: | + ruby -i~ tool/update-bundled_gems.rb gems/bundled_gems + + - name: Maintain updated gems list in NEWS + run: | + #!ruby + require 'json' + news = File.read("NEWS.md") + prev = news[/since the \*+(\d+\.\d+\.\d+)\*+/, 1] + prevs = [prev, prev.sub(/\.\d+\z/, '')] + %W[bundled].each do |type| + last = JSON.parse(File.read("#{type}_gems.json"))['gems'].filter_map do |g| + v = g['versions'].values_at(*prevs).compact.first + g = g['gem'] + g = 'RubyGems' if g == 'rubygems' + [g, v] if v + end.to_h + changed = File.foreach("gems/#{type}_gems").filter_map do |l| + next if l.start_with?("#") + g, v = l.split(" ", 3) + [g, v] unless last[g] == v + end + changed, added = changed.partition {|g, _| last[g]} + news.sub!(/^\*( +)The following #{type} gems? are updated\.\n+\K(?: \1\*( +).*\n)*/) do + mark = "#{$1} *#{$2}" + changed.map {|g, v|"#{mark}#{g} #{v}\n"}.join("") + end or next + news.sub!(/^\*( +)The following default gems are now bundled gems\.\n+\K(?: \1\*( +).*\n)*/) do + mark = "#{$1} *#{$2}" + added.map {|g, v|"#{mark}#{g} #{v}\n"}.join("") + end or next unless added.empty? + File.write("NEWS.md", news) + end + shell: ruby {0} + + - name: Check diffs + id: diff + run: | + git add -- NEWS.md + git diff --no-ext-diff --ignore-submodules --quiet -- gems/bundled_gems + continue-on-error: true + + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y build-essential libssl-dev libyaml-dev libreadline6-dev zlib1g-dev libncurses5-dev libffi-dev bison autoconf ruby + if: ${{ steps.diff.outcome == 'failure' }} + + - name: Build + run: | + ./autogen.sh + ./configure -C --disable-install-doc + make + if: ${{ steps.diff.outcome == 'failure' }} + + - name: Prepare bundled gems + run: | + make -s prepare-gems + if: ${{ steps.diff.outcome == 'failure' }} + + - name: Test bundled gems + run: | + make -s test-bundled-gems + git add -- gems/bundled_gems + timeout-minutes: 30 + env: + RUBY_TESTOPTS: "-q --tty=no" + TEST_BUNDLED_GEMS_ALLOW_FAILURES: "" + if: ${{ steps.diff.outcome == 'failure' }} + + - name: Show diffs + id: show + run: | + git diff --cached --color --no-ext-diff --ignore-submodules --exit-code -- + continue-on-error: true + + - name: Commit + run: | + git pull --ff-only origin ${GITHUB_REF#refs/heads/} + message="Update bundled gems list at " + if [ ${{ steps.diff.outcome }} = success ]; then + git commit --message="${message}${GITHUB_SHA:0:30} [ci skip]" + else + git commit --message="${message}${TODAY}" + fi + git push origin ${GITHUB_REF#refs/heads/} + env: + EMAIL: svn-admin@ruby-lang.org + GIT_AUTHOR_NAME: git + GIT_COMMITTER_NAME: git + if: ${{ github.repository == 'ruby/ruby' && !startsWith(github.event_name, 'pull') && steps.show.outcome == 'failure' }} + + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / update", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} diff --git a/.github/workflows/check_dependencies.yml b/.github/workflows/check_dependencies.yml new file mode 100644 index 0000000000..79b2916feb --- /dev/null +++ b/.github/workflows/check_dependencies.yml @@ -0,0 +1,78 @@ +name: Check Dependencies +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + update-deps: + strategy: + matrix: + os: [ubuntu-22.04] + fail-fast: true + runs-on: ${{ matrix.os }} + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y build-essential libssl-dev libyaml-dev libreadline6-dev zlib1g-dev libncurses5-dev libffi-dev bison autoconf ruby + if: ${{ contains(matrix.os, 'ubuntu') }} + - name: Install libraries + run: | + brew install gmp libffi openssl@1.1 zlib autoconf automake libtool readline + if: ${{ contains(matrix.os, 'macos') }} + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: .downloaded-cache + key: downloaded-cache + - run: ./autogen.sh + - name: Run configure + run: ./configure -C --disable-install-doc --disable-rubygems --with-gcc 'optflags=-O0' 'debugflags=-save-temps=obj -g' + - run: make all golf + - run: ruby tool/update-deps --fix + - run: git diff --no-ext-diff --ignore-submodules --exit-code + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ matrix.os }} / Dependencies need to update", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} diff --git a/.github/workflows/codeql-analysis.yml b/.github/workflows/codeql-analysis.yml new file mode 100644 index 0000000000..8dba76fbe2 --- /dev/null +++ b/.github/workflows/codeql-analysis.yml @@ -0,0 +1,75 @@ +name: "Code scanning - action" + +on: + # push: + # paths-ignore: + # - 'doc/**' + # - '**/man' + # - '**.md' + # - '**.rdoc' + # - '**/.document' + # pull_request: + # paths-ignore: + # - 'doc/**' + # - '**/man' + # - '**.md' + # - '**.rdoc' + # - '**/.document' + schedule: + - cron: '0 12 * * *' + workflow_dispatch: + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: # added using https://github.com/step-security/secure-workflows + contents: read + +jobs: + CodeQL-Build: + + # CodeQL runs on ubuntu-latest and windows-latest + permissions: + actions: read # for github/codeql-action/init to get workflow details + contents: read # for actions/checkout to fetch code + security-events: write # for github/codeql-action/autobuild to send a status report + runs-on: ubuntu-latest + # CodeQL fails to run pull requests from dependabot due to missing write access to upload results. + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') && github.event.head_commit.pusher.name != 'dependabot[bot]' }} + + env: + enable_install_doc: no + + steps: + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y build-essential libssl-dev libyaml-dev libreadline6-dev zlib1g-dev libncurses5-dev libffi-dev bison autoconf ruby + + - name: Checkout repository + uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: .downloaded-cache + key: downloaded-cache + + - name: Remove an obsolete rubygems vendored file + run: sudo rm /usr/lib/ruby/vendor_ruby/rubygems/defaults/operating_system.rb + + - name: Initialize CodeQL + uses: github/codeql-action/init@959cbb7472c4d4ad70cdfe6f4976053fe48ab394 # v2.1.37 + with: + config-file: ./.github/codeql/codeql-config.yml + trap-caching: false + + - name: Set ENV + run: echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + + - name: Autobuild + uses: github/codeql-action/autobuild@959cbb7472c4d4ad70cdfe6f4976053fe48ab394 # v2.1.37 + + - name: Perform CodeQL Analysis + uses: github/codeql-action/analyze@959cbb7472c4d4ad70cdfe6f4976053fe48ab394 # v2.1.37 diff --git a/.github/workflows/compilers.yml b/.github/workflows/compilers.yml new file mode 100644 index 0000000000..caf12cc0f4 --- /dev/null +++ b/.github/workflows/compilers.yml @@ -0,0 +1,281 @@ +name: Compilations + +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +# GitHub actions does not support YAML anchors. This creative use of +# environment variables (plus the "echo $GITHUB_ENV" hack) is to reroute that +# restriction. +env: + default_cc: clang-15 + append_cc: '' + + # -O1 is faster than -O3 in our tests... Majority of time are consumed trying + # to optimize binaries. Also GitHub Actions run on relatively modern CPUs + # compared to, say, GCC 4 or Clang 3. We don't specify `-march=native` + # because compilers tend not understand what the CPU is. + optflags: '-O1' + + # -g0 disables backtraces when SEGV. Do not set that. + debugflags: '-ggdb3' + + default_configure: >- + --enable-debug-env + --disable-install-doc + --with-ext=-test-/cxxanyargs,+ + append_configure: >- + --without-valgrind + --without-jemalloc + --without-gmp + + UPDATE_UNICODE: >- + UNICODE_FILES=. + UNICODE_PROPERTY_FILES=. + UNICODE_AUXILIARY_FILES=. + UNICODE_EMOJI_FILES=. + CONFIGURE_TTY: never + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + RUBY_DEBUG: ci rgengc + RUBY_TESTOPTS: >- + -q + --color=always + --tty=no + +permissions: + contents: read + +jobs: + compile: + strategy: + fail-fast: false + matrix: + env: + - {} + entry: + - { name: gcc-12, env: { default_cc: gcc-12 } } + - { name: gcc-11, env: { default_cc: gcc-11 } } + - { name: gcc-10, env: { default_cc: gcc-10 } } + - { name: gcc-9, env: { default_cc: gcc-9 } } + - { name: gcc-8, env: { default_cc: gcc-8 } } + - { name: gcc-7, env: { default_cc: gcc-7 } } + - name: 'gcc-13 LTO' + container: gcc-13 + env: + default_cc: 'gcc-13 -flto=auto -ffat-lto-objects -Werror=lto-type-mismatch' + optflags: '-O2' + shared: disable + # check: true + - { name: clang-16, env: { default_cc: clang-16 } } + - { name: clang-15, env: { default_cc: clang-15 } } + - { name: clang-14, env: { default_cc: clang-14 } } + - { name: clang-13, env: { default_cc: clang-13 } } + - { name: clang-12, env: { default_cc: clang-12 } } + - { name: clang-11, env: { default_cc: clang-11 } } + - { name: clang-10, env: { default_cc: clang-10 } } + # llvm-objcopy<=9 doesn't have --wildcard. It compiles, but leaves Rust symbols in libyjit.o. + - { name: clang-9, env: { default_cc: clang-9, append_configure: '--disable-yjit' } } + - { name: clang-8, env: { default_cc: clang-8, append_configure: '--disable-yjit' } } + - { name: clang-7, env: { default_cc: clang-7, append_configure: '--disable-yjit' } } + - { name: clang-6.0, env: { default_cc: clang-6.0, append_configure: '--disable-yjit' } } + - name: 'clang-16 LTO' + container: clang-16 + env: + default_cc: 'clang-16 -flto=auto' + optflags: '-O2' + shared: disable + # check: true + +# - { name: aarch64-linux-gnu, crosshost: aarch64-linux-gnu, container: crossbuild-essential-arm64 } +# - { name: arm-linux-gnueabi, crosshost: arm-linux-gnueabi } +# - { name: arm-linux-gnueabihf, crosshost: arm-linux-gnueabihf } +# - { name: i686-w64-mingw32, crosshost: i686-w64-mingw32 } +# - { name: powerpc-linux-gnu, crosshost: powerpc-linux-gnu } +# - { name: powerpc64le-linux-gnu, crosshost: powerpc64le-linux-gnu, container: crossbuild-essential-ppc64el } +# - { name: s390x-linux-gnu, crosshost: s390x-linux-gnu, container: crossbuild-essential-s390x } +# - { name: x86_64-w64-mingw32, crosshost: x86_64-w64-mingw32, container: mingw-w64 } + + # -Wno-strict-prototypes is necessary with current clang-15 since + # older autoconf generate functions without prototype and -pedantic + # now implies strict-prototypes. Disabling the error but leaving the + # warning generates a lot of noise from use of ANYARGS in + # rb_define_method() and friends. + # See: https://github.com/llvm/llvm-project/commit/11da1b53d8cd3507959022cd790d5a7ad4573d94 + - { name: c99, env: { append_cc: '-std=c99 -Werror=pedantic -pedantic-errors -Wno-strict-prototypes' } } +# - { name: c11, env: { append_cc: '-std=c11 -Werror=pedantic -pedantic-errors -Wno-strict-prototypes' } } +# - { name: c17, env: { append_cc: '-std=c17 -Werror=pedantic -pedantic-errors -Wno-strict-prototypes' } } + - { name: c2x, env: { append_cc: '-std=c2x -Werror=pedantic -pedantic-errors -Wno-strict-prototypes' } } + - { name: c++98, env: { CXXFLAGS: '-std=c++98 -Werror=pedantic -pedantic-errors -Wno-c++11-long-long' } } +# - { name: c++11, env: { CXXFLAGS: '-std=c++11 -Werror=pedantic -pedantic-errors -Wno-c++11-long-long' } } +# - { name: c++14, env: { CXXFLAGS: '-std=c++14 -Werror=pedantic -pedantic-errors -Wno-c++11-long-long' } } +# - { name: c++17, env: { CXXFLAGS: '-std=c++17 -Werror=pedantic -pedantic-errors -Wno-c++11-long-long' } } + - { name: c++2a, env: { CXXFLAGS: '-std=c++2a -Werror=pedantic -pedantic-errors -Wno-c++11-long-long' } } + + - { name: '-O0', env: { optflags: '-O0 -march=x86-64 -mtune=generic' } } +# - { name: '-O3', env: { optflags: '-O3 -march=x86-64 -mtune=generic' }, check: true } + + - { name: gmp, env: { append_configure: '--with-gmp' } } + - { name: jemalloc, env: { append_configure: '--with-jemalloc' } } + - { name: valgrind, env: { append_configure: '--with-valgrind' } } + - { name: 'coroutine=ucontext', env: { append_configure: '--with-coroutine=ucontext' } } + - { name: 'coroutine=pthread', env: { append_configure: '--with-coroutine=pthread' } } + - { name: disable-jit-support, env: { append_configure: '--disable-jit-support' } } + - { name: disable-dln, env: { append_configure: '--disable-dln' } } + - { name: enable-mkmf-verbose, env: { append_configure: '--enable-mkmf-verbose' } } + - { name: disable-rubygems, env: { append_configure: '--disable-rubygems' } } + - { name: RUBY_DEVEL, env: { append_configure: '--enable-devel' } } + + - { name: OPT_THREADED_CODE=1, env: { cppflags: '-DOPT_THREADED_CODE=1' } } + - { name: OPT_THREADED_CODE=2, env: { cppflags: '-DOPT_THREADED_CODE=2' } } + - { name: OPT_THREADED_CODE=3, env: { cppflags: '-DOPT_THREADED_CODE=3' } } + + - { name: NDEBUG, env: { cppflags: '-DNDEBUG' } } + - { name: RUBY_DEBUG, env: { cppflags: '-DRUBY_DEBUG' } } +# - { name: ARRAY_DEBUG, env: { cppflags: '-DARRAY_DEBUG' } } +# - { name: BIGNUM_DEBUG, env: { cppflags: '-DBIGNUM_DEBUG' } } +# - { name: CCAN_LIST_DEBUG, env: { cppflags: '-DCCAN_LIST_DEBUG' } } +# - { name: CPDEBUG=-1, env: { cppflags: '-DCPDEBUG=-1' } } +# - { name: ENC_DEBUG, env: { cppflags: '-DENC_DEBUG' } } +# - { name: GC_DEBUG, env: { cppflags: '-DGC_DEBUG' } } +# - { name: HASH_DEBUG, env: { cppflags: '-DHASH_DEBUG' } } +# - { name: ID_TABLE_DEBUG, env: { cppflags: '-DID_TABLE_DEBUG' } } +# - { name: RGENGC_DEBUG=-1, env: { cppflags: '-DRGENGC_DEBUG=-1' } } +# - { name: SYMBOL_DEBUG, env: { cppflags: '-DSYMBOL_DEBUG' } } + +# - { name: RGENGC_CHECK_MODE, env: { cppflags: '-DRGENGC_CHECK_MODE' } } +# - { name: TRANSIENT_HEAP_CHECK_MODE, env: { cppflags: '-DTRANSIENT_HEAP_CHECK_MODE' } } +# - { name: VM_CHECK_MODE, env: { cppflags: '-DVM_CHECK_MODE' } } + + - { name: USE_EMBED_CI=0, env: { cppflags: '-DUSE_EMBED_CI=0' } } + - name: USE_FLONUM=0, + env: + cppflags: '-DUSE_FLONUM=0' + # yjit requires FLONUM for the pointer tagging scheme + append_configure: '--disable-yjit' +# - { name: USE_GC_MALLOC_OBJ_INFO_DETAILS, env: { cppflags: '-DUSE_GC_MALLOC_OBJ_INFO_DETAILS' } } + - { name: USE_LAZY_LOAD, env: { cppflags: '-DUSE_LAZY_LOAD' } } +# - { name: USE_RINCGC=0, env: { cppflags: '-DUSE_RINCGC=0' } } +# - { name: USE_SYMBOL_GC=0, env: { cppflags: '-DUSE_SYMBOL_GC=0' } } +# - { name: USE_THREAD_CACHE=0, env: { cppflags: '-DUSE_THREAD_CACHE=0' } } +# - { name: USE_TRANSIENT_HEAP=0, env: { cppflags: '-DUSE_TRANSIENT_HEAP=0' } } +# - { name: USE_RUBY_DEBUG_LOG=1, env: { cppflags: '-DUSE_RUBY_DEBUG_LOG=1' } } + - { name: USE_RVARGC=0, env: { cppflags: '-DUSE_RVARGC=0' } } +# - { name: USE_RVARGC=1, env: { cppflags: '-DUSE_RVARGC=1' } } +# - { name: USE_DEBUG_COUNTER, env: { cppflags: '-DUSE_DEBUG_COUNTER=1', RUBY_DEBUG_COUNTER_DISABLE: '1' } } + + - { name: DEBUG_FIND_TIME_NUMGUESS, env: { cppflags: '-DDEBUG_FIND_TIME_NUMGUESS' } } + - { name: DEBUG_INTEGER_PACK, env: { cppflags: '-DDEBUG_INTEGER_PACK' } } +# - { name: ENABLE_PATH_CHECK, env: { cppflags: '-DENABLE_PATH_CHECK' } } + + - { name: GC_DEBUG_STRESS_TO_CLASS, env: { cppflags: '-DGC_DEBUG_STRESS_TO_CLASS' } } +# - { name: GC_ENABLE_LAZY_SWEEP=0, env: { cppflags: '-DGC_ENABLE_LAZY_SWEEP=0' } } +# - { name: GC_PROFILE_DETAIL_MEMOTY, env: { cppflags: '-DGC_PROFILE_DETAIL_MEMOTY' } } +# - { name: GC_PROFILE_MORE_DETAIL, env: { cppflags: '-DGC_PROFILE_MORE_DETAIL' } } + +# - { name: CALC_EXACT_MALLOC_SIZE, env: { cppflags: '-DCALC_EXACT_MALLOC_SIZE' } } +# - { name: MALLOC_ALLOCATED_SIZE_CHECK, env: { cppflags: '-DMALLOC_ALLOCATED_SIZE_CHECK' } } + +# - { name: IBF_ISEQ_ENABLE_LOCAL_BUFFER, env: { cppflags: '-DIBF_ISEQ_ENABLE_LOCAL_BUFFER' } } + +# - { name: RGENGC_ESTIMATE_OLDMALLOC, env: { cppflags: '-DRGENGC_ESTIMATE_OLDMALLOC' } } +# - { name: RGENGC_FORCE_MAJOR_GC, env: { cppflags: '-DRGENGC_FORCE_MAJOR_GC' } } +# - { name: RGENGC_OBJ_INFO, env: { cppflags: '-DRGENGC_OBJ_INFO' } } +# - { name: RGENGC_OLD_NEWOBJ_CHECK, env: { cppflags: '-DRGENGC_OLD_NEWOBJ_CHECK' } } +# - { name: RGENGC_PROFILE, env: { cppflags: '-DRGENGC_PROFILE' } } + +# - { name: VM_DEBUG_BP_CHECK, env: { cppflags: '-DVM_DEBUG_BP_CHECK' } } +# - { name: VM_DEBUG_VERIFY_METHOD_CACHE, env: { cppflags: '-DVM_DEBUG_VERIFY_METHOD_CACHE' } } + + - { name: MJIT_FORCE_ENABLE, env: { cppflags: '-DMJIT_FORCE_ENABLE' } } + - { name: YJIT_FORCE_ENABLE, env: { cppflags: '-DYJIT_FORCE_ENABLE' } } + + name: ${{ matrix.entry.name }} + runs-on: ubuntu-latest + container: + image: ghcr.io/ruby/ruby-ci-image:${{ matrix.entry.container || matrix.entry.env.default_cc || 'clang-15' }} + options: --user root + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + env: ${{ matrix.entry.env || matrix.env }} + steps: + - run: id + working-directory: + - run: mkdir build + working-directory: + - name: setenv + run: | + echo "GNUMAKEFLAGS=-sj$((1 + $(nproc --all)))" >> $GITHUB_ENV + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - run: ./autogen.sh + working-directory: src + - name: Run configure + run: > + ../src/configure -C ${default_configure} ${append_configure} + --${{ + matrix.entry.crosshost && 'host' || 'with-gcc' + }}=${{ + matrix.entry.crosshost || '"${default_cc}${append_cc:+ $append_cc}"' + }} + --${{ matrix.entry.shared || 'enable' }}-shared + - run: make extract-extlibs + - run: make incs + - run: make showflags + - run: make + - run: make leaked-globals + - run: make test + - run: make install + if: ${{ matrix.entry.check }} + - run: make test-tool + if: ${{ matrix.entry.check }} + - run: make test-all TESTS='-- ruby -ext-' + if: ${{ matrix.entry.check }} + - run: make test-spec + env: + CHECK_LEAKS: true + if: ${{ matrix.entry.check }} + - run: make test-annocheck + if: ${{ matrix.entry.check && endsWith(matrix.entry.name, 'annocheck') }} + + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / ${{ matrix.entry.name }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/macos.yml b/.github/workflows/macos.yml new file mode 100644 index 0000000000..d8dc58b119 --- /dev/null +++ b/.github/workflows/macos.yml @@ -0,0 +1,113 @@ +name: macOS +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + make: + strategy: + matrix: + test_task: ["check"] # "test-bundler-parallel", "test-bundled-gems" + os: + - macos-13 + - macos-14 + - macos-15 + fail-fast: false + env: + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + runs-on: ${{ matrix.os }} + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Install libraries + run: | + brew install gmp libffi openssl@1.1 zlib autoconf automake libtool readline bison + working-directory: src + - name: Set ENV + run: | + echo "MAKEFLAGS=-j$((1 + $(sysctl -n hw.activecpu)))" >> $GITHUB_ENV + echo "PATH="/usr/local/opt/bison/bin:/opt/homebrew/opt/bison/bin:$PATH"" >> $GITHUB_ENV + - run: ./autogen.sh + working-directory: src + - name: Run configure + run: ../src/configure -C --disable-install-doc --with-openssl-dir=$(brew --prefix openssl@1.1) --with-readline-dir=$(brew --prefix readline) + - run: make incs + - run: make prepare-gems + if: ${{ matrix.test_task == 'test-bundled-gems' }} + - run: make + - run: make leaked-globals + if: ${{ matrix.test_task == 'check' }} + - name: make ${{ matrix.test_task }} + run: | + make -s ${{ matrix.test_task }} ${TESTS:+TESTS=`echo "$TESTS" | sed 's| |$$/ -n!/|g;s|^|-n!/|;s|$|$$/|'`} + timeout-minutes: 40 + env: + RUBY_TESTOPTS: "-q --tty=no" + TESTS: ${{ matrix.test_task == 'check' && matrix.skipped_tests || '' }} + TEST_BUNDLED_GEMS_ALLOW_FAILURES: "" + PRECHECK_BUNDLED_GEMS: "no" + - name: make skipped tests + run: | + make -s test-all TESTS=`echo "$TESTS" | sed 's| |$$/ -n/|g;s|^|-n/|;s|$|$$/|'` + env: + GNUMAKEFLAGS: "" + RUBY_TESTOPTS: "-v --tty=no" + TESTS: ${{ matrix.skipped_tests }} + PRECHECK_BUNDLED_GEMS: "no" + if: ${{ matrix.test_task == 'check' && matrix.skipped_tests != '' }} + continue-on-error: ${{ matrix.continue-on-skipped_tests || false }} + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ matrix.os }} / ${{ matrix.test_task }}${{ matrix.configure }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/mingw.yml b/.github/workflows/mingw.yml new file mode 100644 index 0000000000..0df917d3d8 --- /dev/null +++ b/.github/workflows/mingw.yml @@ -0,0 +1,179 @@ +name: MinGW +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +# Notes: +# Actions console encoding causes issues, see test-all & test-spec steps +# +jobs: + make: + runs-on: windows-2022 + name: ${{ github.workflow }} (${{ matrix.msystem }}) + env: + MSYSTEM: ${{ matrix.msystem }} + MSYS2_ARCH: x86_64 + CHOST: "x86_64-w64-mingw32" + CFLAGS: "-march=x86-64 -mtune=generic -O3 -pipe" + CXXFLAGS: "-march=x86-64 -mtune=generic -O3 -pipe" + CPPFLAGS: "-D_FORTIFY_SOURCE=2 -D__USE_MINGW_ANSI_STDIO=1 -DFD_SETSIZE=2048" + LDFLAGS: "-pipe" + UPDATE_UNICODE: "UNICODE_FILES=. UNICODE_PROPERTY_FILES=. UNICODE_AUXILIARY_FILES=. UNICODE_EMOJI_FILES=." + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + strategy: + matrix: + include: + # To mitigate flakiness of MinGW CI, we test only one runtime that newer MSYS2 uses. + - msystem: "UCRT64" + base_ruby: head + test_task: "check" + test-all-opts: "--name=!/TestObjSpace#test_reachable_objects_during_iteration/" + fail-fast: false + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: git config + run: | + git config --global core.autocrlf false + git config --global core.eol lf + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Set up Ruby & MSYS2 + uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0 + with: + ruby-version: ${{ matrix.base_ruby }} + - name: set env + run: | + echo "GNUMAKEFLAGS=-j$((2 * NUMBER_OF_PROCESSORS))" >> $GITHUB_ENV + + - name: where check + run: | + # show where + mv /c/Windows/System32/libcrypto-1_1-x64.dll /c/Windows/System32/libcrypto-1_1-x64.dll_ + mv /c/Windows/System32/libssl-1_1-x64.dll /c/Windows/System32/libssl-1_1-x64.dll_ + result=true + for e in gcc.exe ragel.exe make.exe bison.exe libcrypto-1_1-x64.dll libssl-1_1-x64.dll; do + echo '##['group']'$'\033[93m'$e$'\033[m' + where $e || result=false + echo '##['endgroup']' + done + $result + + - name: version check + run: | + # show version + result=true + for e in gcc ragel make bison "openssl version"; do + case "$e" in *" "*) ;; *) e="$e --version";; esac + echo '##['group']'$'\033[93m'$e$'\033[m' + $e || result=false + echo '##['endgroup']' + done + $result + + - name: autogen + run: | + ./autogen.sh + working-directory: src + + - name: configure + run: > + ../src/configure --disable-install-doc --prefix=/. + --build=$CHOST --host=$CHOST --target=$CHOST + + - name: update + run: | + make incs + + - name: download gems + run: | + make update-gems + + - name: make all + timeout-minutes: 30 + run: | + make + + - run: make leaked-globals + + - name: make install + run: | + make DESTDIR=../install install-nodoc + + - name: test + timeout-minutes: 5 + run: | + make test + if: ${{matrix.test_task == 'check' || matrix.test_task == 'test'}} + + - name: test-all + timeout-minutes: 45 + run: | + # Actions uses UTF8, causes test failures, similar to normal OS setup + chcp.com 437 + make ${{ StartsWith(matrix.test_task, 'test/') && matrix.test_task || 'test-all' }} + env: + RUBY_TESTOPTS: >- + --retry --job-status=normal --show-skip --timeout-scale=1.5 + ${{ matrix.test-all-opts }} + BUNDLER_VERSION: + if: ${{matrix.test_task == 'check' || matrix.test_task == 'test-all' || StartsWith(matrix.test_task, 'test/')}} + + - name: test-spec + timeout-minutes: 10 + run: | + make ${{ StartsWith(matrix.test_task, 'spec/') && matrix.test_task || 'test-spec' }} + if: ${{matrix.test_task == 'check' || matrix.test_task == 'test-spec' || StartsWith(matrix.test_task, 'spec/')}} + + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} ${{ matrix.msystem }} / ${{ matrix.test_task }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build + shell: sh diff --git a/.github/workflows/mjit-bindgen.yml b/.github/workflows/mjit-bindgen.yml new file mode 100644 index 0000000000..26f8a1b2aa --- /dev/null +++ b/.github/workflows/mjit-bindgen.yml @@ -0,0 +1,104 @@ +name: MJIT bindgen +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + make: + strategy: + matrix: + include: + - task: mjit-bindgen + fail-fast: false + runs-on: ubuntu-22.04 + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: Set ENV + run: | + echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y \ + build-essential \ + libssl-dev libyaml-dev libreadline6-dev \ + zlib1g-dev libncurses5-dev libffi-dev \ + libclang1-14 \ + bison autoconf + sudo apt-get install -q -y pkg-config || : + - name: Set up Ruby + uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0 + with: + ruby-version: '3.1' + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Fixed world writable dirs + run: | + chmod -v go-w $HOME $HOME/.config + sudo chmod -R go-w /usr/share + sudo bash -c 'IFS=:; for d in '"$PATH"'; do chmod -v go-w $d; done' || : + - run: ./autogen.sh + working-directory: src + - name: Run configure + run: ../src/configure -C --disable-install-doc --prefix=$(pwd)/install --enable-yjit=dev_nodebug + - run: make incs + - run: make + - run: make install + - run: make ${{ matrix.task }} + - run: git diff --exit-code + working-directory: src + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ matrix.os }} / ${{ matrix.test_task }}${{ matrix.configure }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/mjit.yml b/.github/workflows/mjit.yml new file mode 100644 index 0000000000..6f7181489a --- /dev/null +++ b/.github/workflows/mjit.yml @@ -0,0 +1,113 @@ +name: MJIT +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + - '**.[1-8]' + - '**.ronn' + merge_group: + paths-ignore: + - 'doc/**' + - '**.md' + - '**.rdoc' + - '**/.document' + - '**.[1-8]' + - '**.ronn' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + make: + strategy: + matrix: + test_task: [check] # to make job names consistent + mjit_opts: [--mjit-wait] + fail-fast: false + runs-on: ubuntu-latest + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + env: + TESTOPTS: '-q --tty=no' + RUN_OPTS: '--disable-gems ${{ matrix.mjit_opts }} --mjit-debug=-ggdb3' + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + steps: + - run: mkdir build + working-directory: + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y build-essential libssl-dev libyaml-dev libreadline6-dev zlib1g-dev libncurses5-dev libffi-dev bison autoconf ruby + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Fixed world writable dirs + run: | + chmod -v go-w $HOME $HOME/.config + sudo chmod -R go-w /usr/share + sudo bash -c 'IFS=:; for d in '"$PATH"'; do chmod -v go-w $d; done' || : + - name: Set ENV + run: | + echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + - run: ./autogen.sh + working-directory: src + - name: Run configure + run: ../src/configure -C --disable-install-doc cppflags=-DVM_CHECK_MODE + - run: make incs + - run: make + - run: sudo make -s install + - name: Run test + run: | + unset GNUMAKEFLAGS + make -s test RUN_OPTS="$RUN_OPTS" + timeout-minutes: 60 + # - name: Run test-all + # run: | + # ulimit -c unlimited + # make -s test-all RUN_OPTS="$RUN_OPTS" + # timeout-minutes: 60 + - name: Run test-spec + run: | + unset GNUMAKEFLAGS + make -s test-spec RUN_OPTS="$RUN_OPTS" + timeout-minutes: 60 + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / ${{ matrix.test_task }} ${{ matrix.mjit_opts }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml new file mode 100644 index 0000000000..5d4474d978 --- /dev/null +++ b/.github/workflows/publish.yml @@ -0,0 +1,18 @@ +name: Start release workflow +on: + push: + tags: + - '*' + +jobs: + notify: + runs-on: ubuntu-latest + steps: + - name: Build release package + run: | + curl -L -X POST \ + -H "Authorization: Bearer ${{ secrets.MATZBOT_GITHUB_WORKFLOW_TOKEN }}" \ + -H "Accept: application/vnd.github+json" \ + -H "X-GitHub-Api-Version: 2022-11-28" \ + https://api.github.com/repos/ruby/actions/dispatches \ + -d '{"event_type": "${{ github.ref }}"}' diff --git a/.github/workflows/scorecards.yml b/.github/workflows/scorecards.yml new file mode 100644 index 0000000000..c12a95362d --- /dev/null +++ b/.github/workflows/scorecards.yml @@ -0,0 +1,72 @@ +# This workflow uses actions that are not certified by GitHub. They are provided +# by a third-party and are governed by separate terms of service, privacy +# policy, and support documentation. + +name: Scorecards supply-chain security +on: + # For Branch-Protection check. Only the default branch is supported. See + # https://github.com/ossf/scorecard/blob/main/docs/checks.md#branch-protection + branch_protection_rule: + # To guarantee Maintained check is occasionally updated. See + # https://github.com/ossf/scorecard/blob/main/docs/checks.md#maintained + schedule: + - cron: '22 4 * * 2' + push: + branches: [ "master" ] + +# Declare default permissions as read only. +permissions: read-all + +jobs: + analysis: + name: Scorecards analysis + runs-on: ubuntu-latest + permissions: + # Needed to upload the results to code-scanning dashboard. + security-events: write + # Needed to publish results and get a badge (see publish_results below). + id-token: write + # Uncomment the permissions below if installing in a private repository. + # contents: read + # actions: read + + steps: + - name: "Checkout code" + uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + persist-credentials: false + + - name: "Run analysis" + uses: ossf/scorecard-action@ea651e62978af7915d09fe2e282747c798bf2dab # v2.4.1 + with: + results_file: results.sarif + results_format: sarif + # (Optional) Read-only PAT token. Uncomment the `repo_token` line below if: + # - you want to enable the Branch-Protection check on a *public* repository, or + # - you are installing Scorecards on a *private* repository + # To create the PAT, follow the steps in https://github.com/ossf/scorecard-action#authentication-with-pat. + repo_token: ${{ secrets.SCORECARD_READ_TOKEN }} + + # Public repositories: + # - Publish results to OpenSSF REST API for easy access by consumers + # - Allows the repository to include the Scorecard badge. + # - See https://github.com/ossf/scorecard-action#publishing-results. + # For private repositories: + # - `publish_results` will always be set to `false`, regardless + # of the value entered here. + publish_results: true + + # Upload the results as artifacts (optional). Commenting out will disable uploads of run results in SARIF + # format to the repository Actions tab. + - name: "Upload artifact" + uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2 + with: + name: SARIF file + path: results.sarif + retention-days: 5 + + # Upload the results to GitHub's code scanning dashboard. + - name: "Upload to code-scanning" + uses: github/codeql-action/upload-sarif@959cbb7472c4d4ad70cdfe6f4976053fe48ab394 # v2.1.27 + with: + sarif_file: results.sarif diff --git a/.github/workflows/spec_guards.yml b/.github/workflows/spec_guards.yml new file mode 100644 index 0000000000..4521195a2b --- /dev/null +++ b/.github/workflows/spec_guards.yml @@ -0,0 +1,71 @@ +name: Rubyspec Version Guards Check + +on: + push: + paths: + - 'spec/**' + - '!spec/*.md' + pull_request: + paths: + - 'spec/**' + - '!spec/*.md' + merge_group: + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + rubyspec: + name: Rubyspec + + runs-on: ubuntu-22.04 + + if: >- + ${{!(false + || contains(github.event.head_commit.message, '[DOC]') + || contains(github.event.head_commit.message, 'Document') + || contains(github.event.pull_request.title, '[DOC]') + || contains(github.event.pull_request.title, 'Document') + || contains(github.event.pull_request.labels.*.name, 'Document') + || (github.event_name == 'push' && github.actor == 'dependabot[bot]') + )}} + + strategy: + matrix: + # Specs from ruby/spec should still run on all supported Ruby versions. + # This also ensures the needed ruby_version_is guards are there, see spec/README.md. + ruby: + - ruby-3.1 + - ruby-3.2 + + steps: + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + + - uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0 + with: + ruby-version: ${{ matrix.ruby }} + bundler: none + + - run: gem install webrick + + - run: ruby ../mspec/bin/mspec + working-directory: spec/ruby + env: + CHECK_LEAKS: true + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / rubyspec @ ${{ matrix.ruby }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() }} diff --git a/.github/workflows/ubuntu.yml b/.github/workflows/ubuntu.yml new file mode 100644 index 0000000000..4fbca1170e --- /dev/null +++ b/.github/workflows/ubuntu.yml @@ -0,0 +1,146 @@ +name: Ubuntu +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + make: + strategy: + matrix: + # main variables included in the job name + test_task: [check] + configure: [cppflags=-DRUBY_DEBUG] # default to use more assertions + arch: [''] + # specify all jobs with `include` to avoid testing duplicated things + include: + - test_task: check + - test_task: check + arch: i686 + configure: '' # test without -DRUBY_DEBUG as well + - test_task: check + configure: "--enable-shared --enable-load-relative" + - test_task: test-all TESTS=--repeat-count=2 + - test_task: test-bundler-parallel + - test_task: test-bundled-gems + fail-fast: false + env: + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + RUBY_DEBUG: ci + SETARCH: ${{ matrix.arch && format('setarch {0}', matrix.arch) }} + runs-on: ubuntu-22.04 + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: Set ENV + run: | + echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + - name: Install libraries + env: + arch: ${{matrix.arch}} + run: | + set -x + arch=${arch:+:${arch/i[3-6]86/i386}} + ${arch:+sudo dpkg --add-architecture ${arch#:}} + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y \ + ${arch:+cross}build-essential${arch/:/-} \ + libssl-dev${arch} libyaml-dev${arch} libreadline6-dev${arch} \ + zlib1g-dev${arch} libncurses5-dev${arch} libffi-dev${arch} \ + bison autoconf ruby + sudo apt-get install -q -y pkg-config${arch} || : + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Fixed world writable dirs + run: | + chmod -v go-w $HOME $HOME/.config + sudo chmod -R go-w /usr/share + sudo bash -c 'IFS=:; for d in '"$PATH"'; do chmod -v go-w $d; done' || : + - run: ./autogen.sh + working-directory: src + - name: Run configure + env: + arch: ${{matrix.arch}} + run: >- + $SETARCH ../src/configure -C --disable-install-doc ${{ matrix.configure }} + ${arch:+--target=$arch-$OSTYPE --host=$arch-$OSTYPE} + - run: $SETARCH make incs + - run: $SETARCH make prepare-gems + if: ${{ matrix.test_task == 'test-bundled-gems' }} + - run: $SETARCH make + - run: $SETARCH make leaked-globals + if: ${{ matrix.test_task == 'check' }} + - name: Create dummy files in build dir + run: | + $SETARCH ./miniruby -e '(("a".."z").to_a+("A".."Z").to_a+("0".."9").to_a+%w[foo bar test zzz]).each{|basename|File.write("#{basename}.rb", "raise %(do not load #{basename}.rb)")}' + if: ${{ matrix.test_task == 'check' }} + - name: make ${{ matrix.test_task }} + run: | + $SETARCH make -s ${{ matrix.test_task }} ${TESTS:+TESTS=`echo "$TESTS" | sed 's| |$$/ -n!/|g;s|^|-n!/|;s|$|$$/|'`} + timeout-minutes: 40 + env: + RUBY_TESTOPTS: "-q --tty=no" + TESTS: ${{ matrix.test_task == 'check' && matrix.skipped_tests || '' }} + TEST_BUNDLED_GEMS_ALLOW_FAILURES: "" + PRECHECK_BUNDLED_GEMS: "no" + - name: make skipped tests + run: | + $SETARCH make -s test-all TESTS=`echo "$TESTS" | sed 's| |$$/ -n/|g;s|^|-n/|;s|$|$$/|'` + env: + GNUMAKEFLAGS: "" + RUBY_TESTOPTS: "-v --tty=no" + TESTS: ${{ matrix.skipped_tests }} + if: ${{ matrix.test_task == 'check' && matrix.skipped_tests != '' }} + continue-on-error: ${{ matrix.continue-on-skipped_tests || false }} + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / ${{ matrix.test_task }} ${{ matrix.configure }}${{ matrix.arch }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/wasm.yml b/.github/workflows/wasm.yml new file mode 100644 index 0000000000..27920b5821 --- /dev/null +++ b/.github/workflows/wasm.yml @@ -0,0 +1,146 @@ +name: WebAssembly +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: # added using https://github.com/step-security/secure-workflows + contents: read + +jobs: + make: + strategy: + matrix: + entry: +# # wasmtime can't compile non-optimized Asyncified binary due to locals explosion +# - { name: O0-debuginfo, optflags: "-O0", debugflags: "-g", wasmoptflags: "-O1" } +# - { name: O1, optflags: "-O1", debugflags: "" , wasmoptflags: "-O1" } + - { name: O2, optflags: "-O2", debugflags: "" , wasmoptflags: "-O2" } +# - { name: O3, optflags: "-O3", debugflags: "" , wasmoptflags: "-O3" } +# # -O4 is equivalent to -O3 in clang, but it's different in wasm-opt +# - { name: O4, optflags: "-O3", debugflags: "" , wasmoptflags: "-O4" } +# - { name: Oz, optflags: "-Oz", debugflags: "" , wasmoptflags: "-Oz" } + fail-fast: false + env: + RUBY_TESTOPTS: '-q --tty=no' + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + WASI_SDK_VERSION_MAJOR: 14 + WASI_SDK_VERSION_MINOR: 0 + BINARYEN_VERSION: 109 + WASMTIME_VERSION: v0.33.0 + runs-on: ubuntu-22.04 + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - name: Install libraries + run: | + set -ex + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y ruby bison make autoconf git wget + + wasi_sdk_deb="wasi-sdk_${WASI_SDK_VERSION_MAJOR}.${WASI_SDK_VERSION_MINOR}_amd64.deb" + wget "https://github.com/WebAssembly/wasi-sdk/releases/download/wasi-sdk-${WASI_SDK_VERSION_MAJOR}/${wasi_sdk_deb}" + sudo dpkg -i "$wasi_sdk_deb" + rm -f "$wasi_sdk_deb" + + mkdir build-sdk + pushd build-sdk + + wasmtime_url="https://github.com/bytecodealliance/wasmtime/releases/download/${WASMTIME_VERSION}/wasmtime-${WASMTIME_VERSION}-x86_64-linux.tar.xz" + wget -O - "$wasmtime_url" | tar xJf - + sudo ln -fs "$PWD/wasmtime-${WASMTIME_VERSION}-x86_64-linux/wasmtime" /usr/local/bin/wasmtime + + binaryen_tarball="binaryen-version_${BINARYEN_VERSION}-x86_64-linux.tar.gz" + binaryen_url="https://github.com/WebAssembly/binaryen/releases/download/version_${BINARYEN_VERSION}/${binaryen_tarball}" + wget -O - "$binaryen_url" | tar xfz - + sudo ln -fs "$PWD/binaryen-version_${BINARYEN_VERSION}/bin/wasm-opt" /usr/local/bin/wasm-opt + working-directory: src + - name: Set ENV + run: | + echo "MAKEFLAGS=-j$((1 + $(sysctl -n hw.activecpu)))" >> $GITHUB_ENV + echo "WASI_SDK_PATH=/opt/wasi-sdk" >> $GITHUB_ENV + - run: ./autogen.sh + working-directory: src + + - uses: ruby/setup-ruby@13e7a03dc3ac6c3798f4570bfead2aed4d96abfb # v1.244.0 + with: + ruby-version: '3.0' + bundler: none + + - name: Download config.guess with wasi version + run: | + rm tool/config.guess tool/config.sub + ruby tool/downloader.rb -d tool -e gnu config.guess config.sub + working-directory: src + + - name: Run configure + run: | + ../src/configure \ + --host wasm32-unknown-wasi \ + --with-static-linked-ext \ + LDFLAGS=" \ + -Xlinker --stack-first \ + -Xlinker -z -Xlinker stack-size=16777216 \ + " \ + optflags="${{ matrix.entry.optflags }}" \ + debugflags="${{ matrix.entry.debugflags }}" \ + wasmoptflags="${{ matrix.entry.wasmoptflags }} ${{ matrix.entry.debugflags }}" + + # miniruby may not be built when cross-compling + - run: make mini ruby + - name: Run basictest + run: wasmtime run ./../build/miniruby --mapdir /::./ -- basictest/test.rb + working-directory: src + - name: Run bootstraptest (no thread) + run: | + NO_THREAD_TESTS="$(grep -L Thread -R ./bootstraptest | awk -F/ '{ print $NF }' | uniq | sed -n 's/test_\(.*\).rb/\1/p' | paste -s -d, -)" + ruby ./bootstraptest/runner.rb --ruby="$(which wasmtime) run $PWD/../build/ruby --mapdir /::./ -- " --verbose "--sets=$NO_THREAD_TESTS" + working-directory: src + + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / ${{ matrix.name }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.github/workflows/windows.yml b/.github/workflows/windows.yml new file mode 100644 index 0000000000..c2bd4881c2 --- /dev/null +++ b/.github/workflows/windows.yml @@ -0,0 +1,149 @@ +name: Windows +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + make: + strategy: + matrix: + include: + - vs: 2022 + vcvers: -vcvars_ver=14.2 + fail-fast: false + runs-on: windows-${{ matrix.vs }} + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + name: VisualStudio ${{ matrix.vs }} + env: + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + PATCH: C:\msys64\usr\bin\patch.exe + OS_VER: windows-${{ matrix.vs }} + steps: + - run: md build + working-directory: + - uses: msys2/setup-msys2@61f9e5e925871ba6c9e3e8da24ede83ea27fa91f # v2.27.0 + id: setup-msys2 + with: + update: true + install: bison patch + - name: patch path + shell: msys2 {0} + run: echo PATCH=$(cygpath -wa $(command -v patch)) >> $GITHUB_ENV + if: ${{ steps.setup-msys2.outcome == 'success' }} + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: C:\vcpkg\installed + key: ${{ runner.os }}-vcpkg-installed-windows-${{ matrix.vs }}-${{ github.sha }} + restore-keys: | + ${{ runner.os }}-vcpkg-installed-windows-${{ matrix.vs }}- + ${{ runner.os }}-vcpkg-installed-windows- + - name: Install libraries with vcpkg + run: | + iex "& {$(irm get.scoop.sh)} -RunAsAdmin" + Join-Path (Resolve-Path ~).Path "scoop\shims" >> $Env:GITHUB_PATH + scoop install cmake@3.31.6 + vcpkg --triplet x64-windows install libffi libyaml openssl readline zlib + shell: + pwsh + - name: git config + run: | + git config --global core.autocrlf false + git config --global core.eol lf + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: setup env + # %TEMP% is inconsistent with %TMP% and test-all expects they are consistent. + # https://github.com/actions/virtual-environments/issues/712#issuecomment-613004302 + # msys2/setup-msys2 installs MSYS2 to D:/a/_temp/msys64/usr/bin + run: | + set Path=D:/a/_temp/msys64/usr/bin;%Path% + if not "%VCVARS%" == "" goto :vcset + set VCVARS="C:\Program Files (x86)\Microsoft Visual Studio\${{ matrix.vs }}\Enterprise\VC\Auxiliary\Build\vcvars64.bat" + if not exist %VCVARS% set VCVARS="C:\Program Files\Microsoft Visual Studio\${{ matrix.vs }}\Enterprise\VC\Auxiliary\Build\vcvars64.bat" + :vcset + set | C:\msys64\usr\bin\sort > old.env + call %VCVARS% ${{ matrix.vcvers || ''}} + set TMP=%USERPROFILE%\AppData\Local\Temp + set TEMP=%USERPROFILE%\AppData\Local\Temp + set /a TEST_JOBS=(15 * %NUMBER_OF_PROCESSORS% / 10) > nul + set | C:\msys64\usr\bin\sort > new.env + C:\msys64\usr\bin\comm -13 old.env new.env >> %GITHUB_ENV% + del *.env + - name: compiler version + run: cl + - name: link libraries + run: | + for %%I in (C:\vcpkg\installed\x64-windows\bin\*.dll) do ( + if not %%~nI == readline mklink %%~nxI %%I + ) + for %%I in (libcrypto-1_1-x64 libssl-1_1-x64) do ( + ren c:\Windows\System32\%%I.dll %%I.dll_ + ) + - name: Configure + run: >- + ../src/win32/configure.bat --disable-install-doc + --with-opt-dir=C:/vcpkg/installed/x64-windows + - run: nmake incs + - run: nmake extract-extlibs + - run: nmake + env: + YACC: bison.exe + - run: nmake test + timeout-minutes: 5 + - run: nmake test-spec + timeout-minutes: 10 + - run: nmake test-all + env: + RUBY_TESTOPTS: -j${{env.TEST_JOBS}} --job-status=normal + timeout-minutes: 60 + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "VS${{ matrix.vs }} / ${{ matrix.test_task || 'check' }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build + shell: cmd diff --git a/.github/workflows/yjit-ubuntu.yml b/.github/workflows/yjit-ubuntu.yml new file mode 100644 index 0000000000..0b7b9046e9 --- /dev/null +++ b/.github/workflows/yjit-ubuntu.yml @@ -0,0 +1,170 @@ +name: YJIT Ubuntu +on: + push: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + pull_request: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + merge_group: + paths-ignore: + - 'doc/**' + - '**/man' + - '**.md' + - '**.rdoc' + - '**/.document' + +concurrency: + group: ${{ github.workflow }} / ${{ startsWith(github.event_name, 'pull') && github.ref_name || github.sha }} + cancel-in-progress: ${{ startsWith(github.event_name, 'pull') }} + +permissions: + contents: read + +jobs: + cargo: + name: Rust cargo test + # GitHub Action's image seems to already contain a Rust 1.58.0. + runs-on: ubuntu-22.04 + steps: + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + # For now we can't run cargo test --offline because it complains about the + # capstone dependency, even though the dependency is optional + #- run: cargo test --offline + - run: RUST_BACKTRACE=1 cargo test + working-directory: yjit + # Also compile and test with all features enabled + - run: RUST_BACKTRACE=1 cargo test --all-features + working-directory: yjit + # Check that we can build in release mode too + - run: cargo build --release + working-directory: yjit + make: + strategy: + fail-fast: false + matrix: + include: + - test_task: 'yjit-bindgen' + hint: 'To fix: use patch in logs' + configure: '--with-gcc=clang-14 --enable-yjit=dev' + libclang_path: '/usr/lib/llvm-14/lib/libclang.so.1' + + - test_task: "check" + # YJIT should be automatically built in release mode on x86-64 Linux with rustc present + #configure: "--enable-yjit RUSTC='rustc +1.58.0'" + configure: "RUSTC='rustc +1.58.0'" + rust_version: "1.58.0" + + - test_task: "check" + configure: "--enable-yjit=dev" + + - test_task: "check" + configure: "--enable-yjit=dev" + yjit_opts: "--yjit-call-threshold=1 --yjit-verify-ctx" + + - test_task: "test-all TESTS=--repeat-count=2" + configure: "--enable-yjit=dev" + + - test_task: "test-bundled-gems" + configure: "--enable-yjit=dev" + + - test_task: "yjit-bench" + configure: "--enable-yjit=dev" + yjit_bench_opts: "--yjit-stats" + env: + GITPULLOPTIONS: --no-tags origin ${{github.ref}} + RUN_OPTS: ${{ matrix.yjit_opts }} + YJIT_BENCH_OPTS: ${{ matrix.yjit_bench_opts }} + RUBY_DEBUG: ci + BUNDLE_JOBS: 8 # for yjit-bench + runs-on: ubuntu-22.04 + if: ${{ !contains(github.event.head_commit.message, '[DOC]') && !contains(github.event.pull_request.labels.*.name, 'Documentation') }} + steps: + - run: mkdir build + working-directory: + - name: Install libraries + run: | + set -x + sudo apt-get update -q || : + sudo apt-get install --no-install-recommends -q -y build-essential libssl-dev libyaml-dev libreadline6-dev zlib1g-dev libncurses5-dev libffi-dev bison autoconf ruby + - name: Install Rust + if: ${{ matrix.rust_version }} + run: rustup install ${{ matrix.rust_version }} --profile minimal + - name: git config + run: | + git config --global advice.detachedHead 0 + git config --global init.defaultBranch garbage + - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2 + with: + path: src + - uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3 + with: + path: src/.downloaded-cache + key: downloaded-cache + - name: Fixed world writable dirs + run: | + chmod -v go-w $HOME $HOME/.config + sudo chmod -R go-w /usr/share + sudo bash -c 'IFS=:; for d in '"$PATH"'; do chmod -v go-w $d; done' || : + - name: Set ENV + run: | + echo "GNUMAKEFLAGS=-j$((1 + $(nproc --all)))" >> $GITHUB_ENV + - run: ./autogen.sh + working-directory: src + - name: Run configure + run: ../src/configure -C --disable-install-doc --prefix=$(pwd)/install ${{ matrix.configure }} + - run: make incs + - run: make prepare-gems + if: ${{ matrix.test_task == 'test-bundled-gems' }} + - run: make -j + - run: make leaked-globals + if: ${{ matrix.test_task == 'check' }} + - name: Create dummy files in build dir + run: | + ./miniruby -e '(("a".."z").to_a+("A".."Z").to_a+("0".."9").to_a+%w[foo bar test zzz]).each{|basename|File.write("#{basename}.rb", "raise %(do not load #{basename}.rb)")}' + if: ${{ matrix.test_task == 'check' }} + - name: Enable YJIT through ENV + run: echo "RUBY_YJIT_ENABLE=1" >> $GITHUB_ENV + # Check that the binary was built with YJIT + - name: Check YJIT enabled + run: ./miniruby --yjit -v | grep "+YJIT" + - name: make ${{ matrix.test_task }} + run: make -s -j ${{ matrix.test_task }} RUN_OPTS="$RUN_OPTS" YJIT_BENCH_OPTS="$YJIT_BENCH_OPTS" + timeout-minutes: 60 + env: + RUBY_TESTOPTS: "-q --tty=no" + TEST_BUNDLED_GEMS_ALLOW_FAILURES: "" + PRECHECK_BUNDLED_GEMS: "no" + LIBCLANG_PATH: ${{ matrix.libclang_path }} + continue-on-error: ${{ matrix.test_task == 'yjit-bench' }} + - name: Show ${{ github.event.pull_request.base.ref }} GitHub URL for yjit-bench comparison + run: echo "https://github.com/${BASE_REPO}/commit/${BASE_SHA}" + env: + BASE_REPO: ${{ github.event.pull_request.base.repo.full_name }} + BASE_SHA: ${{ github.event.pull_request.base.sha }} + if: ${{ matrix.test_task == 'yjit-bench' && startsWith(github.event_name, 'pull') }} + - uses: ruby/action-slack@0bd85c72233cdbb6a0fe01d37aaeff1d21b5fce1 # v3.2.1 + with: + payload: | + { + "ci": "GitHub Actions", + "env": "${{ github.workflow }} / ${{ matrix.test_task }} ${{ matrix.configure }}", + "url": "https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}", + "commit": "${{ github.sha }}", + "branch": "${{ github.ref_name }}" + } + env: + SLACK_WEBHOOK_URL: ${{ secrets.SIMPLER_ALERTS_URL }} # ruby-lang slack: ruby/simpler-alerts-bot + if: ${{ failure() && github.event_name == 'push' }} + +defaults: + run: + working-directory: build diff --git a/.gitignore b/.gitignore index efc7fd5864..99d32a1825 100644 --- a/.gitignore +++ b/.gitignore @@ -4,25 +4,32 @@ *.a *.bak *.bc +*.bundle *.dSYM *.dmyh *.dylib *.elc *.i +*.ii *.inc *.log *.o +*.o.tmp *.obj *.old *.orig +*.pch *.pdb +*.rbinc *.rej *.s *.sav +*.sl +*.so +*.so.* *.swp *.yarb *~ -.*-* .*.list .*.time .DS_Store @@ -32,6 +39,8 @@ .pc .ppack .svn +.time +.ruby-version Makefile cygruby*.def extconf.h @@ -40,6 +49,7 @@ y.tab.c *.gcda *.gcno *.gcov +*.vscode lcov*.info # / @@ -50,7 +60,11 @@ lcov*.info /*.pc /*.rc /*_prelude.c +/.downloaded-cache +/.top-enc.mk +/build*/ /COPYING.LIB +/ChangeLog /Doxyfile /GNUmakefile /README.atheos @@ -60,6 +74,7 @@ lcov*.info /archive /autom4te*.cache /automake +/benchmark/benchmark-driver /beos /bmlog-* /breakpoints.gdb @@ -107,14 +122,17 @@ lcov*.info /rename2.h /repack /revision.h +/revision.tmp /riscos /rubicon /ruby /ruby-runner /ruby-runner.h /ruby-man.rd.gz +/rubyspec_temp /run.gdb /sizes.c +/static-ruby /test.rb /test-coverage.dat /tmp @@ -125,6 +143,13 @@ lcov*.info /web /yasmdata.rb +# /bin/ + +/bin/*.exe +/bin/*.dll +/bin/goruby +/bin/ruby + # /benchmark/ /benchmark/bm_require.data /benchmark/bmx_*.rb @@ -137,27 +162,35 @@ lcov*.info /enc/jis/props.h /enc/unicode/data +# /coroutine/ +!/coroutine/**/*.s + # /enc/trans/ /enc/trans/*.c /enc/trans/*.def /enc/trans/*.exp /enc/trans/*.lib -/enc/trans/.time # /exe/ +/exe/goruby /exe/ruby -/exe/.time # /ext/ /ext/extinit.c /ext/configure-ext.mk /ext/*/exts.mk +# /ext/-test-/cxxanyargs +/ext/-test-/cxxanyargs/failure*.failed + # /ext/-test-/win32/dln/ /ext/-test-/win32/dln/dlntest.dll /ext/-test-/win32/dln/dlntest.exp /ext/-test-/win32/dln/dlntest.lib +# /ext/-test-/gems +/ext/-test-/gems + # /ext/etc/ /ext/etc/constdefs.h @@ -170,6 +203,7 @@ lcov*.info # /ext/ripper/ /ext/ripper/eventids1.c +/ext/ripper/.eventids2-check /ext/ripper/eventids2table.c /ext/ripper/ripper.* /ext/ripper/ids1 @@ -185,6 +219,13 @@ lcov*.info /gems/src /gems/*-* +# /lib/ +/lib/ruby/[1-9]*.* +/lib/ruby/vendor_ruby + +# /misc/ +/misc/**/__pycache__ + # /spec/bundler /.rspec_status @@ -194,4 +235,16 @@ lcov*.info # /win32/ /win32/*.ico -/win32/.time + +# MJIT +/include/ruby-*/*/rb_mjit_min_header-*.h +/lib/ruby_vm/mjit/instruction.rb +/mjit_config.h +/rb_mjit_header.h + +# YJIT +/yjit-bench +/yjit_exit_locations.dump + +# /wasm/ +/wasm/tests/*.wasm diff --git a/.indent.pro b/.indent.pro index 6a207a0554..1d61cbcad1 100644 --- a/.indent.pro +++ b/.indent.pro @@ -2,12 +2,15 @@ -nbbb -nbc -br +-brs -nbs -ncdb --ce --cli0.5 +-nce +-cdw +-cli2 +-cbi2 -ndj --ei +-ncs -nfc1 -i4 -l120 @@ -16,6 +19,14 @@ -psl -sc -sob +-sbi4 +-nut +-par -TID -TVALUE +-Tst_data_t +-Tst_index_t +-Tst_table +-Trb_data_type_t +-TFILE diff --git a/.rdoc_options b/.rdoc_options new file mode 100644 index 0000000000..760507c7a2 --- /dev/null +++ b/.rdoc_options @@ -0,0 +1,4 @@ +--- +page_dir: doc +main_page: README.md +title: Documentation for Ruby development version diff --git a/.rspec_parallel b/.rspec_parallel new file mode 100644 index 0000000000..aaff198a32 --- /dev/null +++ b/.rspec_parallel @@ -0,0 +1,2 @@ +--format progress +--format ParallelTests::RSpec::RuntimeLogger --out tmp/parallel_runtime_rspec.log diff --git a/.travis.yml b/.travis.yml deleted file mode 100644 index 50b919e08d..0000000000 --- a/.travis.yml +++ /dev/null @@ -1,108 +0,0 @@ -# Copyright (C) 2011 Urabe, Shyouhei. All rights reserved. -# -# This file is a part of the programming language Ruby. Permission is hereby -# granted, to either redistribute or modify this file, provided that the -# conditions mentioned in the file COPYING are met. Consult the file for -# details. - -# This is a Travis-CI build configuration file. The list of configurations -# available is located in -# -# http://about.travis-ci.org/docs/user/build-configuration/ -# -# and as Ruby itself is a project written in C language, -# -# http://about.travis-ci.org/docs/user/languages/c/ -# -# is also a good place to look at. - -language: c -dist: trusty - -# temporary workaround for https://github.com/travis-ci/travis-ci/issues/8892 -sudo: required -group: deprecated-2017Q4 - -compiler: - - gcc - -os: - - linux - -before_install: - - "CONFIG_FLAG=" - - "JOBS=-j`nproc`" - -before_script: - - "echo JOBS=$JOBS" - - "uname -a" - - "uname -r" - - "rm -fr .ext autom4te.cache" - - "echo $TERM" - - "> config.status" - - "sed -f tool/prereq.status Makefile.in common.mk > Makefile" - - "make update-config_files" - - "make touch-unicode-files" - - "make -s $JOBS srcs UNICODE_FILES=." - - "requests=; for req in ${RUBYSPEC_PULL_REQUEST//,/ }; do - requests=\"$requests +refs/pull/$req/merge:\"; - done" - - "${requests:+git -C spec/ruby -c user.email=none -c user.name=none pull --no-edit origin $requests}" - - "${requests:+git -C spec/ruby log --oneline origin/master..@}" - - "rm config.status Makefile rbconfig.rb .rbconfig.time" - - "mkdir build config_1st config_2nd" - - "chmod -R a-w ." - - "chmod u+w build config_1st config_2nd" - - "cd build" - - "../configure -C --disable-install-doc --with-gcc=$CC $CONFIG_FLAG" - - "cp -pr config.cache config.status .ext/include ../config_1st" - - "make reconfig" - - "cp -pr config.cache config.status .ext/include ../config_2nd" - - "(cd .. && exec diff -ru config_1st config_2nd)" - - "make -s $JOBS" - -script: - - "make -s test TESTOPTS=--color=never" - - "make -s $JOBS test-all -o exts TESTOPTS='-q --color=never --job-status=normal'" - - "make -s $JOBS test-spec MSPECOPT=-j" - -# Branch matrix. Not all branches are Travis-ready so we limit branches here. -branches: - only: - - trunk - - ruby_2_2 - - ruby_2_3 - - ruby_2_4 - - ruby_2_5 - - /^feature\// - - /^bug\// - -# We want to be notified when something happens. -notifications: - irc: - channels: - - "chat.freenode.net#ruby-core" - - "chat.freenode.net#ruby-ja" - on_success: change # [always|never|change] # default: always - on_failure: always # [always|never|change] # default: always - template: - - "%{message} by @%{author}: See %{build_url}" - - slack: - rooms: - secure: i1GLETSKye85ea6dGNA3MxI/5myChmMFiZtBd5C69xK+s1sBFqEgOSbaSf9KHc0CYrHVyNhQMaZRruieV7xS+6Pfs0Zvxf1DO6QQTWC2KhkqwFDLvZncAzjoyASdR90hbr+iRPOngQ+HJuE94zemALAwEqNAinzA74PMiJXktqY= - on_pull_requests: false - on_success: change - on_failure: always - - email: - - ko1c-failure@atdot.net - -# Local Variables: -# mode: YAML -# coding: utf-8-unix -# indent-tabs-mode: nil -# tab-width: 4 -# fill-column: 79 -# default-justification: full -# End: @@ -4,10 +4,10 @@ Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright -notice, this list of conditions and the following disclaimer. + notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright -notice, this list of conditions and the following disclaimer in the -documentation and/or other materials provided with the distribution. + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index ffdf2dd4b8..13df6087ca 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -1,4 +1 @@ -Please see the [official issue tracker] and wiki [HowToContribute]. - -[official issue tracker]: https://bugs.ruby-lang.org -[HowToContribute]: https://bugs.ruby-lang.org/projects/ruby/wiki/HowToContribute +See ["Contributing to Ruby"](https://docs.ruby-lang.org/en/master/contributing_md.html), which includes setup and build instructions. @@ -2,55 +2,55 @@ Ruby is copyrighted free software by Yukihiro Matsumoto <matz@netlab.jp>. You can redistribute it and/or modify it under either the terms of the 2-clause BSDL (see the file BSDL), or the conditions below: - 1. You may make and give away verbatim copies of the source form of the - software without restriction, provided that you duplicate all of the - original copyright notices and associated disclaimers. +1. You may make and give away verbatim copies of the source form of the + software without restriction, provided that you duplicate all of the + original copyright notices and associated disclaimers. - 2. You may modify your copy of the software in any way, provided that - you do at least ONE of the following: +2. You may modify your copy of the software in any way, provided that + you do at least ONE of the following: - a) place your modifications in the Public Domain or otherwise - make them Freely Available, such as by posting said - modifications to Usenet or an equivalent medium, or by allowing - the author to include your modifications in the software. + a. place your modifications in the Public Domain or otherwise + make them Freely Available, such as by posting said + modifications to Usenet or an equivalent medium, or by allowing + the author to include your modifications in the software. - b) use the modified software only within your corporation or - organization. + b. use the modified software only within your corporation or + organization. - c) give non-standard binaries non-standard names, with - instructions on where to get the original software distribution. + c. give non-standard binaries non-standard names, with + instructions on where to get the original software distribution. - d) make other distribution arrangements with the author. + d. make other distribution arrangements with the author. - 3. You may distribute the software in object code or binary form, - provided that you do at least ONE of the following: +3. You may distribute the software in object code or binary form, + provided that you do at least ONE of the following: - a) distribute the binaries and library files of the software, - together with instructions (in the manual page or equivalent) - on where to get the original distribution. + a. distribute the binaries and library files of the software, + together with instructions (in the manual page or equivalent) + on where to get the original distribution. - b) accompany the distribution with the machine-readable source of - the software. + b. accompany the distribution with the machine-readable source of + the software. - c) give non-standard binaries non-standard names, with - instructions on where to get the original software distribution. + c. give non-standard binaries non-standard names, with + instructions on where to get the original software distribution. - d) make other distribution arrangements with the author. + d. make other distribution arrangements with the author. - 4. You may modify and include the part of the software into any other - software (possibly commercial). But some files in the distribution - are not written by the author, so that they are not under these terms. +4. You may modify and include the part of the software into any other + software (possibly commercial). But some files in the distribution + are not written by the author, so that they are not under these terms. - For the list of those files and their copying conditions, see the - file LEGAL. + For the list of those files and their copying conditions, see the + file LEGAL. - 5. The scripts and library files supplied as input to or produced as - output from the software do not automatically fall under the - copyright of the software, but belong to whomever generated them, - and may be sold commercially, and may be aggregated with this - software. +5. The scripts and library files supplied as input to or produced as + output from the software do not automatically fall under the + copyright of the software, but belong to whomever generated them, + and may be sold commercially, and may be aggregated with this + software. - 6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR - IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED - WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR - PURPOSE. +6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR + IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + PURPOSE. diff --git a/COPYING.ja b/COPYING.ja index e50d01c8d1..230376bc60 100644 --- a/COPYING.ja +++ b/COPYING.ja @@ -2,50 +2,50 @@ ã¾ãŸã¯ä»¥ä¸‹ã«ç¤ºã™æ¡ä»¶ã§æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã‚’å†é…布ã§ãã¾ã™ 2-clause BSDLã«ã¤ã„ã¦ã¯BSDLファイルをå‚ç…§ã—ã¦ä¸‹ã•ã„． - 1. 複製ã¯åˆ¶é™ãªã自由ã§ã™ï¼Ž +1. 複製ã¯åˆ¶é™ãªã自由ã§ã™ï¼Ž - 2. 以下ã®æ¡ä»¶ã®ã„ãšã‚Œã‹ã‚’満ãŸã™æ™‚ã«æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ã‚½ãƒ¼ã‚¹ã‚’ - 自由ã«å¤‰æ›´ã§ãã¾ã™ï¼Ž +2. 以下ã®æ¡ä»¶ã®ã„ãšã‚Œã‹ã‚’満ãŸã™æ™‚ã«æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ã‚½ãƒ¼ã‚¹ã‚’ + 自由ã«å¤‰æ›´ã§ãã¾ã™ï¼Ž - (a) ãƒãƒƒãƒˆãƒ‹ãƒ¥ãƒ¼ã‚ºã«ãƒã‚¹ãƒˆã—ãŸã‚Šï¼Œä½œè€…ã«å¤‰æ›´ã‚’é€ä»˜ã™ã‚‹ - ãªã©ã®æ–¹æ³•ã§ï¼Œå¤‰æ›´ã‚’公開ã™ã‚‹ï¼Ž + a. ãƒãƒƒãƒˆãƒ‹ãƒ¥ãƒ¼ã‚ºã«ãƒã‚¹ãƒˆã—ãŸã‚Šï¼Œä½œè€…ã«å¤‰æ›´ã‚’é€ä»˜ã™ã‚‹ + ãªã©ã®æ–¹æ³•ã§ï¼Œå¤‰æ›´ã‚’公開ã™ã‚‹ï¼Ž - (b) 変更ã—ãŸæœ¬ãƒ—ãƒã‚°ãƒ©ãƒ を自分ã®æ‰€å±žã™ã‚‹çµ„織内部ã ã‘㧠- 使ã†ï¼Ž + b. 変更ã—ãŸæœ¬ãƒ—ãƒã‚°ãƒ©ãƒ を自分ã®æ‰€å±žã™ã‚‹çµ„織内部ã ã‘㧠+ 使ã†ï¼Ž - (c) 変更点を明示ã—ãŸã†ãˆï¼Œã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã®åå‰ã‚’変更ã™ã‚‹ï¼Ž - ãã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚’é…布ã™ã‚‹æ™‚ã«ã¯å¤‰æ›´å‰ã®æœ¬ãƒ—ãƒã‚°ãƒ© - ムもåŒæ™‚ã«é…布ã™ã‚‹ï¼Žã¾ãŸã¯å¤‰æ›´å‰ã®æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ã‚½ãƒ¼ - スã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž + c. 変更点を明示ã—ãŸã†ãˆï¼Œã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã®åå‰ã‚’変更ã™ã‚‹ï¼Ž + ãã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚’é…布ã™ã‚‹æ™‚ã«ã¯å¤‰æ›´å‰ã®æœ¬ãƒ—ãƒã‚°ãƒ© + ムもåŒæ™‚ã«é…布ã™ã‚‹ï¼Žã¾ãŸã¯å¤‰æ›´å‰ã®æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ã‚½ãƒ¼ + スã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž - (d) ãã®ä»–ã®å¤‰æ›´æ¡ä»¶ã‚’作者ã¨åˆæ„ã™ã‚‹ï¼Ž + d. ãã®ä»–ã®å¤‰æ›´æ¡ä»¶ã‚’作者ã¨åˆæ„ã™ã‚‹ï¼Ž - 3. 以下ã®æ¡ä»¶ã®ã„ãšã‚Œã‹ã‚’満ãŸã™æ™‚ã«æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ をコンパイ - ルã—ãŸã‚ªãƒ–ジェクトコードや実行形å¼ã§ã‚‚é…布ã§ãã¾ã™ï¼Ž +3. 以下ã®æ¡ä»¶ã®ã„ãšã‚Œã‹ã‚’満ãŸã™æ™‚ã«æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ をコンパイ + ルã—ãŸã‚ªãƒ–ジェクトコードや実行形å¼ã§ã‚‚é…布ã§ãã¾ã™ï¼Ž - (a) ãƒã‚¤ãƒŠãƒªã‚’å—ã‘å–ã£ãŸäººãŒã‚½ãƒ¼ã‚¹ã‚’入手ã§ãるよã†ã«ï¼Œ - ソースã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž + a. ãƒã‚¤ãƒŠãƒªã‚’å—ã‘å–ã£ãŸäººãŒã‚½ãƒ¼ã‚¹ã‚’入手ã§ãるよã†ã«ï¼Œ + ソースã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž - (b) 機械å¯èªãªã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã‚’添付ã™ã‚‹ï¼Ž + b. 機械å¯èªãªã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã‚’添付ã™ã‚‹ï¼Ž - (c) 変更を行ã£ãŸãƒã‚¤ãƒŠãƒªã¯åå‰ã‚’変更ã—ãŸã†ãˆï¼Œã‚ªãƒªã‚¸ãƒŠ - ルã®ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž + c. 変更を行ã£ãŸãƒã‚¤ãƒŠãƒªã¯åå‰ã‚’変更ã—ãŸã†ãˆï¼Œã‚ªãƒªã‚¸ãƒŠ + ルã®ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã®å…¥æ‰‹æ³•ã‚’明示ã™ã‚‹ï¼Ž - (d) ãã®ä»–ã®é…布æ¡ä»¶ã‚’作者ã¨åˆæ„ã™ã‚‹ï¼Ž + d. ãã®ä»–ã®é…布æ¡ä»¶ã‚’作者ã¨åˆæ„ã™ã‚‹ï¼Ž - 4. ä»–ã®ãƒ—ãƒã‚°ãƒ©ãƒ ã¸ã®å¼•ç”¨ã¯ã„ã‹ãªã‚‹ç›®çš„ã§ã‚れ自由ã§ã™ï¼ŽãŸ - ã ã—，本プãƒã‚°ãƒ©ãƒ ã«å«ã¾ã‚Œã‚‹ä»–ã®ä½œè€…ã«ã‚ˆã‚‹ã‚³ãƒ¼ãƒ‰ã¯ï¼Œã - ã‚Œãžã‚Œã®ä½œè€…ã®æ„å‘ã«ã‚ˆã‚‹åˆ¶é™ãŒåŠ ãˆã‚‰ã‚Œã‚‹å ´åˆãŒã‚ã‚Šã¾ã™ï¼Ž +4. ä»–ã®ãƒ—ãƒã‚°ãƒ©ãƒ ã¸ã®å¼•ç”¨ã¯ã„ã‹ãªã‚‹ç›®çš„ã§ã‚れ自由ã§ã™ï¼ŽãŸ + ã ã—，本プãƒã‚°ãƒ©ãƒ ã«å«ã¾ã‚Œã‚‹ä»–ã®ä½œè€…ã«ã‚ˆã‚‹ã‚³ãƒ¼ãƒ‰ã¯ï¼Œã + ã‚Œãžã‚Œã®ä½œè€…ã®æ„å‘ã«ã‚ˆã‚‹åˆ¶é™ãŒåŠ ãˆã‚‰ã‚Œã‚‹å ´åˆãŒã‚ã‚Šã¾ã™ï¼Ž - ãれらファイルã®ä¸€è¦§ã¨ãã‚Œãžã‚Œã®é…布æ¡ä»¶ãªã©ã«ä»˜ã„ã¦ã¯ - LEGALファイルをå‚ç…§ã—ã¦ãã ã•ã„． + ãれらファイルã®ä¸€è¦§ã¨ãã‚Œãžã‚Œã®é…布æ¡ä»¶ãªã©ã«ä»˜ã„ã¦ã¯ + LEGALファイルをå‚ç…§ã—ã¦ãã ã•ã„． - 5. 本プãƒã‚°ãƒ©ãƒ ã¸ã®å…¥åŠ›ã¨ãªã‚‹ã‚¹ã‚¯ãƒªãƒ—トãŠã‚ˆã³ï¼Œæœ¬ãƒ—ãƒã‚°ãƒ© - ムã‹ã‚‰ã®å‡ºåŠ›ã®æ¨©åˆ©ã¯æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ä½œè€…ã§ã¯ãªã，ãれ㞠- ã‚Œã®å…¥å‡ºåŠ›ã‚’生æˆã—ãŸäººã«å±žã—ã¾ã™ï¼Žã¾ãŸï¼Œæœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã« - 組ã¿è¾¼ã¾ã‚Œã‚‹ãŸã‚ã®æ‹¡å¼µãƒ©ã‚¤ãƒ–ラリã«ã¤ã„ã¦ã‚‚åŒæ§˜ã§ã™ï¼Ž +5. 本プãƒã‚°ãƒ©ãƒ ã¸ã®å…¥åŠ›ã¨ãªã‚‹ã‚¹ã‚¯ãƒªãƒ—トãŠã‚ˆã³ï¼Œæœ¬ãƒ—ãƒã‚°ãƒ© + ムã‹ã‚‰ã®å‡ºåŠ›ã®æ¨©åˆ©ã¯æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã®ä½œè€…ã§ã¯ãªã，ãれ㞠+ ã‚Œã®å…¥å‡ºåŠ›ã‚’生æˆã—ãŸäººã«å±žã—ã¾ã™ï¼Žã¾ãŸï¼Œæœ¬ãƒ—ãƒã‚°ãƒ©ãƒ ã« + 組ã¿è¾¼ã¾ã‚Œã‚‹ãŸã‚ã®æ‹¡å¼µãƒ©ã‚¤ãƒ–ラリã«ã¤ã„ã¦ã‚‚åŒæ§˜ã§ã™ï¼Ž - 6. 本プãƒã‚°ãƒ©ãƒ ã¯ç„¡ä¿è¨¼ã§ã™ï¼Žä½œè€…ã¯æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ をサãƒãƒ¼ãƒˆ - ã™ã‚‹æ„å¿—ã¯ã‚ã‚Šã¾ã™ãŒï¼Œãƒ—ãƒã‚°ãƒ©ãƒ 自身ã®ãƒã‚°ã‚ã‚‹ã„ã¯æœ¬ãƒ— - ãƒã‚°ãƒ©ãƒ ã®å®Ÿè¡Œãªã©ã‹ã‚‰ç™ºç”Ÿã™ã‚‹ã„ã‹ãªã‚‹æ害ã«å¯¾ã—ã¦ã‚‚責 - 任をæŒã¡ã¾ã›ã‚“． +6. 本プãƒã‚°ãƒ©ãƒ ã¯ç„¡ä¿è¨¼ã§ã™ï¼Žä½œè€…ã¯æœ¬ãƒ—ãƒã‚°ãƒ©ãƒ をサãƒãƒ¼ãƒˆ + ã™ã‚‹æ„å¿—ã¯ã‚ã‚Šã¾ã™ãŒï¼Œãƒ—ãƒã‚°ãƒ©ãƒ 自身ã®ãƒã‚°ã‚ã‚‹ã„ã¯æœ¬ãƒ— + ãƒã‚°ãƒ©ãƒ ã®å®Ÿè¡Œãªã©ã‹ã‚‰ç™ºç”Ÿã™ã‚‹ã„ã‹ãªã‚‹æ害ã«å¯¾ã—ã¦ã‚‚責 + 任をæŒã¡ã¾ã›ã‚“． @@ -1,160 +1,248 @@ -LEGAL NOTICE INFORMATION ------------------------- +# -*- rdoc -*- + += LEGAL NOTICE INFORMATION +-------------------------- All the files in this distribution are covered under either the Ruby's license (see the file COPYING) or public-domain except some files mentioned below. -ccan/build_assert/build_assert.h -ccan/check_type/check_type.h -ccan/container_of/container_of.h -ccan/str/str.h +[addr2line.c] + + A part of this file is from FreeBSD. + + >>> + Copyright (c) 1986, 1988, 1991, 1993:: + The Regents of the University of California. All rights reserved. - These files are licensed under the CC0. + (c) UNIX System Laboratories, Inc. - https://creativecommons.org/choose/zero/ + All or some portions of this file are derived from material licensed + to the University of California by American Telephone and Telegraph + Co. or Unix System Laboratories, Inc. and are reproduced herein with + the permission of UNIX System Laboratories, Inc. -ccan/list/list.h + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + 4. Neither the name of the University nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. - This file is licensed under the MIT License. + THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. - Permission is hereby granted, free of charge, to any person obtaining a copy - of this software and associated documentation files (the "Software"), to deal - in the Software without restriction, including without limitation the rights - to use, copy, modify, merge, publish, distribute, sublicense, and/or sell - copies of the Software, and to permit persons to whom the Software is - furnished to do so, subject to the following conditions: + @(#)subr_prf.c 8.3 (Berkeley) 1/21/94 + + +[ccan/build_assert/build_assert.h] +[ccan/check_type/check_type.h] +[ccan/container_of/container_of.h] +[ccan/str/str.h] + + These files are licensed under the {CC0}[https://creativecommons.org/choose/zero/]. + +[ccan/list/list.h] + + This file is licensed under the {MIT License}[rdoc-label:label-MIT+License]. + +[coroutine] + + Unless otherwise specified, these files are licensed under the + {MIT License}[rdoc-label:label-MIT+License]. + +[include/ruby/onigmo.h] +[include/ruby/oniguruma.h] +[regcomp.c] +[regenc.c] +[regenc.h] +[regerror.c] +[regexec.c] +[regint.h] +[regparse.c] +[regparse.h] +[enc/ascii.c] +[enc/big5.c] +[enc/cp949.c] +[enc/emacs_mule.c] +[enc/encdb.c] +[enc/euc_jp.c] +[enc/euc_kr.c] +[enc/euc_tw.c] +[enc/gb18030.c] +[enc/gb2312.c] +[enc/gbk.c] +[enc/iso_8859_1.c] +[enc/iso_8859_10.c] +[enc/iso_8859_11.c] +[enc/iso_8859_13.c] +[enc/iso_8859_14.c] +[enc/iso_8859_15.c] +[enc/iso_8859_16.c] +[enc/iso_8859_2.c] +[enc/iso_8859_3.c] +[enc/iso_8859_4.c] +[enc/iso_8859_5.c] +[enc/iso_8859_6.c] +[enc/iso_8859_7.c] +[enc/iso_8859_8.c] +[enc/iso_8859_9.c] +[enc/koi8_r.c] +[enc/koi8_u.c] +[enc/shift_jis.c] +[enc/unicode.c] +[enc/us_ascii.c] +[enc/utf_16be.c] +[enc/utf_16le.c] +[enc/utf_32be.c] +[enc/utf_32le.c] +[enc/utf_8.c] +[enc/windows_1251.c] +[enc/windows_31j.c] + + Onigmo (Oniguruma-mod) LICENSE + + >>> + Copyright (c) 2002-2009:: K.Kosako <sndgk393 AT ybb DOT ne DOT jp> + Copyright (c) 2011-2014:: K.Takata <kentkt AT csc DOT jp> + All rights reserved. - The above copyright notice and this permission notice shall be included in - all copies or substantial portions of the Software. + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. - THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN - THE SOFTWARE. - -include/ruby/onigmo.h: -include/ruby/oniguruma.h: -regcomp.c: -regenc.[ch]: -regerror.c: -regexec.c: -regint.h: -regparse.[ch]: -enc/ascii.c -enc/big5.c -enc/cp949.c -enc/emacs_mule.c -enc/encdb.c -enc/euc_jp.c -enc/euc_kr.c -enc/euc_tw.c -enc/gb18030.c -enc/gb2312.c -enc/gbk.c -enc/iso_8859_1.c -enc/iso_8859_10.c -enc/iso_8859_11.c -enc/iso_8859_13.c -enc/iso_8859_14.c -enc/iso_8859_15.c -enc/iso_8859_16.c -enc/iso_8859_2.c -enc/iso_8859_3.c -enc/iso_8859_4.c -enc/iso_8859_5.c -enc/iso_8859_6.c -enc/iso_8859_7.c -enc/iso_8859_8.c -enc/iso_8859_9.c -enc/koi8_r.c -enc/koi8_u.c -enc/shift_jis.c -enc/unicode.c -enc/us_ascii.c -enc/utf_16be.c -enc/utf_16le.c -enc/utf_32be.c -enc/utf_32le.c -enc/utf_8.c -enc/windows_1251.c - -Onigmo (Oniguruma-mod) LICENSE ------------------------------- - -/*- - * Copyright (c) 2002-2009 K.Kosako <sndgk393 AT ybb DOT ne DOT jp> - * Copyright (c) 2011-2014 K.Takata <kentkt AT csc DOT jp> - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - */ - - -Oniguruma LICENSE ------------------ - -/*- - * Copyright (c) 2002-2009 K.Kosako <sndgk393 AT ybb DOT ne DOT jp> - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - */ - -https://github.com/k-takata/Onigmo/ -http://www.geocities.jp/kosako3/oniguruma/ -http://www.ruby-lang.org/cgi-bin/cvsweb.cgi/oniguruma/ -http://www.freebsd.org/cgi/cvsweb.cgi/ports/devel/oniguruma/ - - When this software is partly used or it is distributed with Ruby, - this of Ruby follows the license of Ruby. - -enc/trans/GB/GB12345%UCS.src: -enc/trans/GB/UCS%GB12345.src: -enc/trans/GB/GB2312%UCS.src: -enc/trans/GB/UCS%GB2312.src: + THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + + Oniguruma LICENSE + + >>> + Copyright (c) 2002-2009:: K.Kosako <sndgk393 AT ybb DOT ne DOT jp> + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + + * https://github.com/k-takata/Onigmo/ + * https://github.com/kkos/oniguruma + * https://svnweb.freebsd.org/ports/head/devel/oniguruma/ + + When this software is partly used or it is distributed with Ruby, + this of Ruby follows the license of Ruby. +[enc/windows_1250.c] +[enc/windows_1252.c] + + >>> + Copyright (c) 2006-2007:: Byte <byte AT mail DOT kna DOT ru> + K.Kosako <sndgk393 AT ybb DOT ne DOT jp> + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + +[enc/cesu_8.c] +[enc/windows_1253.c] +[enc/windows_1254.c] +[enc/windows_1257.c] + + >>> + Copyright (c) 2002-2007:: K.Kosako <sndgk393 AT ybb DOT ne DOT jp> + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + +[enc/trans/GB/GB12345%UCS.src] +[enc/trans/GB/UCS%GB12345.src] +[enc/trans/GB/GB2312%UCS.src] +[enc/trans/GB/UCS%GB2312.src] + + These files have this explanatory texts. + + >>> This mapping data was created from files provided by Unicode, Inc. (The Unicode Consortium). The files were used to create a product supporting Unicode, as explicitly permitted in the files' copyright notices. @@ -162,56 +250,82 @@ enc/trans/GB/UCS%GB2312.src: files for any particular purpose, and has ceased to publish the files many years ago. -enc/trans/JIS/JISX0201-KANA%UCS.src: -enc/trans/JIS/JISX0208@1990%UCS.src: -enc/trans/JIS/JISX0212%UCS.src: -enc/trans/JIS/UCS%JISX0201-KANA.src: -enc/trans/JIS/UCS%JISX0208@1990.src: -enc/trans/JIS/UCS%JISX0212.src: +[enc/trans/JIS/JISX0201-KANA%UCS.src] +[enc/trans/JIS/JISX0208\@1990%UCS.src] +[enc/trans/JIS/JISX0212%UCS.src] +[enc/trans/JIS/UCS%JISX0201-KANA.src] +[enc/trans/JIS/UCS%JISX0208@1990.src] +[enc/trans/JIS/UCS%JISX0212.src] + These files are copyrighted as the following. + + >>> © 2015 Unicode®, Inc. + For terms of use, see http://www.unicode.org/terms_of_use.html -enc/trans/JIS/JISX0213-1%UCS@BMP.src: -enc/trans/JIS/JISX0213-1%UCS@SIP.src: -enc/trans/JIS/JISX0213-2%UCS@BMP.src: -enc/trans/JIS/JISX0213-2%UCS@SIP.src: +[enc/trans/JIS/JISX0213-1%UCS@BMP.src] +[enc/trans/JIS/JISX0213-1%UCS@SIP.src] +[enc/trans/JIS/JISX0213-2%UCS@BMP.src] +[enc/trans/JIS/JISX0213-2%UCS@SIP.src] + + These files are copyrighted as the following. + + >>> + Copyright (C) 2001:: earthian@tama.or.jp, All Rights Reserved. + Copyright (C) 2001:: I'O, All Rights Reserved. + Copyright (C) 2006:: Project X0213, All Rights Reserved. + You can use, modify, distribute this table freely. + +[enc/trans/JIS/UCS@BMP%JISX0213-1.src] +[enc/trans/JIS/UCS@BMP%JISX0213-2.src] +[enc/trans/JIS/UCS@SIP%JISX0213-1.src] +[enc/trans/JIS/UCS@SIP%JISX0213-2.src] - Copyright (C) 2001 earthian@tama.or.jp, All Rights Reserved. - Copyright (C) 2001 I'O, All Rights Reserved. - Copyright (C) 2006 Project X0213, All Rights Reserved. - You can use, modify, distribute this table freely. + These files are copyrighted as the following. -enc/trans/JIS/UCS@BMP%JISX0213-1.src: -enc/trans/JIS/UCS@BMP%JISX0213-2.src: -enc/trans/JIS/UCS@SIP%JISX0213-1.src: -enc/trans/JIS/UCS@SIP%JISX0213-2.src: + >>> + Copyright (C) 2001:: earthian@tama.or.jp, All Rights Reserved. + Copyright (C) 2001:: I'O, All Rights Reserved. + You can use, modify, distribute this table freely. - Copyright (C) 2001 earthian@tama.or.jp, All Rights Reserved. - Copyright (C) 2001 I'O, All Rights Reserved. - You can use, modify, distribute this table freely. +[enc/trans/ucm/glibc-BIG5-2.3.3.ucm] +[enc/trans/ucm/glibc-BIG5HKSCS-2.3.3.ucm] -configure: + >>> + Copyright (C) 2001-2005:: International Business Machines + Corporation and others. All Rights Reserved. + +[enc/trans/ucm/windows-950-2000.ucm] +[enc/trans/ucm/windows-950_hkscs-2001.ucm] + + >>> + Copyright (C) 2001-2002:: International Business Machines + Corporation and others. All Rights Reserved. + + +[configure] This file is free software. - Copyright (C) 1992, 93, 94, 95, 96 Free Software Foundation, Inc. + >>> + Copyright (C) 1992-1996, 1998-2012:: Free Software Foundation, Inc. This configure script is free software; the Free Software Foundation gives unlimited permission to copy, distribute and modify it. -tool/config.guess: -tool/config.sub: +[tool/config.guess] +[tool/config.sub] As long as you distribute these files with the file configure, they are covered under the Ruby's license. - Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999 - Free Software Foundation, Inc. + >>> + Copyright 1992-2018:: Free Software Foundation, Inc. This file is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or + the Free Software Foundation; either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but @@ -220,54 +334,85 @@ tool/config.sub: General Public License for more details. You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + along with this program; if not, see <https://www.gnu.org/licenses/>. As a special exception to the GNU General Public License, if you distribute this file as part of a program that contains a configuration script generated by Autoconf, you may include it under - the same distribution terms that you use for the rest of that program. + the same distribution terms that you use for the rest of that + program. This Exception is an additional permission under section 7 + of the GNU General Public License, version 3 ("GPLv3"). + +[tool/lib/test/*] +[tool/lib/core_assertions.rb] -parse.c: + Some of methods on these files are based on MiniTest 4. MiniTest 4 is + distributed under the MIT License. + + >>> + Copyright (c) Ryan Davis, seattle.rb + + Permission is hereby granted, free of charge, to any person obtaining + a copy of this software and associated documentation files (the + 'Software'), to deal in the Software without restriction, including + without limitation the rights to use, copy, modify, merge, publish, + distribute, sublicense, and/or sell copies of the Software, and to + permit persons to whom the Software is furnished to do so, subject to + the following conditions: + + The above copyright notice and this permission notice shall be + included in all copies or substantial portions of the Software. - This file is licensed under the GPL, but is incorporated into Ruby and + THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. + IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY + CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, + TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE + SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +[parse.c] +[parse.h] + + These files are licensed under the GPL, but are incorporated into Ruby and redistributed under the terms of the Ruby license, as permitted by the exception to the GPL below. - Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003, 2004, 2005, 2006 - Free Software Foundation, Inc. + >>> + Copyright (C) 1984, 1989-1990, 2000-2015, 2018:: Free Software Foundation, Inc. - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 3 of the License, or + (at your option) any later version. - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program. If not, see <http://www.gnu.org/licenses/>. - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 51 Franklin Street, Fifth Floor, - Boston, MA 02110-1301, USA. */ + As a special exception, you may create a larger work that contains + part or all of the Bison parser skeleton and distribute that work + under terms of your choice, so long as that work isn't itself a + parser generator using the skeleton or a modified version thereof + as a parser skeleton. Alternatively, if you modify or redistribute + the parser skeleton itself, you may (at your option) remove this + special exception, which will cause the skeleton and the resulting + Bison output files to be licensed under the GNU General Public + License without this special exception. - /* As a special exception, you may create a larger work that contains - part or all of the Bison parser skeleton and distribute that work - under terms of your choice, so long as that work isn't itself a - parser generator using the skeleton or a modified version thereof - as a parser skeleton. Alternatively, if you modify or redistribute - the parser skeleton itself, you may (at your option) remove this - special exception, which will cause the skeleton and the resulting - Bison output files to be licensed under the GNU General Public - License without this special exception. + This special exception was added by the Free Software Foundation in + version 2.2 of Bison. - This special exception was added by the Free Software Foundation in - version 2.2 of Bison. */ +[missing/dtoa.c] -util.c (partly): + This file is under these licenses. - Copyright (c) 1991, 2000, 2001 by Lucent Technologies. + >>> + Copyright (c) 1991, 2000, 2001:: by Lucent Technologies. Permission to use, copy, modify, and distribute this software for any purpose without fee is hereby granted, provided that this entire notice @@ -280,71 +425,74 @@ util.c (partly): REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE. -win32/win32.[ch]: + >>> + Copyright (c) 2004-2008:: David Schultz <das@FreeBSD.ORG> + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + + THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + +[win32/win32.c] +[include/ruby/win32.h] You can apply the Artistic License to these files. (or GPL, alternatively) - Copyright (c) 1993, Intergraph Corporation + >>> + Copyright (c) 1993:: Intergraph Corporation You may distribute under the terms of either the GNU General Public License or the Artistic License, as specified in the perl README file. -util.c (partly): - - Copyright (c) 2004-2008 David Schultz <das@FreeBSD.ORG> - All rights reserved. - - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions - are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - - THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - SUCH DAMAGE. - -random.c +[missing/mt19937.c] This file is under the new-style BSD license. - A C-program for MT19937, with initialization improved 2002/2/10. + >>> + A C-program for MT19937, with initialization improved 2002/2/10.:: Coded by Takuji Nishimura and Makoto Matsumoto. + This is a faster version by taking Shawn Cokus's optimization, Matthe Bellew's simplification, Isaku Wada's real version. Before using, initialize the state by using init_genrand(seed) or init_by_array(init_key, key_length). - Copyright (C) 1997 - 2002, Makoto Matsumoto and Takuji Nishimura, - All rights reserved. + Copyright (C) 1997 - 2002:: Makoto Matsumoto and Takuji Nishimura, + All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. - 3. The names of its contributors may not be used to endorse or promote - products derived from this software without specific prior written - permission. + 3. The names of its contributors may not be used to endorse or promote + products derived from this software without specific prior written + permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT @@ -363,12 +511,15 @@ random.c http://www.math.keio.ac.jp/matumoto/emt.html email: matumoto@math.keio.ac.jp -vm_dump.c:procstat_vm + The Wayback Machine url: http://web.archive.org/web/19990429082237/http://www.math.keio.ac.jp/matumoto/emt.html + +[missing/procstat_vm.c] This file is under the new-style BSD license. - Copyright (c) 2007 Robert N. M. Watson - All rights reserved. + >>> + Copyright (c) 2007:: Robert N. M. Watson + All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions @@ -393,146 +544,65 @@ vm_dump.c:procstat_vm $FreeBSD: head/usr.bin/procstat/procstat_vm.c 261780 2014-02-11 21:57:37Z jhb $ -vsnprintf.c: +[vsnprintf.c] - This file is under the old-style BSD license. Note that the - paragraph 3 below is now null and void. + This file is under the {old-style BSD license}[rdoc-label:label-Old-style+BSD+license]. - Copyright (c) 1990, 1993 - The Regents of the University of California. All rights reserved. + >>> + Copyright (c) 1990, 1993:: + The Regents of the University of California. All rights reserved. This code is derived from software contributed to Berkeley by Chris Torek. - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions - are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - 3. Neither the name of the University nor the names of its contributors - may be used to endorse or promote products derived from this software - without specific prior written permission. - - THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND - ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - SUCH DAMAGE. - - IMPORTANT NOTE: - -------------- - From ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change - paragraph 3 above is now null and void. - -st.c: -strftime.c: -include/ruby/st.h: -missing/acosh.c: -missing/alloca.c: -missing/dup2.c: -missing/erf.c: -missing/finite.c: -missing/hypot.c: -missing/isinf.c: -missing/isnan.c: -missing/lgamma_r.c: -missing/memcmp.c: -missing/memmove.c: -missing/strchr.c: -missing/strerror.c: -missing/strstr.c: -missing/tgamma.c: -ext/date/date_strftime.c: -ext/digest/sha1/sha1.[ch]: -ext/sdbm/_sdbm.c: -ext/sdbm/sdbm.h: +[st.c] +[strftime.c] +[include/ruby/st.h] +[missing/acosh.c] +[missing/alloca.c] +[missing/erf.c] +[missing/hypot.c] +[missing/lgamma_r.c] +[missing/memcmp.c] +[missing/memmove.c] +[missing/strchr.c] +[missing/strerror.c] +[missing/strstr.c] +[missing/tgamma.c] +[ext/date/date_strftime.c] +[ext/digest/sha1/sha1.c] +[ext/digest/sha1/sha1.h] These files are all under public domain. -missing/crypt.c: +[missing/crypt.c] - This file is under the old-style BSD license. Note that the - paragraph 3 below is now null and void. + This file is under the {old-style BSD license}[rdoc-label:label-Old-style+BSD+license]. - Copyright (c) 1989, 1993 - The Regents of the University of California. All rights reserved. + >>> + Copyright (c) 1989, 1993:: + The Regents of the University of California. All rights reserved. This code is derived from software contributed to Berkeley by Tom Truscott. - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions - are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - 3. Neither the name of the University nor the names of its contributors - may be used to endorse or promote products derived from this software - without specific prior written permission. +[missing/setproctitle.c] - THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND - ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - SUCH DAMAGE. - -missing/setproctitle.c - - This file is under the old-style BSD license. Note that the - paragraph 3 below is now null and void. + This file is under the {old-style BSD license}[rdoc-label:label-Old-style+BSD+license]. - Copyright 2003 Damien Miller - Copyright (c) 1983, 1995-1997 Eric P. Allman - Copyright (c) 1988, 1993 - The Regents of the University of California. All rights reserved. - - Redistribution and use in source and binary forms, with or without - modification, are permitted provided that the following conditions - are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. - 3. Neither the name of the University nor the names of its contributors - may be used to endorse or promote products derived from this software - without specific prior written permission. - - THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND - ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - SUCH DAMAGE. + >>> + Copyright 2003:: Damien Miller + Copyright (c) 1983, 1995-1997:: Eric P. Allman + Copyright (c) 1988, 1993:: + The Regents of the University of California. All rights reserved. -missing/strlcat.c -missing/strlcpy.c +[missing/strlcat.c] +[missing/strlcpy.c] These files are under an ISC-style license. - Copyright (c) 1998, 2015 Todd C. Miller <Todd.Miller@courtesan.com> + >>> + Copyright (c) 1998, 2015:: Todd C. Miller <Todd.Miller@courtesan.com> Permission to use, copy, modify, and distribute this software for any purpose with or without fee is hereby granted, provided that the above @@ -546,23 +616,26 @@ missing/strlcpy.c ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. -missing/langinfo.c +[missing/langinfo.c] This file is from http://www.cl.cam.ac.uk/~mgk25/ucs/langinfo.c. Ruby uses a modified version. The file contains the following author/copyright notice: - Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11 - Permission to use, copy, modify, and distribute this software - for any purpose and without fee is hereby granted. The author - disclaims all warranties with regard to this software. + >>> + Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11:: + Permission to use, copy, modify, and distribute this software + for any purpose and without fee is hereby granted. The author + disclaims all warranties with regard to this software. -ext/digest/md5/md5.[ch]: +[ext/digest/md5/md5.c] +[ext/digest/md5/md5.h] These files are under the following license. Ruby uses modified versions of them. - Copyright (C) 1999, 2000 Aladdin Enterprises. All rights reserved. + >>> + Copyright (C) 1999, 2000:: Aladdin Enterprises. All rights reserved. This software is provided 'as-is', without any express or implied warranty. In no event will the authors be held liable for any damages @@ -583,23 +656,27 @@ ext/digest/md5/md5.[ch]: L. Peter Deutsch ghost@aladdin.com -ext/digest/rmd160/rmd160.[ch]: +[ext/digest/rmd160/rmd160.c] +[ext/digest/rmd160/rmd160.h] These files have the following copyright information, and by the author we are allowed to use it under the new-style BSD license. - AUTHOR: Antoon Bosselaers, ESAT-COSIC + >>> + AUTHOR:: Antoon Bosselaers, ESAT-COSIC (Arranged for libc by Todd C. Miller) - DATE: 1 March 1996 + DATE:: 1 March 1996 - Copyright (c) Katholieke Universiteit Leuven + Copyright (c):: Katholieke Universiteit Leuven 1996, All Rights Reserved -ext/digest/sha2/sha2.[ch]: +[ext/digest/sha2/sha2.c] +[ext/digest/sha2/sha2.h] These files are under the new-style BSD license. - Copyright 2000 Aaron D. Gifford. All rights reserved. + >>> + Copyright 2000:: Aaron D. Gifford. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions @@ -625,96 +702,123 @@ ext/digest/sha2/sha2.[ch]: OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. -ext/json/generator/generator.c: +[ext/json/generator/generator.c] + + The file contains the following copyright notice. - Copyright 2001-2004 Unicode, Inc. + >>> + Copyright 2001-2004:: Unicode, Inc. - Disclaimer + Disclaimer:: - This source code is provided as is by Unicode, Inc. No claims are - made as to fitness for any particular purpose. No warranties of any - kind are expressed or implied. The recipient agrees to determine - applicability of information provided. If this file has been - purchased on magnetic or optical media from Unicode, Inc., the - sole remedy for any claim will be exchange of defective media - within 90 days of receipt. + This source code is provided as is by Unicode, Inc. No claims are + made as to fitness for any particular purpose. No warranties of any + kind are expressed or implied. The recipient agrees to determine + applicability of information provided. If this file has been + purchased on magnetic or optical media from Unicode, Inc., the + sole remedy for any claim will be exchange of defective media + within 90 days of receipt. - Limitations on Rights to Redistribute This Code + Limitations on Rights to Redistribute This Code:: - Unicode, Inc. hereby grants the right to freely use the information - supplied in this file in the creation of products supporting the - Unicode Standard, and to make copies of this file in any form - for internal or external distribution as long as this notice - remains attached. + Unicode, Inc. hereby grants the right to freely use the information + supplied in this file in the creation of products supporting the + Unicode Standard, and to make copies of this file in any form + for internal or external distribution as long as this notice + remains attached. -ext/nkf/nkf-utf8/config.h: -ext/nkf/nkf-utf8/nkf.c: -ext/nkf/nkf-utf8/utf8tbl.c: +[ext/nkf/nkf-utf8/config.h] +[ext/nkf/nkf-utf8/nkf.c] +[ext/nkf/nkf-utf8/utf8tbl.c] These files are under the following license. So to speak, it is copyrighted semi-public-domain software. - Copyright (C) 1987, Fujitsu LTD. (Itaru ICHIKAWA) - Everyone is permitted to do anything on this program - including copying, modifying, improving, - as long as you don't try to pretend that you wrote it. - i.e., the above copyright notice has to appear in all copies. - Binary distribution requires original version messages. - You don't have to ask before copying, redistribution or publishing. - THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE. - -ext/psych: -test/psych: - - Copyright 2009 Aaron Patterson, et al. - - Permission is hereby granted, free of charge, to any person obtaining a copy of - this software and associated documentation files (the 'Software'), to deal in - the Software without restriction, including without limitation the rights to - use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies - of the Software, and to permit persons to whom the Software is furnished to do - so, subject to the following conditions: - - The above copyright notice and this permission notice shall be included in all - copies or substantial portions of the Software. - - THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - SOFTWARE. - -ext/psych/yaml: - - Copyright (c) 2006 Kirill Simonov - - Permission is hereby granted, free of charge, to any person obtaining a copy of - this software and associated documentation files (the "Software"), to deal in - the Software without restriction, including without limitation the rights to - use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies - of the Software, and to permit persons to whom the Software is furnished to do - so, subject to the following conditions: - - The above copyright notice and this permission notice shall be included in all - copies or substantial portions of the Software. - - THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, - OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE - SOFTWARE. - -ext/socket/addrinfo.h: -ext/socket/getaddrinfo.c: -ext/socket/getnameinfo.c: + >>> + Copyright (C) 1987:: Fujitsu LTD. (Itaru ICHIKAWA) + + Everyone is permitted to do anything on this program + including copying, modifying, improving, + as long as you don't try to pretend that you wrote it. + i.e., the above copyright notice has to appear in all copies. + Binary distribution requires original version messages. + You don't have to ask before copying, redistribution or publishing. + THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE. + +[ext/psych] +[test/psych] + + The files under these directories are under the following license, except for + ext/psych/yaml. + + >>> + Copyright 2009:: Aaron Patterson, et al. + + Permission is hereby granted, free of charge, to any person obtaining a copy of + this software and associated documentation files (the 'Software'), to deal in + the Software without restriction, including without limitation the rights to + use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies + of the Software, and to permit persons to whom the Software is furnished to do + so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in all + copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. + +[ext/psych/yaml] + + The files under this directory are under the following license. + + >>> + Copyright (c) 2006:: Kirill Simonov + + Permission is hereby granted, free of charge, to any person obtaining a copy of + this software and associated documentation files (the "Software"), to deal in + the Software without restriction, including without limitation the rights to + use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies + of the Software, and to permit persons to whom the Software is furnished to do + so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in all + copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. + +[ext/pty/pty.c] + + >>> + C) Copyright 1998:: by Akinori Ito. + + This software may be redistributed freely for this purpose, in full + or in part, provided that this entire copyright notice is included + on any copies of this software and applications and derivations thereof. + + This software is provided on an "as is" basis, without warranty of any + kind, either expressed or implied, as to any matter including, but not + limited to warranty of fitness of purpose, or merchantability, or + results obtained from use of this software. + +[ext/socket/addrinfo.h] +[ext/socket/getaddrinfo.c] +[ext/socket/getnameinfo.c] These files are under the new-style BSD license. - Copyright (C) 1995, 1996, 1997, 1998, and 1999 WIDE Project. + >>> + Copyright (C) 1995, 1996, 1997, 1998, and 1999:: WIDE Project. All rights reserved. Redistribution and use in source and binary forms, with or without @@ -741,31 +845,35 @@ ext/socket/getnameinfo.c: OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. -ext/win32ole/win32ole.c: +[ext/win32ole/win32ole.c] You can apply the Artistic License to this file. (or GPL, alternatively) - (c) 1995 Microsoft Corporation. All rights reserved. + >>> + (c) 1995:: Microsoft Corporation. All rights reserved. Developed by ActiveWare Internet Corp., http://www.ActiveWare.com - Other modifications Copyright (c) 1997, 1998 by Gurusamy Sarathy + Other modifications Copyright (c) 1997, 1998:: by Gurusamy Sarathy <gsar@umich.edu> and Jan Dubois <jan.dubois@ibm.net> You may distribute under the terms of either the GNU General Public License or the Artistic License, as specified in the README file of the Perl distribution. -lib/rdoc/generator/template/darkfish/css/fonts.css: + The Wayback Machine url: http://web.archive.org/web/19970607104352/http://www.activeware.com:80/ + +[lib/rdoc/generator/template/darkfish/css/fonts.css] - This file is licensed under the SIL Open Font License. + This file is licensed under the {SIL Open Font License}[http://scripts.sil.org/OFL]. - http://scripts.sil.org/OFL +[spec/mspec] +[spec/ruby] -spec/mspec: -spec/ruby: + The files under these directories are under the following license. - Copyright (c) 2008 Engine Yard, Inc. All rights reserved. + >>> + Copyright (c) 2008:: Engine Yard, Inc. All rights reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation @@ -788,60 +896,224 @@ spec/ruby: FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -lib/rubygems.rb: -lib/rubygems: -test/rubygems: +[lib/rubygems.rb] +[lib/rubygems] +[test/rubygems] + + RubyGems is under the following license. + + >>> + RubyGems is copyrighted free software by Chad Fowler, Rich Kilmer, Jim + Weirich and others. You can redistribute it and/or modify it under + either the terms of the {MIT license}[rdoc-label:label-MIT+License], or the conditions + below: - RubyGems is copyrighted free software by Chad Fowler, Rich Kilmer, Jim - Weirich and others. You can redistribute it and/or modify it under - either the terms of the MIT license (see the file MIT.txt), or the - conditions below: + 1. You may make and give away verbatim copies of the source form of the + software without restriction, provided that you duplicate all of the + original copyright notices and associated disclaimers. - 1. You may make and give away verbatim copies of the source form of the - software without restriction, provided that you duplicate all of the - original copyright notices and associated disclaimers. + 2. You may modify your copy of the software in any way, provided that + you do at least ONE of the following: - 2. You may modify your copy of the software in any way, provided that - you do at least ONE of the following: + a. place your modifications in the Public Domain or otherwise + make them Freely Available, such as by posting said + modifications to Usenet or an equivalent medium, or by allowing + the author to include your modifications in the software. - a. place your modifications in the Public Domain or otherwise - make them Freely Available, such as by posting said - modifications to Usenet or an equivalent medium, or by allowing - the author to include your modifications in the software. + b. use the modified software only within your corporation or + organization. - b. use the modified software only within your corporation or - organization. + c. give non-standard executables non-standard names, with + instructions on where to get the original software distribution. - c. give non-standard executables non-standard names, with - instructions on where to get the original software distribution. + d. make other distribution arrangements with the author. - d. make other distribution arrangements with the author. + 3. You may distribute the software in object code or executable + form, provided that you do at least ONE of the following: - 3. You may distribute the software in object code or executable - form, provided that you do at least ONE of the following: + a. distribute the executables and library files of the software, + together with instructions (in the manual page or equivalent) + on where to get the original distribution. - a. distribute the executables and library files of the software, - together with instructions (in the manual page or equivalent) - on where to get the original distribution. + b. accompany the distribution with the machine-readable source of + the software. - b. accompany the distribution with the machine-readable source of - the software. + c. give non-standard executables non-standard names, with + instructions on where to get the original software distribution. - c. give non-standard executables non-standard names, with - instructions on where to get the original software distribution. + d. make other distribution arrangements with the author. - d . make other distribution arrangements with the author. + 4. You may modify and include the part of the software into any other + software (possibly commercial). - 4. You may modify and include the part of the software into any other - software (possibly commercial). + 5. The scripts and library files supplied as input to or produced as + output from the software do not automatically fall under the + copyright of the software, but belong to whomever generated them, + and may be sold commercially, and may be aggregated with this + software. - 5. The scripts and library files supplied as input to or produced as - output from the software do not automatically fall under the - copyright of the software, but belong to whomever generated them, - and may be sold commercially, and may be aggregated with this - software. + 6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR + IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + PURPOSE. - 6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR - IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED - WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR - PURPOSE. +[lib/bundler] +[lib/bundler.rb] +[spec/bundler] + + Bundler is under the following license. + + >>> + Portions copyright (c) 2010:: Andre Arko + Portions copyright (c) 2009:: Engine Yard + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/bundler/vendor/thor] + + Thor is under the following license. + + >>> + Copyright (c) 2008 Yehuda Katz, Eric Hodel, et al. + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/rubygems/resolver/molinillo] + + molinillo is under the following license. + + >>> + Copyright (c) 2014 Samuel E. Giddins segiddins@segiddins.me + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/bundler/vendor/pub_grub] + + pub_grub is under the following license. + + >>> + Copyright (c) 2018 John Hawthorn + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/bundler/vendor/connection_pool] + + connection_pool is under the following license. + + >>> + Copyright (c) 2011 Mike Perham + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/bundler/vendor/net-http-persistent] + + net-http-persistent is under the following license. + + >>> + Copyright (c) Eric Hodel, Aaron Patterson + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/did_you_mean] +[lib/did_you_mean.rb] +[test/did_you_mean] + + did_you_mean is under the following license. + + >>> + Copyright (c) 2014-2016 Yuki Nishijima + + {MIT License}[rdoc-label:label-MIT+License] + +[lib/error_highlight] +[lib/error_highlight.rb] +[test/error_highlight] + + error_highlight is under the following license. + + >>> + Copyright (c) 2021 Yusuke Endoh + + {MIT License}[rdoc-label:label-MIT+License] + +[benchmark/so_ackermann.rb] +[benchmark/so_array.rb] +[benchmark/so_binary_trees.rb] +[benchmark/so_concatenate.rb] +[benchmark/so_count_words.yml] +[benchmark/so_exception.rb] +[benchmark/so_fannkuch.rb] +[benchmark/so_fasta.rb] +[benchmark/so_k_nucleotide.yml] +[benchmark/so_lists.rb] +[benchmark/so_mandelbrot.rb] +[benchmark/so_matrix.rb] +[benchmark/so_meteor_contest.rb] +[benchmark/so_nbody.rb] +[benchmark/so_nested_loop.rb] +[benchmark/so_nsieve.rb] +[benchmark/so_nsieve_bits.rb] +[benchmark/so_object.rb] +[benchmark/so_partial_sums.rb] +[benchmark/so_pidigits.rb] +[benchmark/so_random.rb] +[benchmark/so_reverse_complement.yml] +[benchmark/so_sieve.rb] +[benchmark/so_spectralnorm.rb] + + These files are very old copy of then-called "The Great Computer Language + Shootout". LEGAL SITUATION OF THESE FILES ARE UNCLEAR because the original + site has been lost. Upstream diverged to delete several benchmarks listed + above. + +== MIT License +>>> + Permission is hereby granted, free of charge, to any person obtaining + a copy of this software and associated documentation files (the + "Software"), to deal in the Software without restriction, including + without limitation the rights to use, copy, modify, merge, publish, + distribute, sublicense, and/or sell copies of the Software, and to + permit persons to whom the Software is furnished to do so, subject to + the following conditions: + + The above copyright notice and this permission notice shall be + included in all copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE + LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION + OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION + WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +== Old-style BSD license +>>> + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + 3. Neither the name of the University nor the names of its contributors + may be used to endorse or promote products derived from this software + without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + SUCH DAMAGE. + + IMPORTANT NOTE:: + + From ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change + paragraph 3 above is now null and void. diff --git a/Makefile.in b/Makefile.in deleted file mode 100644 index 48dd90d32c..0000000000 --- a/Makefile.in +++ /dev/null @@ -1,531 +0,0 @@ -SHELL = /bin/sh -NULLCMD = @NULLCMD@ -n=$(NULLCMD) -ECHO1 = $(V:1=$n) -RUNCMD = $(SHELL) -CDPATH = . -CHDIR = @CHDIR@ -exec = exec -NULL = /dev/null -PATH_SEPARATOR = @PATH_SEPARATOR@ - -#### Start of system configuration section. #### - -srcdir = @srcdir@ -top_srcdir = $(srcdir) -hdrdir = $(srcdir)/include -PLATFORM_DIR = @PLATFORM_DIR@ - -CC = @CC@ -CPP = @CPP@ -LD = @LD@ -YACC = bison -PURIFY = -AUTOCONF = autoconf -ACLOCAL = aclocal -CONFIGURE = @CONFIGURE@ -@SET_MAKE@ -MKFILES = @MAKEFILES@ -BASERUBY = @BASERUBY@ -HAVE_BASERUBY = @HAVE_BASERUBY@ -TEST_RUNNABLE = @TEST_RUNNABLE@ -CROSS_COMPILING = @CROSS_COMPILING@ -DOXYGEN = @DOXYGEN@ - -prefix = @prefix@ -exec_prefix = @exec_prefix@ -bindir = @bindir@ -sbindir = @sbindir@ -libdir = @libdir@ -libexecdir = @libexecdir@ -datarootdir = @datarootdir@ -datadir = @datadir@ -arch = @arch@ -sitearch = @sitearch@ -sitedir = @sitedir@ -archlibdir = @archlibdir@ -ruby_version = @ruby_version@ - -TESTUI = console -TESTS = -INSTALLDOC = @INSTALLDOC@ -DOCTARGETS = @RDOCTARGET@ @CAPITARGET@ - -EXTOUT = @EXTOUT@ -arch_hdrdir = $(EXTOUT)/include/$(arch) -VPATH = $(arch_hdrdir)/ruby:$(hdrdir)/ruby:$(srcdir):$(srcdir)/missing - -empty = -CC_VERSION = @CC_VERSION@ -OUTFLAG = @OUTFLAG@$(empty) -COUTFLAG = @COUTFLAG@$(empty) -ARCH_FLAG = @ARCH_FLAG@ -CFLAGS_NO_ARCH = @CFLAGS@ -CFLAGS = $(CFLAGS_NO_ARCH) $(ARCH_FLAG) -cflags = @cflags@ -optflags = @optflags@ -debugflags = @debugflags@ -warnflags = @warnflags@ @strict_warnflags@ -cppflags = @cppflags@ -XCFLAGS = @XCFLAGS@ -CPPFLAGS = @CPPFLAGS@ $(INCFLAGS) -LDFLAGS = @STATIC@ $(CFLAGS) @LDFLAGS@ -EXTLDFLAGS = @EXTLDFLAGS@ -XLDFLAGS = @XLDFLAGS@ $(EXTLDFLAGS) -EXTLIBS = -LIBS = @LIBS@ $(EXTLIBS) -MISSING = @LIBOBJS@ @ALLOCA@ -ENABLE_SHARED = @ENABLE_SHARED@ -LDSHARED = @LIBRUBY_LDSHARED@ -DLDFLAGS = @LIBRUBY_DLDFLAGS@ $(XLDFLAGS) $(ARCH_FLAG) -SOLIBS = @SOLIBS@ -ENABLE_DEBUG_ENV = @ENABLE_DEBUG_ENV@ -MAINLIBS = @MAINLIBS@ -ARCHMINIOBJS = @MINIOBJS@ -DLNOBJ = @DLNOBJ@ -ENCOBJS = @ENCOBJS@ -EXTOBJS = @EXTOBJS@ -BUILTIN_ENCOBJS = @BUILTIN_ENCOBJS@ -BUILTIN_TRANSSRCS = @BUILTIN_TRANSSRCS@ -BUILTIN_TRANSOBJS = @BUILTIN_TRANSOBJS@ -POSTLINK = @POSTLINK@ - -RUBY_BASE_NAME=@RUBY_BASE_NAME@ -RUBY_PROGRAM_VERSION=@RUBY_PROGRAM_VERSION@ -RUBY_API_VERSION=@RUBY_API_VERSION@ -RUBY_INSTALL_NAME=@RUBY_INSTALL_NAME@ -RUBY_SO_NAME=@RUBY_SO_NAME@ -EXEEXT = @EXEEXT@ -LIBEXT = @LIBEXT@ -PROGRAM=$(RUBY_INSTALL_NAME)$(EXEEXT) -RUBY = $(RUBY_INSTALL_NAME) -MINIRUBY = @MINIRUBY@\ - $(MINIRUBYOPT) -# RUNRUBY_COMMAND:: runruby.rb or baseruby. do not append options directly -RUNRUBY_COMMAND = @RUNRUBY_COMMAND@ -# RUNRUBY:: run ruby with RUN_OPTS which is passed to ruby -RUNRUBY = @RUNRUBY@ $(RUN_OPTS) -# RUNRUBY_DEBUGGER:: debugging option for runruby.rb -RUNRUBY_DEBUGGER = --debugger='gdb -x run.gdb --quiet --args' -XRUBY = @XRUBY@ -BTESTRUBY = @BTESTRUBY@\ - $(MINIRUBYOPT) -INSTALL = @INSTALL@ -INSTALL_DATA = @INSTALL_DATA@ -INSTALL_SCRIPT = @INSTALL_SCRIPT@ -INSTALL_PROGRAM = @INSTALL_PROGRAM@ -XRUBY_LIBDIR = @XRUBY_LIBDIR@ -XRUBY_RUBYLIBDIR = @XRUBY_RUBYLIBDIR@ -XRUBY_RUBYHDRDIR = @XRUBY_RUBYHDRDIR@ -BOOTSTRAPRUBY = @BOOTSTRAPRUBY@ - -#### End of system configuration section. #### - -MAJOR= @MAJOR@ -MINOR= @MINOR@ -TEENY= @TEENY@ -RUBY_PROGRAM_VERSION = @RUBY_PROGRAM_VERSION@ - -LIBRUBY_A = @LIBRUBY_A@ -LIBRUBY_SO = @LIBRUBY_SO@ -LIBRUBY_SONAME= @LIBRUBY_SONAME@ -LIBRUBY_ALIASES= @LIBRUBY_ALIASES@ -LIBRUBY = @LIBRUBY@ -LIBRUBYARG = @LIBRUBYARG@ -LIBRUBYARG_STATIC = @LIBRUBYARG_STATIC@ -LIBRUBYARG_SHARED = @LIBRUBYARG_SHARED@ -LIBRUBY_RELATIVE = @LIBRUBY_RELATIVE@ -LIBRUBY_A_OBJS = @LIBRUBY_A_OBJS@ - -DTRACE_REBUILD_OBJS = $(DTRACE_REBUILD:yes=$(DTRACE_DEPENDENT_OBJS)) - -DTRACE_DEPENDENT_OBJS = array.$(OBJEXT) \ - eval.$(OBJEXT) \ - gc.$(OBJEXT) \ - hash.$(OBJEXT) \ - load.$(OBJEXT) \ - object.$(OBJEXT) \ - parse.$(OBJEXT) \ - string.$(OBJEXT) \ - symbol.$(OBJEXT) \ - vm.$(OBJEXT) - -THREAD_MODEL = @THREAD_MODEL@ - -PREP = @PREP@ -ARCHFILE = @ARCHFILE@ -SETUP = -EXTSTATIC = @EXTSTATIC@ -ENCSTATIC = @ENCSTATIC@ -SET_LC_MESSAGES = env LC_MESSAGES=C - -MAKEDIRS = @MKDIR_P@ -CP = cp -MV = mv -RM = rm -f -RMDIR = @RMDIR@ -RMDIRS = @RMDIRS@ -RMALL = @RMALL@ -NM = @NM@ -AR = @AR@ -ARFLAGS = @ARFLAGS@$(empty) -RANLIB = @RANLIB@ -AS = @AS@ -ASFLAGS = @ASFLAGS@ $(INCFLAGS) -IFCHANGE = $(srcdir)/tool/ifchange -SET_LC_MESSAGES = env LC_MESSAGES=C -OBJDUMP = @OBJDUMP@ -OBJCOPY = @OBJCOPY@ -HAVE_GIT = @HAVE_GIT@ -GIT = @GIT@ -VCS = @VCS@ -VCSUP = @VCSUP@ -DTRACE = @DTRACE@ @DTRACE_OPT@ -DTRACE_EXT = @DTRACE_EXT@ -DTRACE_OBJ = @DTRACE_OBJ@ -DTRACE_REBUILD= @DTRACE_REBUILD@ -DTRACE_GLOMMED_OBJ = $(DTRACE_REBUILD:yes=ruby-glommed.$(OBJEXT)) - -OBJEXT = @OBJEXT@ -ASMEXT = S -SOEXT = @SOEXT@ -DLEXT = @DLEXT@ -MANTYPE = @MANTYPE@ -SYMBOL_PREFIX = @SYMBOL_PREFIX@ - -INSTALLED_LIST= .installed.list - -NEWLINE_C = enc/trans/newline.c -MINIPRELUDE_C = miniprelude.c -PRELUDE_C = prelude.c -RBCONFIG = .rbconfig.time - -MAINSRC = $(MAINOBJ:@OBJEXT@=c) - -SRC_FILE = $< -OS_SRC_FILE = $< -DEST_FILE = $@ -OS_DEST_FILE = $@ - -MESSAGE_BEGIN = @for line in -MESSAGE_END = ; do echo "$$line"; done -ECHO_BEGIN = @sep=''; for word in -ECHO_END = ; do echo @ECHO_N@ "$$sep'$$word'@ECHO_C@"; sep=' '; done; echo - -DESTDIR = @DESTDIR@ - -configure_args = @configure_args@ -#### End of variables - -.SUFFIXES: .inc .h .c .y .i .$(DTRACE_EXT) - -all: - -# Prevent GNU make v3 from overflowing arg limit on SysV. -.NOEXPORT: - -miniruby$(EXEEXT): - @-if test -f $@; then $(MV) -f $@ $@.old; $(RM) $@.old; fi - $(ECHO) linking $@ - $(Q) $(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(NORMALMAINOBJ) $(MINIOBJS) $(COMMONOBJS) $(MAINLIBS) $(LIBS) $(OUTFLAG)$@ - $(Q) $(POSTLINK) - -$(PROGRAM): - @$(RM) $@ - $(ECHO) linking $@ - $(Q) $(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ) $(EXTOBJS) $(LIBRUBYARG) $(MAINLIBS) $(LIBS) $(EXTLIBS) $(OUTFLAG)$@ - $(Q) $(POSTLINK) - -# We must `rm' the library each time this rule is invoked because "updating" a -# MAB library on Apple/NeXT (see --enable-fat-binary in configure) is not -# supported. -$(LIBRUBY_A): - @$(RM) $@ - $(ECHO) linking static-library $@ - $(Q) $(AR) $(ARFLAGS) $@ $(LIBRUBY_A_OBJS) $(INITOBJS) - @-$(RANLIB) $@ 2> /dev/null || true - -verify-static-library: $(LIBRUBY_A) - $(ECHO) verifying static-library $@ - @$(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ) $(LIBRUBY_A) $(MAINLIBS) $(EXTLIBS) $(LIBS) $(OUTFLAG)conftest$(EXEEXT) - @$(RMALL) conftest$(EXEEXT) conftest.c conftest.dSYM - -$(LIBRUBY_SO): - @-$(PRE_LIBRUBY_UPDATE) - $(ECHO) linking shared-library $@ - $(Q) $(LDSHARED) $(DLDFLAGS) $(OBJS) $(DLDOBJS) $(SOLIBS) $(EXTSOLIBS) $(OUTFLAG)$@ - -$(Q) $(OBJCOPY) -w -L '$(SYMBOL_PREFIX)Init_*' -L '$(SYMBOL_PREFIX)ruby_static_id_*' \ - -L '$(SYMBOL_PREFIX)*_threadptr_*' -L '$(SYMBOL_PREFIX)*_ec_*' $@ - $(Q) $(POSTLINK) - @-$(MINIRUBY) -e 'ARGV.each{|link| File.delete link rescue nil; \ - File.symlink "$(LIBRUBY_SO)", link}' \ - $(LIBRUBY_ALIASES) || true - -ruby_pc = @ruby_pc@ -$(ruby_pc): - @./config.status --file=$@:$(srcdir)/template/ruby.pc.in - -ruby-runner.h: template/ruby-runner.h.in - @./config.status --file=$@:$(srcdir)/template/$(@F).in - -$(RBCONFIG): $(PREP) - -rbconfig.rb: $(RBCONFIG) - -install-cross: $(arch)-fake.rb $(RBCONFIG) rbconfig.rb $(arch_hdrdir)/ruby/config.h \ - $(LIBRUBY_A) $(LIBRUBY_SO) $(ARCHFILE) - $(ECHO) installing cross-compiling stuff - $(Q) $(MAKEDIRS) $(XRUBY_RUBYLIBDIR)/$(arch) $(XRUBY_RUBYHDRDIR)/$(arch)/ruby - $(Q) sed '/^\$$:\.unshift/q' $(arch)-fake.rb > fake.rb - $(Q) $(BASERUBY) -p \ - -e '~/^\s*CONFIG\["LDFLAGS"\]/ and' \ - -e '$$_[/(?=\s*"$$)/] = %q[ #{(CONFIG["LIBPATHFLAG"]%File.dirname(__FILE__)).strip}]' \ - rbconfig.rb > fake-rbconfig.rb - $(INSTALL_SCRIPT) fake.rb $(XRUBY_RUBYLIBDIR)/$(arch)/fake.rb - $(INSTALL_SCRIPT) fake-rbconfig.rb $(XRUBY_RUBYLIBDIR)/$(arch)/rbconfig.rb - @$(RM) fake.rb fake-rbconfig.rb - $(INSTALL_DATA) $(arch_hdrdir)/ruby/config.h $(XRUBY_RUBYHDRDIR)/$(arch)/ruby - $(INSTALL_DATA) $(top_srcdir)/include/ruby/win32.h $(XRUBY_RUBYHDRDIR)/ruby - $(INSTALL_DATA) $(LIBRUBY) $(LIBRUBY_A) $(XRUBY_RUBYLIBDIR)/$(arch) - $(INSTALL_PROGRAM) $(LIBRUBY_SO) $(XRUBY_RUBYLIBDIR)/$(arch) - -Makefile: $(srcdir)/Makefile.in $(srcdir)/enc/Makefile.in - -$(MKFILES): config.status $(srcdir)/version.h - @[ -f $@ ] && mv $@ $@.old - MAKE=$(MAKE) $(SHELL) ./config.status $@ - @cmp $@ $@.old > /dev/null 2>&1 && echo $@ unchanged && exit 0; \ - { \ - echo "all:; -@rm -f conftest.mk"; \ - echo "conftest.mk: .force; @echo AUTO_REMAKE"; \ - echo ".force:"; \ - } > conftest.mk || exit 1; \ - $(MAKE) -f conftest.mk | grep '^AUTO_REMAKE$$' >/dev/null 2>&1 || \ - { echo "$@ updated, restart."; exit 1; } - -uncommon.mk: $(srcdir)/common.mk - sed 's/{\$$([^(){}]*)[^{}]*}//g' $< > $@ - -.PHONY: reconfig -reconfig-args = $(srcdir)/$(CONFIGURE) $(configure_args) -config.status-args = ./config.status --recheck -reconfig-exec-0 = test -t 1 && { CONFIGURE_TTY=yes; export CONFIGURE_TTY; }; exec 3>&1; exit `exec 4>&1; { "$$@" 3>&- 4>&-; echo $$? 1>&4; } | fgrep -v '(cached)' 1>&3 3>&- 4>&-` -reconfig-exec-1 = set -x; "$$@" - -reconfig config.status: $(srcdir)/$(CONFIGURE) $(srcdir)/enc/Makefile.in \ - $(srcdir)/include/ruby/version.h - @PWD= MINIRUBY="$(MINIRUBY)"; export MINIRUBY; \ - set $(SHELL) $($@-args); $(reconfig-exec-$(V)) - -$(srcdir)/$(CONFIGURE): $(srcdir)/configure.ac $(srcdir)/aclocal.m4 - $(CHDIR) $(srcdir) && exec $(AUTOCONF) -o $(@F) - -$(srcdir)/aclocal.m4: - $(CHDIR) $(srcdir) && \ - type $(ACLOCAL) >/dev/null 2>&1 && exec $(ACLOCAL); \ - touch $(@F) - -prereq: $(srcdir)/$(CONFIGURE) - -incs: id.h -all-incs: probes.h - -# Things which should be considered: -# * with gperf v.s. without gperf -# * committers may have various versions of gperf -# * ./configure v.s. ../ruby/configure -# * GNU make v.s. HP-UX make # HP-UX make invokes the action if lex.c and keywords has same mtime. -# * svn checkout generate a file with mtime as current time -# * ext4 and XFS has a mtime with fractional part -lex.c: defs/keywords - @\ - if cmp -s $(srcdir)/defs/lex.c.src $?; then \ - [ $(Q) ] && echo copying $@ || set -x; \ - $(CP) $(srcdir)/lex.c.blt $@; \ - else \ - [ $(Q) ] && echo generating $@ || set -x; \ - gperf -C -P -p -j1 -i 1 -g -o -t -N rb_reserved_word -k1,3,$$ $? \ - | sed -f $(srcdir)/tool/gperf.sed \ - > $@.tmp && \ - $(MV) $@.tmp $@ && \ - $(CP) $? $(srcdir)/defs/lex.c.src && \ - $(CP) $@ $(srcdir)/lex.c.blt; \ - fi - -JIS_PROPS_OPTIONS = -k1,3 -7 -c -j1 -i1 -t -C -P -t --ignore-case -H onig_jis_property_hash -Q onig_jis_property_pool -N onig_jis_property - -$(srcdir)/enc/jis/props.h: enc/jis/props.kwd - $(MAKEDIRS) $(@D) - @set +e; \ - if cmp -s $(?:.kwd=.src) $?; then \ - set -x; \ - $(CP) $(?:.kwd=.h.blt) $@; \ - else \ - set -x; \ - gperf $(JIS_PROPS_OPTIONS) $? | \ - sed -f $(srcdir)/tool/gperf.sed > $@ && \ - $(CP) $? $(?:.kwd=.src) && \ - $(CP) $@ $(?:.kwd=.h.blt); \ - fi - -.c.@OBJEXT@: - @$(ECHO) compiling $< - $(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -c $< - -.s.@OBJEXT@: - @$(ECHO) assembling $< - $(Q) $(AS) $(ASFLAGS) -o $@ $< - -.c.S: - @$(ECHO) translating $< - $(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -S $< - -.c.i: - @$(ECHO) preprocessing $< - $(Q) $(CPP) $(warnflags) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -E $< > $@ - -.d.h: - @$(ECHO) translating probes $< - $(Q) $(DTRACE) -o $@.tmp -h -C $(INCFLAGS) -s $< - $(Q) sed -e 's/RUBY_/RUBY_DTRACE_/g' -e 's/PROBES_H_TMP/RUBY_PROBES_H/' -e 's/(char \*/(const char */g' -e 's/, char \*/, const char */g' $@.tmp > $@ - $(Q) $(RM) $@.tmp - -.dmyh.h: - @$(ECHO) making dummy $(DEST_FILE) - $(Q)echo '#include "$(*F).dmyh"' > $@ - -probes.stamp: $(DTRACE_REBUILD_OBJS) - $(Q) if test -f $@ -o -f probes.$(OBJEXT); then \ - $(RM) $(DTRACE_REBUILD_OBJS) $@; \ - $(ECHO0) "rebuilding objects which were modified by \"dtrace -G\""; \ - $(MAKE) $(DTRACE_REBUILD_OBJS); \ - fi - $(Q) touch $@ - -probes.@OBJEXT@: $(srcdir)/probes.d $(DTRACE_REBUILD:yes=probes.stamp) - @$(ECHO) processing probes in object files - $(Q) $(RM) $@ - $(Q) $(DTRACE) -G -C $(INCFLAGS) -s $(srcdir)/probes.d -o $@ $(DTRACE_REBUILD_OBJS) - -# DTrace static library hacks described here: -# http://mail.opensolaris.org/pipermail/dtrace-discuss/2005-August/000207.html -ruby-glommed.$(OBJEXT): - @$(ECHO) generating a glommed object with DTrace probes for static library - $(Q) $(LD) -r -o $@ $(OBJS) - -clean-local:: - $(Q)$(RM) ext/extinit.c ext/extinit.$(OBJEXT) ext/ripper/y.output \ - enc/encinit.c enc/encinit.$(OBJEXT) - -$(Q)$(RM) $(pkgconfig_DATA) - -$(Q)$(RMALL) exe/ ruby-runner.h *.dSYM - -distclean-local:: - $(Q)$(RM) ext/config.cache $(RBCONFIG) Doxyfile - -$(Q)$(RM) run.gdb - -$(Q)$(RM) $(INSTALLED_LIST) $(arch_hdrdir)/ruby/config.h verconf.h - -$(Q)$(RMDIRS) $(arch_hdrdir)/ruby 2> /dev/null || true - -ext/clean.sub gems/clean.sub:: ext/clean.mk -ext/distclean.sub gems/distclean.sub:: ext/distclean.mk -ext/realclean.sub gems/realclean.sub:: ext/realclean.mk - -ext/clean.mk ext/distclean.mk ext/realclean.mk:: - -$(Q) if [ -f $(EXTS_MK) ]; then exec $(MAKE) -f $(EXTS_MK) $(@F:.mk=); fi - -ext/clean:: ext/clean.sub -ext/distclean:: ext/distclean.sub -ext/realclean:: ext/realclean.sub -gems/clean:: gems/clean.sub -gems/distclean:: gems/distclean.sub -gems/realclean:: gems/realclean.sub - -ext/clean.sub ext/distclean.sub ext/realclean.sub \ -gems/clean.sub gems/distclean.sub gems/realclean.sub:: - $(Q) set dummy `echo "${EXTS}" | tr , ' '`; shift; \ - test "$$#" = 0 && set .; \ - set dummy `\ - cd $(@D) 2>/dev/null && \ - find "$$@" \( -name Makefile -o -name exts.mk \) -print | \ - sed -n 's:^\./::;s:^:$(@D)/:;s:/[^/][^/]*$$::p' | sort -u; \ - `; shift; \ - for dir do \ - $(RM) "$$dir/exts.mk"; \ - if [ -f "$$dir/Makefile" ]; then \ - echo $(@F:.sub=)ing "$$dir"; \ - (cd "$$dir" && exec $(MAKE) $(mflags) $(@F:.sub=)); \ - fi; \ - done || true - -ext/distclean ext/realclean gems/distclean gems/realclean:: - $(Q) set dummy `echo "${EXTS}" | tr , ' '`; shift; \ - test "$$#" = 0 && set .; \ - cd $(@D) 2>/dev/null && \ - find "$$@" -type d -empty -exec $(RMDIRS) {} + 2> /dev/null || true - $(Q) $(RMDIRS) $(@D) 2> /dev/null || true - -clean-enc distclean-enc realclean-enc: - @test -f "$(ENC_MK)" || exit 0; \ - echo $(@:-enc=ing) encodings; \ - exec $(MAKE) $(MAKE_ENC) $(@:-enc=) - -ext/extinit.$(OBJEXT): ext/extinit.c $(SETUP) - $(ECHO) compiling $@ - $(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -c ext/extinit.c - -enc/encinit.$(OBJEXT): enc/encinit.c $(SETUP) - -test-bundled-gems-run: - $(Q) set -e; while read gem _; do \ - echo testing $$gem gem && \ - $(XRUBY) -C $(srcdir)/gems/src/$$gem -Ilib ../../../.bundle/bin/rake; \ - done < $(srcdir)/gems/bundled_gems - -update-src:: - @$(CHDIR) "$(srcdir)" && LC_TIME=C exec $(VCSUP) - -update-download:: update-config_files - -after-update:: prereq - -gcov: - $(Q) $(BASERUBY) $(srcdir)/tool/run-gcov.rb - -lcov: - $(Q) $(BASERUBY) $(srcdir)/tool/run-lcov.rb - -update-doclie: - $(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \ - --branch $(DOCLIE_GIT_REF) \ - $(DOCLIE_GIT_URL) doclie $(GIT_OPTS) - -update-simplecov-html: - $(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \ - --branch $(SIMPLECOV_HTML_GIT_REF) \ - $(SIMPLECOV_HTML_GIT_URL) simplecov-html $(GIT_OPTS) - -update-simplecov: - $(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \ - --branch $(SIMPLECOV_GIT_REF) \ - $(SIMPLECOV_GIT_URL) simplecov $(GIT_OPTS) - -update-coverage: update-simplecov update-simplecov-html update-doclie - -INSNS = opt_sc.inc optinsn.inc optunifs.inc insns.inc insns_info.inc \ - vmtc.inc vm.inc - -$(INSNS): $(srcdir)/insns.def vm_opts.h \ - $(srcdir)/defs/opt_operand.def $(srcdir)/defs/opt_insn_unif.def \ - $(srcdir)/tool/insns2vm.rb - $(ECHO) generating $@ - $(Q) $(BASERUBY) -Ku $(srcdir)/tool/insns2vm.rb $(INSNS2VMOPT) $@ - -verconf.h: $(RBCONFIG) - -loadpath: verconf.h - @$(CPP) $(XCFLAGS) $(CPPFLAGS) $(srcdir)/loadpath.c | \ - sed -e '1,/^const char ruby_initial_load_paths/d;/;/,$$d' \ - -e '/^ /!d;s/ *"\\0"$$//;s/" *"//g' - -un-runnable: - $(ECHO) cannot make runnable, configure with --enable-load-relative. - $(Q) exit 1 @@ -1,43 +0,0 @@ -# -*- rdoc -*- - -= NEWS for Ruby 2.6.0 - -This document is a list of user visible feature changes made between -releases except for bug fixes. - -Note that each entry is kept so brief that no reason behind or -reference information is supplied with. For a full list of changes -with all sufficient information, see the ChangeLog file or Redmine -(e.g. <tt>https://bugs.ruby-lang.org/issues/$FEATURE_OR_BUG_NUMBER</tt>) - -== Changes since the 2.5.0 release - -=== Language changes - -* $SAFE is a process global state and we can set 0 again. [Feature #14250] - -=== Core classes updates (outstanding ones only) - -* Proc - - * Proc#call doesn't change $SAFE any more. [Feature #14250] - -=== Stdlib updates (outstanding ones only) - -* Matrix - - * New method: - - * Matrix#antisymmetric? - -=== Compatibility issues (excluding feature bug fixes) - -=== Stdlib compatibility issues (excluding feature bug fixes) - -=== C API updates - -=== Supported platform changes - -=== Implementation improvements - -=== Miscellaneous changes diff --git a/NEWS.md b/NEWS.md new file mode 100644 index 0000000000..f6c3c6fc97 --- /dev/null +++ b/NEWS.md @@ -0,0 +1,820 @@ +# NEWS for Ruby 3.2.0 + +This document is a list of user-visible feature changes +since the **3.1.0** release, except for bug fixes. + +Note that each entry is kept to a minimum, see links for details. + +## Language changes + +* Anonymous rest and keyword rest arguments can now be passed as + arguments, instead of just used in method parameters. + [[Feature #18351]] + + ```ruby + def foo(*) + bar(*) + end + def baz(**) + quux(**) + end + ``` + +* A proc that accepts a single positional argument and keywords will + no longer autosplat. [[Bug #18633]] + + ```ruby + proc{|a, **k| a}.call([1, 2]) + # Ruby 3.1 and before + # => 1 + # Ruby 3.2 and after + # => [1, 2] + ``` + +* Constant assignment evaluation order for constants set on explicit + objects has been made consistent with single attribute assignment + evaluation order. With this code: + + ```ruby + foo::BAR = baz + ``` + + `foo` is now called before `baz`. Similarly, for multiple assignments + to constants, left-to-right evaluation order is used. With this + code: + + ```ruby + foo1::BAR1, foo2::BAR2 = baz1, baz2 + ``` + + The following evaluation order is now used: + + 1. `foo1` + 2. `foo2` + 3. `baz1` + 4. `baz2` + + [[Bug #15928]] + +* "Find pattern" is no longer experimental. + [[Feature #18585]] + +* Methods taking a rest parameter (like `*args`) and wishing to delegate keyword + arguments through `foo(*args)` must now be marked with `ruby2_keywords` + (if not already the case). In other words, all methods wishing to delegate + keyword arguments through `*args` must now be marked with `ruby2_keywords`, + with no exception. This will make it easier to transition to other ways of + delegation once a library can require Ruby 3+. Previously, the `ruby2_keywords` + flag was kept if the receiving method took `*args`, but this was a bug and an + inconsistency. A good technique to find the potentially-missing `ruby2_keywords` + is to run the test suite, for where it fails find the last method which must + receive keyword arguments, use `puts nil, caller, nil` there, and check each + method/block on the call chain which must delegate keywords is correctly marked + as `ruby2_keywords`. [[Bug #18625]] [[Bug #16466]] + + ```ruby + def target(**kw) + end + + # Accidentally worked without ruby2_keywords in Ruby 2.7-3.1, ruby2_keywords + # needed in 3.2+. Just like (*args, **kwargs) or (...) would be needed on + # both #foo and #bar when migrating away from ruby2_keywords. + ruby2_keywords def bar(*args) + target(*args) + end + + ruby2_keywords def foo(*args) + bar(*args) + end + + foo(k: 1) + ``` + +## Core classes updates + +Note: We're only listing outstanding class updates. + +* Fiber + + * Introduce Fiber.[] and Fiber.[]= for inheritable fiber storage. + Introduce Fiber#storage and Fiber#storage= (experimental) for + getting and resetting the current storage. Introduce + `Fiber.new(storage:)` for setting the storage when creating a + fiber. [[Feature #19078]] + + Existing Thread and Fiber local variables can be tricky to use. + Thread-local variables are shared between all fibers, making it + hard to isolate, while Fiber-local variables can be hard to + share. It is often desirable to define unit of execution + ("execution context") such that some state is shared between all + fibers and threads created in that context. This is what Fiber + storage provides. + + ```ruby + def log(message) + puts "#{Fiber[:request_id]}: #{message}" + end + + def handle_requests + while request = read_request + Fiber.schedule do + Fiber[:request_id] = SecureRandom.uuid + + request.messages.each do |message| + Fiber.schedule do + log("Handling #{message}") # Log includes inherited request_id. + end + end + end + end + end + ``` + + You should generally consider Fiber storage for any state which + you want to be shared implicitly between all fibers and threads + created in a given context, e.g. a connection pool, a request + id, a logger level, environment variables, configuration, etc. + +* Fiber::Scheduler + + * Introduce `Fiber::Scheduler#io_select` for non-blocking IO.select. + [[Feature #19060]] + +* IO + + * Introduce IO#timeout= and IO#timeout which can cause + IO::TimeoutError to be raised if a blocking operation exceeds the + specified timeout. [[Feature #18630]] + + ```ruby + STDIN.timeout = 1 + STDIN.read # => Blocking operation timed out! (IO::TimeoutError) + ``` + + * Introduce `IO.new(..., path:)` and promote `File#path` to `IO#path`. + [[Feature #19036]] + +* Class + + * Class#attached_object, which returns the object for which + the receiver is the singleton class. Raises TypeError if the + receiver is not a singleton class. + [[Feature #12084]] + + ```ruby + class Foo; end + + Foo.singleton_class.attached_object #=> Foo + Foo.new.singleton_class.attached_object #=> #<Foo:0x000000010491a370> + Foo.attached_object #=> TypeError: `Foo' is not a singleton class + nil.singleton_class.attached_object #=> TypeError: `NilClass' is not a singleton class + ``` + +* Data + + * New core class to represent simple immutable value object. The class is + similar to Struct and partially shares an implementation, but has more + lean and strict API. [[Feature #16122]] + + ```ruby + Measure = Data.define(:amount, :unit) + distance = Measure.new(100, 'km') #=> #<data Measure amount=100, unit="km"> + weight = Measure.new(amount: 50, unit: 'kg') #=> #<data Measure amount=50, unit="kg"> + weight.with(amount: 40) #=> #<data Measure amount=40, unit="kg"> + weight.amount #=> 50 + weight.amount = 40 #=> NoMethodError: undefined method `amount=' + ``` + +* Encoding + + * Encoding#replicate has been deprecated and will be removed in 3.3. [[Feature #18949]] + * The dummy `Encoding::UTF_16` and `Encoding::UTF_32` encodings no longer + try to dynamically guess the endian based on a byte order mark. + Use `Encoding::UTF_16BE`/`UTF_16LE` and `Encoding::UTF_32BE`/`UTF_32LE` instead. + This change speeds up getting the encoding of a String. [[Feature #18949]] + * Limit maximum encoding set size by 256. + If exceeding maximum size, `EncodingError` will be raised. [[Feature #18949]] + +* Enumerator + + * Enumerator.product has been added. Enumerator::Product is the implementation. [[Feature #18685]] + +* Exception + + * Exception#detailed_message has been added. + The default error printer calls this method on the Exception object + instead of #message. [[Feature #18564]] + +* Hash + + * Hash#shift now always returns nil if the hash is + empty, instead of returning the default value or + calling the default proc. [[Bug #16908]] + +* Integer + + * Integer#ceildiv has been added. [[Feature #18809]] + +* Kernel + + * Kernel#binding raises RuntimeError if called from a non-Ruby frame + (such as a method defined in C). [[Bug #18487]] + +* MatchData + + * MatchData#byteoffset has been added. [[Feature #13110]] + * MatchData#deconstruct has been added. [[Feature #18821]] + * MatchData#deconstruct_keys has been added. [[Feature #18821]] + +* Module + + * Module.used_refinements has been added. [[Feature #14332]] + * Module#refinements has been added. [[Feature #12737]] + * Module#const_added has been added. [[Feature #17881]] + * Module#undefined_instance_methods has been added. [[Feature #12655]] + +* Proc + + * Proc#dup returns an instance of subclass. [[Bug #17545]] + * Proc#parameters now accepts lambda keyword. [[Feature #15357]] + +* Process + * Added `RLIMIT_NPTS` constant to FreeBSD platform + +* Regexp + + * The cache-based optimization is introduced. + Many (but not all) Regexp matching is now in linear time, which + will prevent regular expression denial of service (ReDoS) + vulnerability. [[Feature #19104]] + + * Regexp.linear_time? is introduced. [[Feature #19194]] + + * Regexp.new now supports passing the regexp flags not only as an Integer, + but also as a String. Unknown flags raise ArgumentError. + Otherwise, anything other than `true`, `false`, `nil` or Integer will be warned. + [[Feature #18788]] + + * Regexp.timeout= has been added. Also, Regexp.new new supports timeout keyword. + See [[Feature #17837]] + +* Refinement + + * Refinement#refined_class has been added. [[Feature #12737]] + +* RubyVM::AbstractSyntaxTree + + * Add `error_tolerant` option for `parse`, `parse_file` and `of`. [[Feature #19013]] + With this option + + 1. SyntaxError is suppressed + 2. AST is returned for invalid input + 3. `end` is complemented when a parser reaches to the end of input but `end` is insufficient + 4. `end` is treated as keyword based on indent + + ```ruby + # Without error_tolerant option + root = RubyVM::AbstractSyntaxTree.parse(<<~RUBY) + def m + a = 10 + if + end + RUBY + # => <internal:ast>:33:in `parse': syntax error, unexpected `end' (SyntaxError) + + # With error_tolerant option + root = RubyVM::AbstractSyntaxTree.parse(<<~RUBY, error_tolerant: true) + def m + a = 10 + if + end + RUBY + p root # => #<RubyVM::AbstractSyntaxTree::Node:SCOPE@1:0-4:3> + + # `end` is treated as keyword based on indent + root = RubyVM::AbstractSyntaxTree.parse(<<~RUBY, error_tolerant: true) + module Z + class Foo + foo. + end + + def bar + end + end + RUBY + p root.children[-1].children[-1].children[-1].children[-2..-1] + # => [#<RubyVM::AbstractSyntaxTree::Node:CLASS@2:2-4:5>, #<RubyVM::AbstractSyntaxTree::Node:DEFN@6:2-7:5>] + ``` + + * Add `keep_tokens` option for `parse`, `parse_file` and `of`. Add `#tokens` and `#all_tokens` + for RubyVM::AbstractSyntaxTree::Node [[Feature #19070]] + + ```ruby + root = RubyVM::AbstractSyntaxTree.parse("x = 1 + 2", keep_tokens: true) + root.tokens # => [[0, :tIDENTIFIER, "x", [1, 0, 1, 1]], [1, :tSP, " ", [1, 1, 1, 2]], ...] + root.tokens.map{_1[2]}.join # => "x = 1 + 2" + ``` + +* Set + + * Set is now available as a built-in class without the need for `require "set"`. [[Feature #16989]] + It is currently autoloaded via the Set constant or a call to Enumerable#to_set. + +* String + + * String#byteindex and String#byterindex have been added. [[Feature #13110]] + * Update Unicode to Version 15.0.0 and Emoji Version 15.0. [[Feature #18639]] + (also applies to Regexp) + * String#bytesplice has been added. [[Feature #18598]] + * String#dedup has been added as an alias to String#-@. [[Feature #18595]] + +* Struct + + * A Struct class can also be initialized with keyword arguments + without `keyword_init: true` on Struct.new [[Feature #16806]] + + ```ruby + Post = Struct.new(:id, :name) + Post.new(1, "hello") #=> #<struct Post id=1, name="hello"> + # From Ruby 3.2, the following code also works without keyword_init: true. + Post.new(id: 1, name: "hello") #=> #<struct Post id=1, name="hello"> + ``` + +* Thread + + * Thread.each_caller_location is added. [[Feature #16663]] + +* Thread::Queue + + * Thread::Queue#pop(timeout: sec) is added. [[Feature #18774]] + +* Thread::SizedQueue + + * Thread::SizedQueue#pop(timeout: sec) is added. [[Feature #18774]] + * Thread::SizedQueue#push(timeout: sec) is added. [[Feature #18944]] + +* Time + + * Time#deconstruct_keys is added, allowing to use Time instances + in pattern-matching expressions [[Feature #19071]] + + * Time.new now can parse a string like generated by Time#inspect + and return a Time instance based on the given argument. + [[Feature #18033]] + +* SyntaxError + * SyntaxError#path has been added. [[Feature #19138]] + +* TracePoint + + * TracePoint#binding now returns `nil` for `c_call`/`c_return` TracePoints. + [[Bug #18487]] + * TracePoint#enable `target_thread` keyword argument now defaults to the + current thread if a block is given and `target` and `target_line` keyword + arguments are not passed. [[Bug #16889]] + +* UnboundMethod + + * `UnboundMethod#==` returns `true` if the actual method is same. For example, + `String.instance_method(:object_id) == Array.instance_method(:object_id)` + returns `true`. [[Feature #18798]] + + * `UnboundMethod#inspect` does not show the receiver of `instance_method`. + For example `String.instance_method(:object_id).inspect` returns + `"#<UnboundMethod: Kernel#object_id()>"` + (was `"#<UnboundMethod: String(Kernel)#object_id()>"`). + +* GC + + * Expose `need_major_gc` via `GC.latest_gc_info`. [GH-6791] + +* ObjectSpace + + * `ObjectSpace.dump_all` dump shapes as well. [GH-6868] + +## Stdlib updates + +* Bundler + + * Bundler now uses [PubGrub] resolver instead of [Molinillo] for performance improvement. + * Add --ext=rust support to bundle gem for creating simple gems with Rust extensions. + [[GH-rubygems-6149]] + * Make cloning git repos faster [[GH-rubygems-4475]] + +* RubyGems + + * Add mswin support for cargo builder. [[GH-rubygems-6167]] + +* CGI + + * `CGI.escapeURIComponent` and `CGI.unescapeURIComponent` are added. + [[Feature #18822]] + +* Coverage + + * `Coverage.setup` now accepts `eval: true`. By this, `eval` and related methods are + able to generate code coverage. [[Feature #19008]] + + * `Coverage.supported?(mode)` enables detection of what coverage modes are + supported. [[Feature #19026]] + +* Date + + * Added `Date#deconstruct_keys` and `DateTime#deconstruct_keys` same as [[Feature #19071]] + +* ERB + + * `ERB::Util.html_escape` is made faster than `CGI.escapeHTML`. + * It no longer allocates a String object when no character needs to be escaped. + * It skips calling `#to_s` method when an argument is already a String. + * `ERB::Escape.html_escape` is added as an alias to `ERB::Util.html_escape`, + which has not been monkey-patched by Rails. + * `ERB::Util.url_encode` is made faster using `CGI.escapeURIComponent`. + * `-S` option is removed from `erb` command. + +* FileUtils + + * Add FileUtils.ln_sr method and `relative:` option to FileUtils.ln_s. + [[Feature #18925]] + +* IRB + + * debug.gem integration commands have been added: `debug`, `break`, `catch`, + `next`, `delete`, `step`, `continue`, `finish`, `backtrace`, `info` + * They work even if you don't have `gem "debug"` in your Gemfile. + * See also: [What's new in Ruby 3.2's IRB?](https://st0012.dev/whats-new-in-ruby-3-2-irb) + * More Pry-like commands and features have been added. + * `edit` and `show_cmds` (like Pry's `help`) are added. + * `ls` takes `-g` or `-G` option to filter out outputs. + * `show_source` is aliased from `$` and accepts unquoted inputs. + * `whereami` is aliased from `@`. + +* Net::Protocol + + * Improve `Net::BufferedIO` performance. [[GH-net-protocol-14]] + +* Pathname + + * Added `Pathname#lutime`. [[GH-pathname-20]] + +* Socket + + * Added the following constants for supported platforms. + * `SO_INCOMING_CPU` + * `SO_INCOMING_NAPI_ID` + * `SO_RTABLE` + * `SO_SETFIB` + * `SO_USER_COOKIE` + * `TCP_KEEPALIVE` + * `TCP_CONNECTION_INFO` + +* SyntaxSuggest + + * The feature of `syntax_suggest` formerly `dead_end` is integrated in Ruby. + [[Feature #18159]] + +* UNIXSocket + + * Add support for UNIXSocket on Windows. Emulate anonymous sockets. Add + support for File.socket? and File::Stat#socket? where possible. + [[Feature #19135]] + +* The following default gems are updated. + + * RubyGems 3.4.1 + * abbrev 0.1.1 + * benchmark 0.2.1 + * bigdecimal 3.1.3 + * bundler 2.4.1 + * cgi 0.3.6 + * csv 3.2.6 + * date 3.3.3 + * delegate 0.3.0 + * did_you_mean 1.6.3 + * digest 3.1.1 + * drb 2.1.1 + * english 0.7.2 + * erb 4.0.2 + * error_highlight 0.5.1 + * etc 1.4.2 + * fcntl 1.0.2 + * fiddle 1.1.1 + * fileutils 1.7.0 + * forwardable 1.3.3 + * getoptlong 0.2.0 + * io-console 0.6.0 + * io-nonblock 0.2.0 + * io-wait 0.3.0 + * ipaddr 1.2.5 + * irb 1.6.2 + * json 2.6.3 + * logger 1.5.3 + * mutex_m 0.1.2 + * net-http 0.4.0 + * net-protocol 0.2.1 + * nkf 0.1.2 + * open-uri 0.3.0 + * open3 0.1.2 + * openssl 3.1.0 + * optparse 0.3.1 + * ostruct 0.5.5 + * pathname 0.2.1 + * pp 0.4.0 + * pstore 0.1.2 + * psych 5.0.1 + * racc 1.6.2 + * rdoc 6.5.0 + * readline-ext 0.1.5 + * reline 0.3.2 + * resolv 0.2.2 + * resolv-replace 0.1.1 + * securerandom 0.2.2 + * set 1.0.3 + * stringio 3.0.4 + * strscan 3.0.5 + * syntax_suggest 1.0.2 + * syslog 0.1.1 + * tempfile 0.1.3 + * time 0.2.1 + * timeout 0.3.1 + * tmpdir 0.1.3 + * tsort 0.1.1 + * un 0.2.1 + * uri 0.12.0 + * weakref 0.1.2 + * win32ole 1.8.9 + * yaml 0.2.1 + * zlib 3.0.0 + +* The following bundled gems are updated. + + * minitest 5.16.3 + * power_assert 2.0.3 + * test-unit 3.5.7 + * net-ftp 0.2.0 + * net-imap 0.3.4 + * net-pop 0.1.2 + * net-smtp 0.3.3 + * rbs 2.8.2 + * typeprof 0.21.3 + * debug 1.7.1 + +See GitHub releases like [GitHub Releases of Logger](https://github.com/ruby/logger/releases) or changelog for details of the default gems or bundled gems. + +## Supported platforms + +* WebAssembly/WASI is added. See [wasm/README.md] and [ruby.wasm] for more details. [[Feature #18462]] + +## Compatibility issues + +* `String#to_c` currently treat a sequence of underscores as an end of Complex + string. [[Bug #19087]] + +* Now `ENV.clone` raises `TypeError` as well as `ENV.dup` [[Bug #17767]] + +### Removed constants + +The following deprecated constants are removed. + +* `Fixnum` and `Bignum` [[Feature #12005]] +* `Random::DEFAULT` [[Feature #17351]] +* `Struct::Group` +* `Struct::Passwd` + +### Removed methods + +The following deprecated methods are removed. + +* `Dir.exists?` [[Feature #17391]] +* `File.exists?` [[Feature #17391]] +* `Kernel#=~` [[Feature #15231]] +* `Kernel#taint`, `Kernel#untaint`, `Kernel#tainted?` + [[Feature #16131]] +* `Kernel#trust`, `Kernel#untrust`, `Kernel#untrusted?` + [[Feature #16131]] +* `Method#public?`, `Method#private?`, `Method#protected?`, + `UnboundMethod#public?`, `UnboundMethod#private?`, `UnboundMethod#protected?` + [[Bug #18729]] [[Bug #18751]] [[Bug #18435]] + +### Source code incompatibility of extension libraries + +* Extension libraries provide PRNG, subclasses of Random, need updates. + See [PRNG update] below for more information. [[Bug #19100]] + +### Error printer + +* Ruby no longer escapes control characters and backslashes in an + error message. [[Feature #18367]] + +### Constant lookup when defining a class/module + +* When defining a class/module directly under the Object class by class/module + statement, if there is already a class/module defined by `Module#include` + with the same name, the statement was handled as "open class" in Ruby 3.1 or before. + Since Ruby 3.2, a new class is defined instead. [[Feature #18832]] + +## Stdlib compatibility issues + +* Psych no longer bundles libyaml sources. + And also Fiddle no longer bundles libffi sources. + Users need to install the libyaml/libffi library themselves via the package + manager like apt, yum, brew, etc. + + Psych and fiddle supported the static build with specific version of libyaml + and libffi sources. You can build psych with libyaml-0.2.5 like this. + + ```bash + $ ./configure --with-libyaml-source-dir=/path/to/libyaml-0.2.5 + ``` + + And you can build fiddle with libffi-3.4.4 like this. + + ```bash + $ ./configure --with-libffi-source-dir=/path/to/libffi-3.4.4 + ``` + + [[Feature #18571]] + +* Check cookie name/path/domain characters in `CGI::Cookie`. [[CVE-2021-33621]] + +* `URI.parse` return empty string in host instead of nil. [[sec-156615]] + +## C API updates + +### Updated C APIs + +The following APIs are updated. + +* PRNG update + + `rb_random_interface_t` in ruby/random.h updated and versioned. + Extension libraries which use this interface and built for older + versions need to rebuild with adding `init_int32` function. + +### Added C APIs + +* `VALUE rb_hash_new_capa(long capa)` was added to created hashes with the desired capacity. +* `rb_internal_thread_add_event_hook` and `rb_internal_thread_add_event_hook` were added to instrument threads scheduling. + The following events are available: + * `RUBY_INTERNAL_THREAD_EVENT_STARTED` + * `RUBY_INTERNAL_THREAD_EVENT_READY` + * `RUBY_INTERNAL_THREAD_EVENT_RESUMED` + * `RUBY_INTERNAL_THREAD_EVENT_SUSPENDED` + * `RUBY_INTERNAL_THREAD_EVENT_EXITED` +* `rb_debug_inspector_current_depth` and `rb_debug_inspector_frame_depth` are added for debuggers. + +### Removed C APIs + +The following deprecated APIs are removed. + +* `rb_cData` variable. +* "taintedness" and "trustedness" functions. [[Feature #16131]] + +## Implementation improvements + +* Fixed several race conditions in Kernel#autoload. [[Bug #18782]] +* Cache invalidation for expressions referencing constants is now + more fine-grained. `RubyVM.stat(:global_constant_state)` was + removed because it was closely tied to the previous caching scheme + where setting any constant invalidates all caches in the system. + New keys, `:constant_cache_invalidations` and `:constant_cache_misses`, + were introduced to help with use cases for `:global_constant_state`. + [[Feature #18589]] +* The cache-based optimization for Regexp matching is introduced. + [[Feature #19104]] +* [Variable Width Allocation](https://shopify.engineering/ruby-variable-width-allocation) + is now enabled by default. [[Feature #18239]] +* Added a new instance variable caching mechanism, called object shapes, which + improves inline cache hits for most objects and allows us to generate very + efficient JIT code. Objects whose instance variables are defined in a + consistent order will see the most performance benefits. + [[Feature #18776]] +* Speed up marking instruction sequences by using a bitmap to find "markable" + objects. This change results in faster major collections. + [[Feature #18875]] + +## JIT + +### YJIT + +* YJIT is no longer experimental + * Has been tested on production workloads for over a year and proven to be quite stable. +* YJIT now supports both x86-64 and arm64/aarch64 CPUs on Linux, MacOS, BSD and other UNIX platforms. + * This release brings support for Mac M1/M2, AWS Graviton and Raspberry Pi 4. +* Building YJIT now requires Rust 1.58.0+. [[Feature #18481]] + * In order to ensure that CRuby is built with YJIT, please install `rustc` >= 1.58.0 + before running `./configure` + * Please reach out to the YJIT team should you run into any issues. +* Physical memory for JIT code is lazily allocated. Unlike Ruby 3.1, + the RSS of a Ruby process is minimized because virtual memory pages + allocated by `--yjit-exec-mem-size` will not be mapped to physical + memory pages until actually utilized by JIT code. +* Introduce Code GC that frees all code pages when the memory consumption + by JIT code reaches `--yjit-exec-mem-size`. + * `RubyVM::YJIT.runtime_stats` returns Code GC metrics in addition to + existing `inline_code_size` and `outlined_code_size` keys: + `code_gc_count`, `live_page_count`, `freed_page_count`, and `freed_code_size`. +* Most of the statistics produced by `RubyVM::YJIT.runtime_stats` are now available in release builds. + * Simply run ruby with `--yjit-stats` to compute and dump stats (incurs some run-time overhead). +* YJIT is now optimized to take advantage of object shapes. [[Feature #18776]] +* Take advantage of finer-grained constant invalidation to invalidate less code when defining new constants. [[Feature #18589]] +* The default `--yjit-exec-mem-size` is changed to 64 (MiB). +* The default `--yjit-call-threshold` is changed to 30. + +### MJIT + +* The MJIT compiler is re-implemented in Ruby as `ruby_vm/mjit/compiler`. +* MJIT compiler is executed under a forked Ruby process instead of + doing it in a native thread called MJIT worker. [[Feature #18968]] + * As a result, Microsoft Visual Studio (MSWIN) is no longer supported. +* MinGW is no longer supported. [[Feature #18824]] +* Rename `--mjit-min-calls` to `--mjit-call-threshold`. +* Change default `--mjit-max-cache` back from 10000 to 100. + +[Feature #12005]: https://bugs.ruby-lang.org/issues/12005 +[Feature #12084]: https://bugs.ruby-lang.org/issues/12084 +[Feature #12655]: https://bugs.ruby-lang.org/issues/12655 +[Feature #12737]: https://bugs.ruby-lang.org/issues/12737 +[Feature #13110]: https://bugs.ruby-lang.org/issues/13110 +[Feature #14332]: https://bugs.ruby-lang.org/issues/14332 +[Feature #15231]: https://bugs.ruby-lang.org/issues/15231 +[Feature #15357]: https://bugs.ruby-lang.org/issues/15357 +[Bug #15928]: https://bugs.ruby-lang.org/issues/15928 +[Feature #16122]: https://bugs.ruby-lang.org/issues/16122 +[Feature #16131]: https://bugs.ruby-lang.org/issues/16131 +[Bug #16466]: https://bugs.ruby-lang.org/issues/16466 +[Feature #16663]: https://bugs.ruby-lang.org/issues/16663 +[Feature #16806]: https://bugs.ruby-lang.org/issues/16806 +[Bug #16889]: https://bugs.ruby-lang.org/issues/16889 +[Bug #16908]: https://bugs.ruby-lang.org/issues/16908 +[Feature #16989]: https://bugs.ruby-lang.org/issues/16989 +[Feature #17351]: https://bugs.ruby-lang.org/issues/17351 +[Feature #17391]: https://bugs.ruby-lang.org/issues/17391 +[Bug #17545]: https://bugs.ruby-lang.org/issues/17545 +[Bug #17767]: https://bugs.ruby-lang.org/issues/17767 +[Feature #17837]: https://bugs.ruby-lang.org/issues/17837 +[Feature #17881]: https://bugs.ruby-lang.org/issues/17881 +[Feature #18033]: https://bugs.ruby-lang.org/issues/18033 +[Feature #18159]: https://bugs.ruby-lang.org/issues/18159 +[Feature #18239]: https://bugs.ruby-lang.org/issues/18239#note-17 +[Feature #18351]: https://bugs.ruby-lang.org/issues/18351 +[Feature #18367]: https://bugs.ruby-lang.org/issues/18367 +[Bug #18435]: https://bugs.ruby-lang.org/issues/18435 +[Feature #18462]: https://bugs.ruby-lang.org/issues/18462 +[Feature #18481]: https://bugs.ruby-lang.org/issues/18481 +[Bug #18487]: https://bugs.ruby-lang.org/issues/18487 +[Feature #18564]: https://bugs.ruby-lang.org/issues/18564 +[Feature #18571]: https://bugs.ruby-lang.org/issues/18571 +[Feature #18585]: https://bugs.ruby-lang.org/issues/18585 +[Feature #18589]: https://bugs.ruby-lang.org/issues/18589 +[Feature #18595]: https://bugs.ruby-lang.org/issues/18595 +[Feature #18598]: https://bugs.ruby-lang.org/issues/18598 +[Bug #18625]: https://bugs.ruby-lang.org/issues/18625 +[Feature #18630]: https://bugs.ruby-lang.org/issues/18630 +[Bug #18633]: https://bugs.ruby-lang.org/issues/18633 +[Feature #18639]: https://bugs.ruby-lang.org/issues/18639 +[Feature #18685]: https://bugs.ruby-lang.org/issues/18685 +[Bug #18729]: https://bugs.ruby-lang.org/issues/18729 +[Bug #18751]: https://bugs.ruby-lang.org/issues/18751 +[Feature #18774]: https://bugs.ruby-lang.org/issues/18774 +[Feature #18776]: https://bugs.ruby-lang.org/issues/18776 +[Bug #18782]: https://bugs.ruby-lang.org/issues/18782 +[Feature #18788]: https://bugs.ruby-lang.org/issues/18788 +[Feature #18798]: https://bugs.ruby-lang.org/issues/18798 +[Feature #18809]: https://bugs.ruby-lang.org/issues/18809 +[Feature #18821]: https://bugs.ruby-lang.org/issues/18821 +[Feature #18822]: https://bugs.ruby-lang.org/issues/18822 +[Feature #18824]: https://bugs.ruby-lang.org/issues/18824 +[Feature #18832]: https://bugs.ruby-lang.org/issues/18832 +[Feature #18875]: https://bugs.ruby-lang.org/issues/18875 +[Feature #18925]: https://bugs.ruby-lang.org/issues/18925 +[Feature #18944]: https://bugs.ruby-lang.org/issues/18944 +[Feature #18949]: https://bugs.ruby-lang.org/issues/18949 +[Feature #18968]: https://bugs.ruby-lang.org/issues/18968 +[Feature #19008]: https://bugs.ruby-lang.org/issues/19008 +[Feature #19013]: https://bugs.ruby-lang.org/issues/19013 +[Feature #19026]: https://bugs.ruby-lang.org/issues/19026 +[Feature #19036]: https://bugs.ruby-lang.org/issues/19036 +[Feature #19060]: https://bugs.ruby-lang.org/issues/19060 +[Feature #19070]: https://bugs.ruby-lang.org/issues/19070 +[Feature #19071]: https://bugs.ruby-lang.org/issues/19071 +[Feature #19078]: https://bugs.ruby-lang.org/issues/19078 +[Bug #19087]: https://bugs.ruby-lang.org/issues/19087 +[Bug #19100]: https://bugs.ruby-lang.org/issues/19100 +[Feature #19104]: https://bugs.ruby-lang.org/issues/19104 +[Feature #19135]: https://bugs.ruby-lang.org/issues/19135 +[Feature #19138]: https://bugs.ruby-lang.org/issues/19138 +[Feature #19194]: https://bugs.ruby-lang.org/issues/19194 +[Molinillo]: https://github.com/CocoaPods/Molinillo +[PubGrub]: https://github.com/jhawthorn/pub_grub +[GH-net-protocol-14]: https://github.com/ruby/net-protocol/pull/14 +[GH-pathname-20]: https://github.com/ruby/pathname/pull/20 +[GH-6791]: https://github.com/ruby/ruby/pull/6791 +[GH-6868]: https://github.com/ruby/ruby/pull/6868 +[GH-rubygems-4475]: https://github.com/rubygems/rubygems/pull/4475 +[GH-rubygems-6149]: https://github.com/rubygems/rubygems/pull/6149 +[GH-rubygems-6167]: https://github.com/rubygems/rubygems/pull/6167 +[sec-156615]: https://hackerone.com/reports/156615 +[CVE-2021-33621]: https://www.ruby-lang.org/en/news/2022/11/22/http-response-splitting-in-cgi-cve-2021-33621/ +[wasm/README.md]: https://github.com/ruby/ruby/blob/master/wasm/README.md +[ruby.wasm]: https://github.com/ruby/ruby.wasm diff --git a/README.ja.md b/README.ja.md index 2902216f99..93c0131690 100644 --- a/README.ja.md +++ b/README.ja.md @@ -1,3 +1,10 @@ +[](https://github.com/ruby/ruby/actions?query=workflow%3A"MinGW") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"MJIT") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"Ubuntu") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"Windows") +[](https://ci.appveyor.com/project/ruby/ruby/branch/master) +[](https://app.travis-ci.com/ruby/ruby) + # Rubyã¨ã¯ Rubyã¯ã‚·ãƒ³ãƒ—ルã‹ã¤å¼·åŠ›ãªã‚ªãƒ–ジェクト指å‘スクリプト言語ã§ã™ï¼Ž Rubyã¯ç´”粋ãªã‚ªãƒ–ジェクト指å‘言語ã¨ã—ã¦è¨è¨ˆã•ã‚Œã¦ã„ã‚‹ã®ã§ï¼Œ @@ -10,40 +17,48 @@ Rubyã¯ãƒ†ã‚スト処ç†é–¢ä¿‚ã®èƒ½åŠ›ãªã©ã«å„ªã‚Œï¼ŒPerlã¨åŒã˜ãらㄠ* シンプルãªæ–‡æ³• * 普通ã®ã‚ªãƒ–ジェクト指å‘機能(クラス，メソッドコールãªã©) -* 特殊ãªã‚ªãƒ–ジェクト指å‘機能(Mixin, 特異メソッドãªã©) +* 特殊ãªã‚ªãƒ–ジェクト指å‘機能(Mixin，特異メソッドãªã©) * 演算åオーãƒãƒ¼ãƒãƒ¼ãƒ‰ * 例外処ç†æ©Ÿèƒ½ * イテレータã¨ã‚¯ãƒãƒ¼ã‚¸ãƒ£ * ガーベージコレクタ * ダイナミックãƒãƒ¼ãƒ‡ã‚£ãƒ³ã‚° (アーã‚テクãƒãƒ£ã«ã‚ˆã‚‹) -* 移æ¤æ€§ãŒé«˜ã„．多ãã®Unix-like/POSIX互æ›ãƒ—ラットフォーム上ã§å‹•ãã ã‘ã§ãªã，Windows， Mac OS - X，Haikuãªã©ã®ä¸Šã§ã‚‚å‹•ã cf. - https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/SupportedPlatformsJa +* 移æ¤æ€§ãŒé«˜ã„．多ãã®Unix-like/POSIX互æ›ãƒ—ラットフォーム上ã§å‹•ãã ã‘ã§ãªã，Windows， macOS， + Haikuãªã©ã®ä¸Šã§ã‚‚å‹•ã cf. + https://github.com/ruby/ruby/blob/master/doc/contributing.rdoc#platform-maintainers ## 入手法 -### FTP㧠+サードパーティーツールを使ã£ãŸæ–¹æ³•ã‚’å«ã‚€Rubyã®ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«æ–¹æ³•ã®ä¸€è¦§ã¯ -以下ã®å ´æ‰€ã«ãŠã„ã¦ã‚ã‚Šã¾ã™ï¼Ž +https://www.ruby-lang.org/ja/downloads/ -ftp://ftp.ruby-lang.org/pub/ruby/ +ã‚’å‚ç…§ã—ã¦ãã ã•ã„． -### Subversion㧠+### Git -開発先端ã®ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã¯æ¬¡ã®ã‚³ãƒžãƒ³ãƒ‰ã§å–å¾—ã§ãã¾ã™ï¼Ž +ミラーをGitHubã«å…¬é–‹ã—ã¦ã„ã¾ã™ï¼Ž 以下ã®ã‚³ãƒžãƒ³ãƒ‰ã§ãƒªãƒã‚¸ãƒˆãƒªã‚’å–å¾—ã§ãã¾ã™ï¼Ž - $ svn co https://svn.ruby-lang.org/repos/ruby/trunk/ ruby + $ git clone https://github.com/ruby/ruby.git -ä»–ã«é–‹ç™ºä¸ã®ãƒ–ランãƒã®ä¸€è¦§ã¯æ¬¡ã®ã‚³ãƒžãƒ³ãƒ‰ã§è¦‹ã‚‰ã‚Œã¾ã™ï¼Ž +ä»–ã®ãƒ–ランãƒã®ä¸€è¦§ã¯æ¬¡ã®ã‚³ãƒžãƒ³ãƒ‰ã§è¦‹ã‚‰ã‚Œã¾ã™ï¼Ž - $ svn ls https://svn.ruby-lang.org/repos/ruby/branches/ + $ git ls-remote https://github.com/ruby/ruby.git + +Rubyリãƒã‚¸ãƒˆãƒªã®æœ¬æ¥ã®master㯠https://git.ruby-lang.org/ruby.git ã«ã‚ã‚Šã¾ã™ï¼Ž +コミッタã¯ã“ã¡ã‚‰ã‚’使ã„ã¾ã™ï¼Ž + +### Subversion -### Git㧠+å¤ã„Rubyã®ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã®ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã¯æ¬¡ã®ã‚³ãƒžãƒ³ãƒ‰ã§ã‚‚å–å¾—ã§ãã¾ã™ï¼Ž -Subversionã®ãƒŸãƒ©ãƒ¼ã‚’GitHubã«å…¬é–‹ã—ã¦ã„ã¾ã™ï¼Ž 以下ã®ã‚³ãƒžãƒ³ãƒ‰ã§ãƒªãƒã‚¸ãƒˆãƒªã‚’å–å¾—ã§ãã¾ã™ï¼Ž + $ svn co https://svn.ruby-lang.org/repos/ruby/branches/ruby_2_6/ ruby + +ä»–ã®ãƒ–ランãƒã®ä¸€è¦§ã¯æ¬¡ã®ã‚³ãƒžãƒ³ãƒ‰ã§è¦‹ã‚‰ã‚Œã¾ã™ï¼Ž + + $ svn ls https://svn.ruby-lang.org/repos/ruby/branches/ - $ git clone git://github.com/ruby/ruby.git ## ホームページ @@ -55,38 +70,33 @@ https://www.ruby-lang.org/ ## メーリングリスト -Rubyã®ãƒ¡ãƒ¼ãƒªãƒ³ã‚°ãƒªã‚¹ãƒˆãŒã‚ã‚Šã¾ã™ï¼Žå‚åŠ å¸Œæœ›ã®æ–¹ã¯ - -mailto:ruby-list-request@ruby-lang.org - -ã¾ã§æœ¬æ–‡ã« +Rubyã®ãƒ¡ãƒ¼ãƒªãƒ³ã‚°ãƒªã‚¹ãƒˆãŒã‚ã‚Šã¾ã™ï¼Žå‚åŠ å¸Œæœ›ã®æ–¹ã¯ [ruby-list-request@ruby-lang.org] ã¾ã§æœ¬æ–‡ã« subscribe ã¨æ›¸ã„ã¦é€ã£ã¦ä¸‹ã•ã„． -Ruby開発者å‘ã‘メーリングリストもã‚ã‚Šã¾ã™ï¼Žã“ã¡ã‚‰ã§ã¯rubyã®ãƒã‚°ï¼Œå°†æ¥ã®ä»•æ§˜æ‹¡å¼µãªã©å®Ÿè£…上ã®å•é¡Œã«ã¤ã„ã¦è°è«–ã•ã‚Œã¦ã„ã¾ã™ï¼Ž å‚åŠ å¸Œæœ›ã®æ–¹ã¯ - -mailto:ruby-dev-request@ruby-lang.org - -ã¾ã§ruby-listã¨åŒæ§˜ã®æ–¹æ³•ã§ãƒ¡ãƒ¼ãƒ«ã—ã¦ãã ã•ã„． +Ruby開発者å‘ã‘メーリングリストもã‚ã‚Šã¾ã™ï¼Žã“ã¡ã‚‰ã§ã¯rubyã®ãƒã‚°ï¼Œå°†æ¥ã®ä»•æ§˜æ‹¡å¼µãªã©å®Ÿè£…上ã®å•é¡Œã«ã¤ã„ã¦è°è«–ã•ã‚Œã¦ã„ã¾ã™ï¼Ž +å‚åŠ å¸Œæœ›ã®æ–¹ã¯ [ruby-dev-request@ruby-lang.org] ã¾ã§ruby-listã¨åŒæ§˜ã®æ–¹æ³•ã§ãƒ¡ãƒ¼ãƒ«ã—ã¦ãã ã•ã„． Ruby拡張モジュールã«ã¤ã„ã¦è©±ã—åˆã†ruby-extメーリングリストã¨æ•°å¦é–¢ä¿‚ã®è©±é¡Œã«ã¤ã„ã¦è©±ã—åˆã†ruby-mathメーリングリスト㨠英語ã§rubyã«ã¤ã„ã¦è©±ã—åˆã†ruby-talkメーリングリストもã‚ã‚Šã¾ã™ï¼Žå‚åŠ æ–¹æ³•ã¯ã©ã‚Œã‚‚åŒã˜ã§ã™ï¼Ž +[ruby-list-request@ruby-lang.org]: mailto:ruby-list-request@ruby-lang.org?subject=Join%20Ruby%20Mailing%20List&body=subscribe +[ruby-dev-request@ruby-lang.org]: mailto:ruby-dev-request@ruby-lang.org?subject=Join%20Ruby%20Mailing%20List&body=subscribe + ## コンパイル・インストール 以下ã®æ‰‹é †ã§è¡Œã£ã¦ãã ã•ã„． -1. ã‚‚ã— `configure` ファイルãŒè¦‹ã¤ã‹ã‚‰ãªã„，もã—ã㯠`configure.ac` よりå¤ã„よã†ãªã‚‰ï¼Œ `autoconf` を実行ã—㦠- æ–°ã—ã `configure` を生æˆã™ã‚‹ +1. (Gitリãƒã‚¸ãƒˆãƒªã‹ã‚‰å–å¾—ã—ãŸã‚½ãƒ¼ã‚¹ã‚’ビルドã™ã‚‹å ´åˆ) `./autogen.sh` を実行ã—ã¦æ–°ã—ã `configure` を生æˆã™ã‚‹ 2. `configure` を実行ã—㦠`Makefile` ãªã©ã‚’生æˆã™ã‚‹ 環境ã«ã‚ˆã£ã¦ã¯ãƒ‡ãƒ•ã‚©ãƒ«ãƒˆã®Cコンパイラ用オプションãŒä»˜ãã¾ã™ï¼Ž `configure` オプション㧠`optflags=..` `warnflags=..` ç‰ã§ä¸Šæ›¸ãã§ãã¾ã™ï¼Ž -3. (å¿…è¦ãªã‚‰ã°)`defines.h` を編集ã™ã‚‹ +3. (å¿…è¦ãªã‚‰ã°)`include/ruby/defines.h` を編集ã™ã‚‹ 多分，必è¦ç„¡ã„ã¨æ€ã„ã¾ã™ï¼Ž @@ -95,7 +105,8 @@ Ruby拡張モジュールã«ã¤ã„ã¦è©±ã—åˆã†ruby-extメーリングリスム`ext/Setup` ã«è¨˜è¿°ã—ãŸãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«ã¯é™çš„ã«ãƒªãƒ³ã‚¯ã•ã‚Œã¾ã™ï¼Ž ダイナミックãƒãƒ¼ãƒ‡ã‚£ãƒ³ã‚°ã‚’サãƒãƒ¼ãƒˆã—ã¦ã„ãªã„アーã‚テクãƒãƒ£ã§ã¯ `Setup` ã®1行目ã®ã€Œ`option nodynamic`ã€ã¨ã„ã†è¡Œã®ã‚³ - メントを外ã™å¿…è¦ãŒã‚ã‚Šã¾ã™ï¼Žã¾ãŸï¼Œã“ã®ã‚¢ãƒ¼ã‚テクãƒãƒ£ã§æ‹¡å¼µãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«ã‚’利用ã™ã‚‹ãŸã‚ã«ã¯ï¼Œã‚らã‹ã˜ã‚é™çš„ã«ãƒªãƒ³ã‚¯ã‚’ã—ã¦ãŠãå¿…è¦ãŒã‚ã‚Šã¾ã™ï¼Ž + メントを外ã™å¿…è¦ãŒã‚ã‚Šã¾ã™ï¼Ž + ã¾ãŸï¼Œã“ã®ã‚¢ãƒ¼ã‚テクãƒãƒ£ã§æ‹¡å¼µãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«ã‚’利用ã™ã‚‹ãŸã‚ã«ã¯ï¼Œã‚らã‹ã˜ã‚é™çš„ã«ãƒªãƒ³ã‚¯ã‚’ã—ã¦ãŠãå¿…è¦ãŒã‚ã‚Šã¾ã™ï¼Ž 5. `make` を実行ã—ã¦ã‚³ãƒ³ãƒ‘イルã™ã‚‹ @@ -155,17 +166,17 @@ UNIXã§ã‚れ㰠`configure` ãŒã»ã¨ã‚“ã©ã®å·®ç•°ã‚’å¸åŽã—ã¦ãれる㯠## フィードãƒãƒƒã‚¯ -Rubyã«é–¢ã™ã‚‹è³ªå•ã¯ Ruby-Talk（英語）や Ruby-List（日本語） (https://www.ruby-lang.org/ja/community/mailing-lists) や， -stackoverflow (https://ja.stackoverflow.com/) ãªã©ã®Webサイトã«æŠ•ç¨¿ã—ã¦ãã ã•ã„． +Rubyã«é–¢ã™ã‚‹è³ªå•ã¯ [Ruby-Talk]（英語）や [Ruby-List]（日本語）や， +[stackoverflow] ãªã©ã®Webサイトã«æŠ•ç¨¿ã—ã¦ãã ã•ã„． ãƒã‚°å ±å‘Šã¯ https://bugs.ruby-lang.org ã§å—ã‘付ã‘ã¦ã„ã¾ã™ï¼Ž +[Ruby-Talk]: https://www.ruby-lang.org/en/community/mailing-lists +[Ruby-List]: https://www.ruby-lang.org/ja/community/mailing-lists +[stackoverflow]: https://ja.stackoverflow.com/ ## 著者 Rubyã®ã‚ªãƒªã‚¸ãƒŠãƒ«ç‰ˆã¯ï¼Œ1995å¹´ã«ã¾ã¤ã‚‚ã¨ã‚†ãã²ã‚æ°ã«ã‚ˆã£ã¦è¨è¨ˆãƒ»é–‹ç™ºã•ã‚Œã¾ã—ãŸï¼Ž <mailto:matz@ruby-lang.org> - ---- -created at: Thu Aug 3 11:57:36 JST 1995 @@ -1,175 +1,91 @@ -[](https://travis-ci.org/ruby/ruby) -[](https://ci.appveyor.com/project/ruby/ruby/branch/trunk) +[](https://github.com/ruby/ruby/actions?query=workflow%3A"MinGW") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"MJIT") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"Ubuntu") +[](https://github.com/ruby/ruby/actions?query=workflow%3A"Windows") +[](https://ci.appveyor.com/project/ruby/ruby/branch/master) +[](https://app.travis-ci.com/ruby/ruby) -# What's Ruby +# What is Ruby? -Ruby is the interpreted scripting language for quick and easy object-oriented -programming. It has many features to process text files and to do system -management tasks (as in Perl). It is simple, straight-forward, and -extensible. +Ruby is an interpreted object-oriented programming language often +used for web development. It also offers many scripting features +to process plain text and serialized files, or manage system tasks. +It is simple, straightforward, and extensible. ## Features of Ruby -* Simple Syntax -* **Normal** Object-oriented Features (e.g. class, method calls) -* **Advanced** Object-oriented Features (e.g. mix-in, singleton-method) -* Operator Overloading -* Exception Handling -* Iterators and Closures -* Garbage Collection -* Dynamic Loading of Object Files (on some architectures) -* Highly Portable (works on many Unix-like/POSIX compatible platforms as - well as Windows, macOS, Haiku, etc.) cf. - https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/SupportedPlatforms +* Simple Syntax +* **Normal** Object-oriented Features (e.g. class, method calls) +* **Advanced** Object-oriented Features (e.g. mix-in, singleton-method) +* Operator Overloading +* Exception Handling +* Iterators and Closures +* Garbage Collection +* Dynamic Loading of Object Files (on some architectures) +* Highly Portable (works on many Unix-like/POSIX compatible platforms as + well as Windows, macOS, etc.) cf. + https://github.com/ruby/ruby/blob/master/doc/maintainers.rdoc#label-Platform+Maintainers - -## How to get Ruby +## How to get Ruby with Git For a complete list of ways to install Ruby, including using third-party tools like rvm, see: https://www.ruby-lang.org/en/downloads/ -The Ruby distribution files can be found on the following FTP site: - -ftp://ftp.ruby-lang.org/pub/ruby/ - -The trunk of the Ruby source tree can be checked out with the following -command: - - $ svn co https://svn.ruby-lang.org/repos/ruby/trunk/ ruby - -Or if you are using git then use the following command: +The mirror of the Ruby source tree can be checked out with the following command: $ git clone https://github.com/ruby/ruby.git There are some other branches under development. Try the following command to see the list of branches: - $ svn ls https://svn.ruby-lang.org/repos/ruby/branches/ + $ git ls-remote https://github.com/ruby/ruby.git -Or if you are using git then use the following command: +You may also want to use https://git.ruby-lang.org/ruby.git (actual master of Ruby source) +if you are a committer. - $ git ls-remote git://github.com/ruby/ruby.git +## How to build -## Ruby home page +see [Building Ruby](doc/contributing/building_ruby.md) -The URL of the Ruby home page is: +## Ruby home page https://www.ruby-lang.org/ +## Documentation + +- [English](https://docs.ruby-lang.org/en/master/index.html) +- [Japanese](https://docs.ruby-lang.org/ja/master/index.html) + ## Mailing list -There is a mailing list to talk about Ruby. To subscribe to this list, please +There is a mailing list to discuss Ruby. To subscribe to this list, please send the following phrase: subscribe -in the mail body (not subject) to the address -<mailto:ruby-talk-request@ruby-lang.org>. - -## How to compile and install - -This is what you need to do to compile and install Ruby: - -1. If you want to use Microsoft Visual C++ to compile Ruby, read - [win32/README.win32](win32/README.win32) instead of this document. - -2. If `./configure` does not exist or is older than `configure.ac`, run - `autoconf` to (re)generate configure. - -3. Run `./configure`, which will generate `config.h` and `Makefile`. - - Some C compiler flags may be added by default depending on your - environment. Specify `optflags=..` and `warnflags=..` as necessary to - override them. - -4. Edit `defines.h` if you need. Usually this step will not be needed. - -5. Remove comment mark(`#`) before the module names from `ext/Setup` (or add - module names if not present), if you want to link modules statically. - - If you don't want to compile non static extension modules (probably on - architectures which do not allow dynamic loading), remove comment mark - from the line "`#option nodynamic`" in `ext/Setup`. +in the mail body (not subject) to the address [ruby-talk-request@ruby-lang.org]. - Usually this step will not be needed. - -6. Run `make`. - - * On Mac, set RUBY\_CODESIGN environment variable with a signing identity. - It uses the identity to sign `ruby` binary. See also codesign(1). - -7. Optionally, run '`make check`' to check whether the compiled Ruby - interpreter works well. If you see the message "`check succeeded`", your - Ruby works as it should (hopefully). - -8. Optionally, run `make update-gems` and `make extract-gems`. - - If you want to install bundled gems, run `make update-gems` and - `make extract-gems` before running `make install`. - -9. Run '`make install`'. - - This command will create the following directories and install files into - them. - - * `${DESTDIR}${prefix}/bin` - * `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}` - * `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}` - * `${DESTDIR}${prefix}/lib` - * `${DESTDIR}${prefix}/lib/ruby` - * `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}` - * `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}` - * `${DESTDIR}${prefix}/lib/ruby/site_ruby` - * `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}` - * `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}` - * `${DESTDIR}${prefix}/lib/ruby/vendor_ruby` - * `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}` - * `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}` - * `${DESTDIR}${prefix}/lib/ruby/gems/${MAJOR}.${MINOR}.${TEENY}` - * `${DESTDIR}${prefix}/share/man/man1` - * `${DESTDIR}${prefix}/share/ri/${MAJOR}.${MINOR}.${TEENY}/system` - - - If Ruby's API version is '*x.y.z*', the `${MAJOR}` is '*x*', the - `${MINOR}` is '*y*', and the `${TEENY}` is '*z*'. - - **NOTE**: teeny of the API version may be different from one of Ruby's - program version - - You may have to be a super user to install Ruby. - - -If you fail to compile Ruby, please send the detailed error report with the -error log and machine/OS type, to help others. - -Some extension libraries may not get compiled because of lack of necessary -external libraries and/or headers, then you will need to run '`make distclean-ext`' -to remove old configuration after installing them in such case. +[ruby-talk-request@ruby-lang.org]: mailto:ruby-talk-request@ruby-lang.org?subject=Join%20Ruby%20Mailing%20List&body=subscribe ## Copying -See the file [COPYING](COPYING). +See the file [COPYING](rdoc-ref:COPYING). ## Feedback -Questions about the Ruby language can be asked on the Ruby-Talk mailing list -(https://www.ruby-lang.org/en/community/mailing-lists) or on websites like -(https://stackoverflow.com). +Questions about the Ruby language can be asked on the [Ruby-Talk](https://www.ruby-lang.org/en/community/mailing-lists) mailing list +or on websites like https://stackoverflow.com. -Bug reports should be filed at https://bugs.ruby-lang.org. Read [HowToReport] for more information. - -[HowToReport]: https://bugs.ruby-lang.org/projects/ruby/wiki/HowToReport +Bugs should be reported at https://bugs.ruby-lang.org. Read ["Reporting Issues"](https://docs.ruby-lang.org/en/master/contributing/reporting_issues_md.html) for more information. ## Contributing -See the file [CONTRIBUTING.md](CONTRIBUTING.md) - +See ["Contributing to Ruby"](https://docs.ruby-lang.org/en/master/contributing_md.html), which includes setup and build instructions. ## The Author -Ruby was originally designed and developed by Yukihiro Matsumoto (Matz) in -1995. +Ruby was originally designed and developed by Yukihiro Matsumoto (Matz) in 1995. -<mailto:matz@ruby-lang.org> +<matz@ruby-lang.org> diff --git a/acinclude.m4 b/acinclude.m4 deleted file mode 100644 index 99b24e6d93..0000000000 --- a/acinclude.m4 +++ /dev/null @@ -1,46 +0,0 @@ -# -*- autoconf -*- - -AC_DEFUN([_COLORIZE_RESULT_PREPARE], [ - msg_checking= msg_result_yes= msg_result_no= msg_result_other= msg_reset= - AS_IF([test "x${CONFIGURE_TTY}" = xyes -o -t 1], [ - msg_begin="`tput smso 2>/dev/null`" - AS_CASE(["$msg_begin"], ['@<:@'*m], - [msg_begin="`echo "$msg_begin" | sed ['s/[0-9]*m$//']`" - msg_checking="${msg_begin}33m" - AS_IF([test ${TEST_COLORS:+set}], [ - msg_result_yes=[`expr ":$TEST_COLORS:" : ".*:pass=\([^:]*\):"`] - msg_result_no=[`expr ":$TEST_COLORS:" : ".*:fail=\([^:]*\):"`] - msg_result_other=[`expr ":$TEST_COLORS:" : ".*:skip=\([^:]*\):"`] - ]) - msg_result_yes="${msg_begin}${msg_result_yes:-32;1}m" - msg_result_no="${msg_begin}${msg_result_no:-31;1}m" - msg_result_other="${msg_begin}${msg_result_other:-33;1}m" - msg_reset="${msg_begin}m" - ]) - AS_UNSET(msg_begin) - ]) - AS_REQUIRE_SHELL_FN([colorize_result], - [AS_FUNCTION_DESCRIBE([colorize_result], [MSG], [Colorize result])], - [AS_CASE(["$[]1"], - [yes], [AS_ECHO(["${msg_result_yes}$[]1${msg_reset}]")], - [no], [AS_ECHO(["${msg_result_no}$[]1${msg_reset}]")], - [AS_ECHO(["${msg_result_other}$[]1${msg_reset}]")])]) -]) - -AC_DEFUN([COLORIZE_RESULT], [AC_REQUIRE([_COLORIZE_RESULT_PREPARE])dnl - AS_LITERAL_IF([$1], - [m4_case([$1], - [yes], [AS_ECHO(["${msg_result_yes}$1${msg_reset}"])], - [no], [AS_ECHO(["${msg_result_no}$1${msg_reset}"])], - [AS_ECHO(["${msg_result_other}$1${msg_reset}"])])], - [colorize_result "$1"]) dnl -]) - -AC_DEFUN([AC_CHECKING],[dnl -AC_REQUIRE([_COLORIZE_RESULT_PREPARE])dnl -AS_MESSAGE([checking ${msg_checking}$1${msg_reset}...])]) - -AC_DEFUN([AC_MSG_RESULT], [dnl -{ _AS_ECHO_LOG([result: $1]) -COLORIZE_RESULT([$1]); dnl -}]) diff --git a/aclocal.m4 b/aclocal.m4 index 18ba297b05..e69de29bb2 100644 --- a/aclocal.m4 +++ b/aclocal.m4 @@ -1,15 +0,0 @@ -# generated automatically by aclocal 1.15.1 -*- Autoconf -*- - -# Copyright (C) 1996-2017 Free Software Foundation, Inc. - -# This file is free software; the Free Software Foundation -# gives unlimited permission to copy and/or distribute it, -# with or without modifications, as long as this notice is preserved. - -# This program is distributed in the hope that it will be useful, -# but WITHOUT ANY WARRANTY, to the extent permitted by law; without -# even the implied warranty of MERCHANTABILITY or FITNESS FOR A -# PARTICULAR PURPOSE. - -m4_ifndef([AC_CONFIG_MACRO_DIRS], [m4_defun([_AM_CONFIG_MACRO_DIRS], [])m4_defun([AC_CONFIG_MACRO_DIRS], [_AM_CONFIG_MACRO_DIRS($@)])]) -m4_include([acinclude.m4]) diff --git a/addr2line.c b/addr2line.c index 09fcc3c225..e5f25293e2 100644 --- a/addr2line.c +++ b/addr2line.c @@ -8,20 +8,26 @@ **********************************************************************/ -#include "ruby/config.h" +#if defined(__clang__) +#pragma clang diagnostic ignored "-Wgnu-empty-initializer" +#pragma clang diagnostic ignored "-Wgcc-compat" +#endif + +#include "ruby/internal/config.h" +#include "ruby/defines.h" #include "ruby/missing.h" #include "addr2line.h" #include <stdio.h> #include <errno.h> -#ifdef HAVE_STDBOOL_H -#include <stdbool.h> -#else -#include "missing/stdbool.h" +#ifdef HAVE_LIBPROC_H +#include <libproc.h> #endif -#ifdef USE_ELF +#include "ruby/internal/stdbool.h" + +#if defined(USE_ELF) || defined(HAVE_MACH_O_LOADER_H) #include <fcntl.h> #include <limits.h> @@ -34,12 +40,6 @@ #include <sys/stat.h> #include <unistd.h> -#ifdef __OpenBSD__ -#include <elf_abi.h> -#else -#include <elf.h> -#endif - /* Make alloca work the best possible way. */ #ifdef __GNUC__ # ifndef alloca @@ -63,24 +63,20 @@ void *alloca(); # include <dlfcn.h> #endif -#define DW_LNS_copy 0x01 -#define DW_LNS_advance_pc 0x02 -#define DW_LNS_advance_line 0x03 -#define DW_LNS_set_file 0x04 -#define DW_LNS_set_column 0x05 -#define DW_LNS_negate_stmt 0x06 -#define DW_LNS_set_basic_block 0x07 -#define DW_LNS_const_add_pc 0x08 -#define DW_LNS_fixed_advance_pc 0x09 -#define DW_LNS_set_prologue_end 0x0a /* DWARF3 */ -#define DW_LNS_set_epilogue_begin 0x0b /* DWARF3 */ -#define DW_LNS_set_isa 0x0c /* DWARF3 */ +#ifdef HAVE_MACH_O_LOADER_H +# include <crt_externs.h> +# include <mach-o/fat.h> +# include <mach-o/loader.h> +# include <mach-o/nlist.h> +# include <mach-o/stab.h> +#endif -/* Line number extended opcode name. */ -#define DW_LNE_end_sequence 0x01 -#define DW_LNE_set_address 0x02 -#define DW_LNE_define_file 0x03 -#define DW_LNE_set_discriminator 0x04 /* DWARF4 */ +#ifdef USE_ELF +# ifdef __OpenBSD__ +# include <elf_abi.h> +# else +# include <elf.h> +# endif #ifndef ElfW # if SIZEOF_VOIDP == 8 @@ -96,17 +92,44 @@ void *alloca(); # define ELF_ST_TYPE ELF32_ST_TYPE # endif #endif +#endif + +#ifdef SHF_COMPRESSED +# if defined(ELFCOMPRESS_ZLIB) && defined(HAVE_LIBZ) + /* FreeBSD 11.0 lacks ELFCOMPRESS_ZLIB */ +# include <zlib.h> +# define SUPPORT_COMPRESSED_DEBUG_LINE +# endif +#else /* compatibility with glibc < 2.22 */ +# define SHF_COMPRESSED 0 +#endif + #ifndef PATH_MAX #define PATH_MAX 4096 #endif -#ifndef SHF_COMPRESSED /* compatibility with glibc < 2.22 */ -#define SHF_COMPRESSED 0 -#endif +#define DW_LNS_copy 0x01 +#define DW_LNS_advance_pc 0x02 +#define DW_LNS_advance_line 0x03 +#define DW_LNS_set_file 0x04 +#define DW_LNS_set_column 0x05 +#define DW_LNS_negate_stmt 0x06 +#define DW_LNS_set_basic_block 0x07 +#define DW_LNS_const_add_pc 0x08 +#define DW_LNS_fixed_advance_pc 0x09 +#define DW_LNS_set_prologue_end 0x0a /* DWARF3 */ +#define DW_LNS_set_epilogue_begin 0x0b /* DWARF3 */ +#define DW_LNS_set_isa 0x0c /* DWARF3 */ -int kprintf(const char *fmt, ...); +/* Line number extended opcode name. */ +#define DW_LNE_end_sequence 0x01 +#define DW_LNE_set_address 0x02 +#define DW_LNE_define_file 0x03 +#define DW_LNE_set_discriminator 0x04 /* DWARF4 */ -typedef struct { +PRINTF_ARGS(static int kprintf(const char *fmt, ...), 1, 2); + +typedef struct line_info { const char *dirname; const char *filename; const char *path; /* object path */ @@ -115,27 +138,72 @@ typedef struct { uintptr_t base_addr; uintptr_t saddr; const char *sname; /* function name */ + + struct line_info *next; } line_info_t; -typedef struct obj_info obj_info_t; -struct obj_info { + +struct dwarf_section { + char *ptr; + size_t size; + uint64_t flags; +}; + +typedef struct obj_info { const char *path; /* object path */ - int fd; - void *mapped; + char *mapped; size_t mapped_size; + void *uncompressed; uintptr_t base_addr; - obj_info_t *next; + uintptr_t vmaddr; + struct dwarf_section debug_abbrev; + struct dwarf_section debug_info; + struct dwarf_section debug_line; + struct dwarf_section debug_ranges; + struct dwarf_section debug_str_offsets; + struct dwarf_section debug_addr; + struct dwarf_section debug_rnglists; + struct dwarf_section debug_str; + struct dwarf_section debug_line_str; + struct obj_info *next; +} obj_info_t; + +#define DWARF_SECTION_COUNT 9 + +static struct dwarf_section * +obj_dwarf_section_at(obj_info_t *obj, int n) +{ + struct dwarf_section *ary[] = { + &obj->debug_abbrev, + &obj->debug_info, + &obj->debug_line, + &obj->debug_ranges, + &obj->debug_str_offsets, + &obj->debug_addr, + &obj->debug_rnglists, + &obj->debug_str, + &obj->debug_line_str + }; + if (n < 0 || DWARF_SECTION_COUNT <= n) { + abort(); + } + return ary[n]; +} + +struct debug_section_definition { + const char *name; + struct dwarf_section *dwarf; }; /* Avoid consuming stack as this module may be used from signal handler */ -static char binary_filename[PATH_MAX]; +static char binary_filename[PATH_MAX + 1]; static unsigned long -uleb128(char **p) +uleb128(const char **p) { unsigned long r = 0; int s = 0; for (;;) { - unsigned char b = *(unsigned char *)(*p)++; + unsigned char b = (unsigned char)*(*p)++; if (b < 0x80) { r += (unsigned long)b << s; break; @@ -147,12 +215,12 @@ uleb128(char **p) } static long -sleb128(char **p) +sleb128(const char **p) { long r = 0; int s = 0; for (;;) { - unsigned char b = *(unsigned char *)(*p)++; + unsigned char b = (unsigned char)*(*p)++; if (b < 0x80) { if (b & 0x40) { r -= (0x80 - b) << s; @@ -169,7 +237,7 @@ sleb128(char **p) } static const char * -get_nth_dirname(unsigned long dir, char *p) +get_nth_dirname(unsigned long dir, const char *p) { if (!dir--) { return ""; @@ -186,141 +254,191 @@ get_nth_dirname(unsigned long dir, char *p) return p; } +static const char *parse_ver5_debug_line_header(const char *p, int idx, uint8_t format, obj_info_t *obj, const char **out_path, uint64_t *out_directory_index); + static void -fill_filename(int file, char *include_directories, char *filenames, - line_info_t *line) +fill_filename(int file, uint8_t format, uint16_t version, const char *include_directories, const char *filenames, line_info_t *line, obj_info_t *obj) { int i; - char *p = filenames; - char *filename; + const char *p = filenames; + const char *filename; unsigned long dir; - for (i = 1; i <= file; i++) { - filename = p; - if (!*p) { - /* Need to output binary file name? */ - kprintf("Unexpected file number %d in %s\n", - file, binary_filename); - return; - } - while (*p) p++; - p++; - dir = uleb128(&p); - /* last modified. */ - uleb128(&p); - /* size of the file. */ - uleb128(&p); - - if (i == file) { - line->filename = filename; - line->dirname = get_nth_dirname(dir, include_directories); - } + if (version >= 5) { + const char *path; + uint64_t directory_index = -1; + parse_ver5_debug_line_header(filenames, file, format, obj, &path, &directory_index); + line->filename = path; + parse_ver5_debug_line_header(include_directories, (int)directory_index, format, obj, &path, NULL); + line->dirname = path; + } + else { + for (i = 1; i <= file; i++) { + filename = p; + if (!*p) { + /* Need to output binary file name? */ + kprintf("Unexpected file number %d in %s at %tx\n", + file, binary_filename, filenames - obj->mapped); + return; + } + while (*p) p++; + p++; + dir = uleb128(&p); + /* last modified. */ + uleb128(&p); + /* size of the file. */ + uleb128(&p); + + if (i == file) { + line->filename = filename; + line->dirname = get_nth_dirname(dir, include_directories); + } + } } } static void fill_line(int num_traces, void **traces, uintptr_t addr, int file, int line, - char *include_directories, char *filenames, + uint8_t format, uint16_t version, const char *include_directories, const char *filenames, obj_info_t *obj, line_info_t *lines, int offset) { int i; - addr += obj->base_addr; + addr += obj->base_addr - obj->vmaddr; for (i = offset; i < num_traces; i++) { uintptr_t a = (uintptr_t)traces[i]; /* We assume one line code doesn't result >100 bytes of native code. We may want more reliable way eventually... */ if (addr < a && a < addr + 100) { - fill_filename(file, include_directories, filenames, &lines[i]); + fill_filename(file, format, version, include_directories, filenames, &lines[i], obj); lines[i].line = line; } } } +struct LineNumberProgramHeader { + uint64_t unit_length; + uint16_t version; + uint8_t format; /* 4 or 8 */ + uint64_t header_length; + uint8_t minimum_instruction_length; + uint8_t maximum_operations_per_instruction; + uint8_t default_is_stmt; + int8_t line_base; + uint8_t line_range; + uint8_t opcode_base; + /* uint8_t standard_opcode_lengths[opcode_base-1]; */ + const char *include_directories; + const char *filenames; + const char *cu_start; + const char *cu_end; +}; + static int -parse_debug_line_cu(int num_traces, void **traces, char **debug_line, - obj_info_t *obj, line_info_t *lines, int offset) +parse_debug_line_header(obj_info_t *obj, const char **pp, struct LineNumberProgramHeader *header) { - char *p, *cu_end, *cu_start, *include_directories, *filenames; - unsigned long unit_length; - int default_is_stmt, line_base; - unsigned int header_length, minimum_instruction_length, line_range, - opcode_base; - /* unsigned char *standard_opcode_lengths; */ - - /* The registers. */ - unsigned long addr = 0; - unsigned int file = 1; - unsigned int line = 1; - /* unsigned int column = 0; */ - int is_stmt; - /* int basic_block = 0; */ - /* int end_sequence = 0; */ - /* int prologue_end = 0; */ - /* int epilogue_begin = 0; */ - /* unsigned int isa = 0; */ - - p = *debug_line; - - unit_length = *(unsigned int *)p; - p += sizeof(unsigned int); - if (unit_length == 0xffffffff) { - unit_length = *(unsigned long *)p; - p += sizeof(unsigned long); + const char *p = *pp; + header->unit_length = *(uint32_t *)p; + p += sizeof(uint32_t); + + header->format = 4; + if (header->unit_length == 0xffffffff) { + header->unit_length = *(uint64_t *)p; + p += sizeof(uint64_t); + header->format = 8; } - cu_end = p + unit_length; + header->cu_end = p + header->unit_length; - /*dwarf_version = *(unsigned short *)p;*/ - p += 2; + header->version = *(uint16_t *)p; + p += sizeof(uint16_t); + if (header->version > 5) return -1; - header_length = *(unsigned int *)p; - p += sizeof(unsigned int); + if (header->version >= 5) { + /* address_size = *(uint8_t *)p++; */ + /* segment_selector_size = *(uint8_t *)p++; */ + p += 2; + } - cu_start = p + header_length; + header->header_length = header->format == 4 ? *(uint32_t *)p : *(uint64_t *)p; + p += header->format; + header->cu_start = p + header->header_length; - minimum_instruction_length = *(unsigned char *)p; - p++; + header->minimum_instruction_length = *(uint8_t *)p++; - is_stmt = default_is_stmt = *(unsigned char *)p; - p++; + if (header->version >= 4) { + /* maximum_operations_per_instruction = *(uint8_t *)p; */ + if (*p != 1) return -1; /* For non-VLIW architectures, this field is 1 */ + p++; + } - line_base = *(signed char *)p; - p++; + header->default_is_stmt = *(uint8_t *)p++; + header->line_base = *(int8_t *)p++; + header->line_range = *(uint8_t *)p++; + header->opcode_base = *(uint8_t *)p++; + /* header->standard_opcode_lengths = (uint8_t *)p - 1; */ + p += header->opcode_base - 1; + + if (header->version >= 5) { + header->include_directories = p; + p = parse_ver5_debug_line_header(p, -1, header->format, obj, NULL, NULL); + header->filenames = p; + } + else { + header->include_directories = p; - line_range = *(unsigned char *)p; - p++; + /* temporary measure for compress-debug-sections */ + if (p >= header->cu_end) return -1; - opcode_base = *(unsigned char *)p; - p++; + /* skip include directories */ + while (*p) { + p = memchr(p, '\0', header->cu_end - p); + if (!p) return -1; + p++; + } + p++; - /* standard_opcode_lengths = (unsigned char *)p - 1; */ - p += opcode_base - 1; + header->filenames = p; + } - include_directories = p; + *pp = header->cu_start; - /* temporary measure for compress-debug-sections */ - if (p >= cu_end) return -1; + return 0; +} - /* skip include directories */ - while (*p) { - p = memchr(p, '\0', cu_end - p); - if (!p) return -1; - p++; - } - p++; +static int +parse_debug_line_cu(int num_traces, void **traces, const char **debug_line, + obj_info_t *obj, line_info_t *lines, int offset) +{ + const char *p = (const char *)*debug_line; + struct LineNumberProgramHeader header; - filenames = p; + /* The registers. */ + unsigned long addr = 0; + unsigned int file = 1; + unsigned int line = 1; + /* unsigned int column = 0; */ + int is_stmt; + /* int basic_block = 0; */ + /* int end_sequence = 0; */ + /* int prologue_end = 0; */ + /* int epilogue_begin = 0; */ + /* unsigned int isa = 0; */ - p = cu_start; + if (parse_debug_line_header(obj, &p, &header)) + return -1; + is_stmt = header.default_is_stmt; #define FILL_LINE() \ do { \ fill_line(num_traces, traces, addr, file, line, \ - include_directories, filenames, \ + header.format, \ + header.version, \ + header.include_directories, \ + header.filenames, \ obj, lines, offset); \ /*basic_block = prologue_end = epilogue_begin = 0;*/ \ } while (0) - while (p < cu_end) { + while (p < header.cu_end) { unsigned long a; unsigned char op = *p++; switch (op) { @@ -328,7 +446,7 @@ parse_debug_line_cu(int num_traces, void **traces, char **debug_line, FILL_LINE(); break; case DW_LNS_advance_pc: - a = uleb128(&p); + a = uleb128(&p) * header.minimum_instruction_length; addr += a; break; case DW_LNS_advance_line: { @@ -349,12 +467,13 @@ parse_debug_line_cu(int num_traces, void **traces, char **debug_line, /*basic_block = 1; */ break; case DW_LNS_const_add_pc: - a = ((255 - opcode_base) / line_range) * - minimum_instruction_length; + a = ((255UL - header.opcode_base) / header.line_range) * + header.minimum_instruction_length; addr += a; break; case DW_LNS_fixed_advance_pc: - a = *(unsigned char *)p++; + a = *(uint16_t *)p; + p += sizeof(uint16_t); addr += a; break; case DW_LNS_set_prologue_end: @@ -367,7 +486,7 @@ parse_debug_line_cu(int num_traces, void **traces, char **debug_line, /* isa = (unsigned int)*/(void)uleb128(&p); break; case 0: - a = *(unsigned char *)p++; + a = uleb128(&p); op = *p++; switch (op) { case DW_LNE_end_sequence: @@ -377,7 +496,7 @@ parse_debug_line_cu(int num_traces, void **traces, char **debug_line, file = 1; line = 1; /* column = 0; */ - is_stmt = default_is_stmt; + is_stmt = header.default_is_stmt; /* end_sequence = 0; */ /* isa = 0; */ break; @@ -399,27 +518,25 @@ parse_debug_line_cu(int num_traces, void **traces, char **debug_line, } break; default: { - unsigned long addr_incr; - unsigned long line_incr; - a = op - opcode_base; - addr_incr = (a / line_range) * minimum_instruction_length; - line_incr = line_base + (a % line_range); - addr += (unsigned int)addr_incr; - line += (unsigned int)line_incr; + uint8_t adjusted_opcode = op - header.opcode_base; + uint8_t operation_advance = adjusted_opcode / header.line_range; + /* NOTE: this code doesn't support VLIW */ + addr += operation_advance * header.minimum_instruction_length; + line += header.line_base + (adjusted_opcode % header.line_range); FILL_LINE(); } } } - *debug_line = p; + *debug_line = (char *)p; return 0; } static int parse_debug_line(int num_traces, void **traces, - char *debug_line, unsigned long size, + const char *debug_line, unsigned long size, obj_info_t *obj, line_info_t *lines, int offset) { - char *debug_line_end = debug_line + size; + const char *debug_line_end = debug_line + size; while (debug_line < debug_line_end) { if (parse_debug_line_cu(num_traces, traces, &debug_line, obj, lines, offset)) return -1; @@ -444,13 +561,26 @@ append_obj(obj_info_t **objp) *objp = newobj; } +#ifdef USE_ELF +/* Ideally we should check 4 paths to follow gnu_debuglink: + * + * - /usr/lib/debug/.build-id/ab/cdef1234.debug + * - /usr/bin/ruby.debug + * - /usr/bin/.debug/ruby.debug + * - /usr/lib/debug/usr/bin/ruby.debug. + * + * but we handle only two cases for now as the two formats are + * used by some linux distributions. + * + * See GDB's info for detail. + * https://sourceware.org/gdb/onlinedocs/gdb/Separate-Debug-Files.html + */ + +// check the path pattern of "/usr/lib/debug/usr/bin/ruby.debug" static void follow_debuglink(const char *debuglink, int num_traces, void **traces, obj_info_t **objp, line_info_t *lines, int offset) { - /* Ideally we should check 4 paths to follow gnu_debuglink, - but we handle only one case for now as this format is used - by some linux distributions. See GDB's info for detail. */ static const char global_debug_dir[] = "/usr/lib/debug"; const size_t global_debug_dir_len = sizeof(global_debug_dir) - 1; char *p; @@ -478,6 +608,1398 @@ follow_debuglink(const char *debuglink, int num_traces, void **traces, fill_lines(num_traces, traces, 0, objp, lines, offset); } +// check the path pattern of "/usr/lib/debug/.build-id/ab/cdef1234.debug" +static void +follow_debuglink_build_id(const char *build_id, size_t build_id_size, int num_traces, void **traces, + obj_info_t **objp, line_info_t *lines, int offset) +{ + static const char global_debug_dir[] = "/usr/lib/debug/.build-id/"; + const size_t global_debug_dir_len = sizeof(global_debug_dir) - 1; + char *p; + obj_info_t *o1 = *objp, *o2; + size_t i; + + if (PATH_MAX < global_debug_dir_len + 1 + build_id_size * 2 + 6) return; + + memcpy(binary_filename, global_debug_dir, global_debug_dir_len); + p = binary_filename + global_debug_dir_len; + for (i = 0; i < build_id_size; i++) { + static const char tbl[] = "0123456789abcdef"; + unsigned char n = build_id[i]; + *p++ = tbl[n / 16]; + *p++ = tbl[n % 16]; + if (i == 0) *p++ = '/'; + } + strcpy(p, ".debug"); + + append_obj(objp); + o2 = *objp; + o2->base_addr = o1->base_addr; + o2->path = o1->path; + fill_lines(num_traces, traces, 0, objp, lines, offset); +} +#endif + +enum +{ + DW_TAG_compile_unit = 0x11, + DW_TAG_inlined_subroutine = 0x1d, + DW_TAG_subprogram = 0x2e, +}; + +/* Attributes encodings */ +enum +{ + DW_AT_sibling = 0x01, + DW_AT_location = 0x02, + DW_AT_name = 0x03, + /* Reserved 0x04 */ + /* Reserved 0x05 */ + /* Reserved 0x06 */ + /* Reserved 0x07 */ + /* Reserved 0x08 */ + DW_AT_ordering = 0x09, + /* Reserved 0x0a */ + DW_AT_byte_size = 0x0b, + /* Reserved 0x0c */ + DW_AT_bit_size = 0x0d, + /* Reserved 0x0e */ + /* Reserved 0x0f */ + DW_AT_stmt_list = 0x10, + DW_AT_low_pc = 0x11, + DW_AT_high_pc = 0x12, + DW_AT_language = 0x13, + /* Reserved 0x14 */ + DW_AT_discr = 0x15, + DW_AT_discr_value = 0x16, + DW_AT_visibility = 0x17, + DW_AT_import = 0x18, + DW_AT_string_length = 0x19, + DW_AT_common_reference = 0x1a, + DW_AT_comp_dir = 0x1b, + DW_AT_const_value = 0x1c, + DW_AT_containing_type = 0x1d, + DW_AT_default_value = 0x1e, + /* Reserved 0x1f */ + DW_AT_inline = 0x20, + DW_AT_is_optional = 0x21, + DW_AT_lower_bound = 0x22, + /* Reserved 0x23 */ + /* Reserved 0x24 */ + DW_AT_producer = 0x25, + /* Reserved 0x26 */ + DW_AT_prototyped = 0x27, + /* Reserved 0x28 */ + /* Reserved 0x29 */ + DW_AT_return_addr = 0x2a, + /* Reserved 0x2b */ + DW_AT_start_scope = 0x2c, + /* Reserved 0x2d */ + DW_AT_bit_stride = 0x2e, + DW_AT_upper_bound = 0x2f, + /* Reserved 0x30 */ + DW_AT_abstract_origin = 0x31, + DW_AT_accessibility = 0x32, + DW_AT_address_class = 0x33, + DW_AT_artificial = 0x34, + DW_AT_base_types = 0x35, + DW_AT_calling_convention = 0x36, + DW_AT_count = 0x37, + DW_AT_data_member_location = 0x38, + DW_AT_decl_column = 0x39, + DW_AT_decl_file = 0x3a, + DW_AT_decl_line = 0x3b, + DW_AT_declaration = 0x3c, + DW_AT_discr_list = 0x3d, + DW_AT_encoding = 0x3e, + DW_AT_external = 0x3f, + DW_AT_frame_base = 0x40, + DW_AT_friend = 0x41, + DW_AT_identifier_case = 0x42, + /* Reserved 0x43 */ + DW_AT_namelist_item = 0x44, + DW_AT_priority = 0x45, + DW_AT_segment = 0x46, + DW_AT_specification = 0x47, + DW_AT_static_link = 0x48, + DW_AT_type = 0x49, + DW_AT_use_location = 0x4a, + DW_AT_variable_parameter = 0x4b, + DW_AT_virtuality = 0x4c, + DW_AT_vtable_elem_location = 0x4d, + DW_AT_allocated = 0x4e, + DW_AT_associated = 0x4f, + DW_AT_data_location = 0x50, + DW_AT_byte_stride = 0x51, + DW_AT_entry_pc = 0x52, + DW_AT_use_UTF8 = 0x53, + DW_AT_extension = 0x54, + DW_AT_ranges = 0x55, + DW_AT_trampoline = 0x56, + DW_AT_call_column = 0x57, + DW_AT_call_file = 0x58, + DW_AT_call_line = 0x59, + DW_AT_description = 0x5a, + DW_AT_binary_scale = 0x5b, + DW_AT_decimal_scale = 0x5c, + DW_AT_small = 0x5d, + DW_AT_decimal_sign = 0x5e, + DW_AT_digit_count = 0x5f, + DW_AT_picture_string = 0x60, + DW_AT_mutable = 0x61, + DW_AT_threads_scaled = 0x62, + DW_AT_explicit = 0x63, + DW_AT_object_pointer = 0x64, + DW_AT_endianity = 0x65, + DW_AT_elemental = 0x66, + DW_AT_pure = 0x67, + DW_AT_recursive = 0x68, + DW_AT_signature = 0x69, + DW_AT_main_subprogram = 0x6a, + DW_AT_data_bit_offset = 0x6b, + DW_AT_const_expr = 0x6c, + DW_AT_enum_class = 0x6d, + DW_AT_linkage_name = 0x6e, + DW_AT_string_length_bit_size = 0x6f, + DW_AT_string_length_byte_size = 0x70, + DW_AT_rank = 0x71, + DW_AT_str_offsets_base = 0x72, + DW_AT_addr_base = 0x73, + DW_AT_rnglists_base = 0x74, + /* Reserved 0x75 */ + DW_AT_dwo_name = 0x76, + DW_AT_reference = 0x77, + DW_AT_rvalue_reference = 0x78, + DW_AT_macros = 0x79, + DW_AT_call_all_calls = 0x7a, + DW_AT_call_all_source_calls = 0x7b, + DW_AT_call_all_tail_calls = 0x7c, + DW_AT_call_return_pc = 0x7d, + DW_AT_call_value = 0x7e, + DW_AT_call_origin = 0x7f, + DW_AT_call_parameter = 0x80, + DW_AT_call_pc = 0x81, + DW_AT_call_tail_call = 0x82, + DW_AT_call_target = 0x83, + DW_AT_call_target_clobbered = 0x84, + DW_AT_call_data_location = 0x85, + DW_AT_call_data_value = 0x86, + DW_AT_noreturn = 0x87, + DW_AT_alignment = 0x88, + DW_AT_export_symbols = 0x89, + DW_AT_deleted = 0x8a, + DW_AT_defaulted = 0x8b, + DW_AT_loclists_base = 0x8c, + DW_AT_lo_user = 0x2000, + DW_AT_hi_user = 0x3fff +}; + +/* Attribute form encodings */ +enum +{ + DW_FORM_addr = 0x01, + /* Reserved 0x02 */ + DW_FORM_block2 = 0x03, + DW_FORM_block4 = 0x04, + DW_FORM_data2 = 0x05, + DW_FORM_data4 = 0x06, + DW_FORM_data8 = 0x07, + DW_FORM_string = 0x08, + DW_FORM_block = 0x09, + DW_FORM_block1 = 0x0a, + DW_FORM_data1 = 0x0b, + DW_FORM_flag = 0x0c, + DW_FORM_sdata = 0x0d, + DW_FORM_strp = 0x0e, + DW_FORM_udata = 0x0f, + DW_FORM_ref_addr = 0x10, + DW_FORM_ref1 = 0x11, + DW_FORM_ref2 = 0x12, + DW_FORM_ref4 = 0x13, + DW_FORM_ref8 = 0x14, + DW_FORM_ref_udata = 0x15, + DW_FORM_indirect = 0x16, + DW_FORM_sec_offset = 0x17, + DW_FORM_exprloc = 0x18, + DW_FORM_flag_present = 0x19, + DW_FORM_strx = 0x1a, + DW_FORM_addrx = 0x1b, + DW_FORM_ref_sup4 = 0x1c, + DW_FORM_strp_sup = 0x1d, + DW_FORM_data16 = 0x1e, + DW_FORM_line_strp = 0x1f, + DW_FORM_ref_sig8 = 0x20, + DW_FORM_implicit_const = 0x21, + DW_FORM_loclistx = 0x22, + DW_FORM_rnglistx = 0x23, + DW_FORM_ref_sup8 = 0x24, + DW_FORM_strx1 = 0x25, + DW_FORM_strx2 = 0x26, + DW_FORM_strx3 = 0x27, + DW_FORM_strx4 = 0x28, + DW_FORM_addrx1 = 0x29, + DW_FORM_addrx2 = 0x2a, + DW_FORM_addrx3 = 0x2b, + DW_FORM_addrx4 = 0x2c +}; + +/* Range list entry encodings */ +enum { + DW_RLE_end_of_list = 0x00, + DW_RLE_base_addressx = 0x01, + DW_RLE_startx_endx = 0x02, + DW_RLE_startx_length = 0x03, + DW_RLE_offset_pair = 0x04, + DW_RLE_base_address = 0x05, + DW_RLE_start_end = 0x06, + DW_RLE_start_length = 0x07 +}; + +enum { + VAL_none = 0, + VAL_cstr = 1, + VAL_data = 2, + VAL_uint = 3, + VAL_int = 4, + VAL_addr = 5 +}; + +# define ABBREV_TABLE_SIZE 256 +typedef struct { + obj_info_t *obj; + const char *file; + uint8_t current_version; + const char *current_cu; + uint64_t current_low_pc; + uint64_t current_str_offsets_base; + uint64_t current_addr_base; + uint64_t current_rnglists_base; + const char *debug_line_cu_end; + uint8_t debug_line_format; + uint16_t debug_line_version; + const char *debug_line_files; + const char *debug_line_directories; + const char *p; + const char *cu_end; + const char *pend; + const char *q0; + const char *q; + int format; // 4 or 8 + uint8_t address_size; + int level; + const char *abbrev_table[ABBREV_TABLE_SIZE]; +} DebugInfoReader; + +typedef struct { + ptrdiff_t pos; + int tag; + int has_children; +} DIE; + +typedef struct { + union { + const char *ptr; + uint64_t uint64; + int64_t int64; + uint64_t addr_idx; + } as; + uint64_t off; + uint64_t at; + uint64_t form; + size_t size; + int type; +} DebugInfoValue; + +#if defined(WORDS_BIGENDIAN) +#define MERGE_2INTS(a,b,sz) (((uint64_t)(a)<<sz)|(b)) +#else +#define MERGE_2INTS(a,b,sz) (((uint64_t)(b)<<sz)|(a)) +#endif + +static uint16_t +get_uint16(const uint8_t *p) +{ + return (uint16_t)MERGE_2INTS(p[0],p[1],8); +} + +static uint32_t +get_uint32(const uint8_t *p) +{ + return (uint32_t)MERGE_2INTS(get_uint16(p),get_uint16(p+2),16); +} + +static uint64_t +get_uint64(const uint8_t *p) +{ + return MERGE_2INTS(get_uint32(p),get_uint32(p+4),32); +} + +static uint8_t +read_uint8(const char **ptr) +{ + const char *p = *ptr; + *ptr = (p + 1); + return (uint8_t)*p; +} + +static uint16_t +read_uint16(const char **ptr) +{ + const char *p = *ptr; + *ptr = (p + 2); + return get_uint16((const uint8_t *)p); +} + +static uint32_t +read_uint24(const char **ptr) +{ + const char *p = *ptr; + *ptr = (p + 3); + return ((uint8_t)*p << 16) | get_uint16((const uint8_t *)p+1); +} + +static uint32_t +read_uint32(const char **ptr) +{ + const char *p = *ptr; + *ptr = (p + 4); + return get_uint32((const uint8_t *)p); +} + +static uint64_t +read_uint64(const char **ptr) +{ + const unsigned char *p = (const unsigned char *)*ptr; + *ptr = (char *)(p + 8); + return get_uint64(p); +} + +static uintptr_t +read_uintptr(const char **ptr) +{ + const unsigned char *p = (const unsigned char *)*ptr; + *ptr = (char *)(p + SIZEOF_VOIDP); +#if SIZEOF_VOIDP == 8 + return get_uint64(p); +#else + return get_uint32(p); +#endif +} + +static uint64_t +read_uint(DebugInfoReader *reader) +{ + if (reader->format == 4) { + return read_uint32(&reader->p); + } else { /* 64 bit */ + return read_uint64(&reader->p); + } +} + +static uint64_t +read_uleb128(DebugInfoReader *reader) +{ + return uleb128(&reader->p); +} + +static int64_t +read_sleb128(DebugInfoReader *reader) +{ + return sleb128(&reader->p); +} + +static void +debug_info_reader_init(DebugInfoReader *reader, obj_info_t *obj) +{ + reader->file = obj->mapped; + reader->obj = obj; + reader->p = obj->debug_info.ptr; + reader->pend = obj->debug_info.ptr + obj->debug_info.size; + reader->debug_line_cu_end = obj->debug_line.ptr; + reader->current_low_pc = 0; + reader->current_str_offsets_base = 0; + reader->current_addr_base = 0; + reader->current_rnglists_base = 0; +} + +static void +di_skip_die_attributes(const char **p) +{ + for (;;) { + uint64_t at = uleb128(p); + uint64_t form = uleb128(p); + if (!at && !form) break; + switch (form) { + default: + break; + case DW_FORM_implicit_const: + sleb128(p); + break; + } + } +} + +static void +di_read_debug_abbrev_cu(DebugInfoReader *reader) +{ + uint64_t prev = 0; + const char *p = reader->q0; + for (;;) { + uint64_t abbrev_number = uleb128(&p); + if (abbrev_number <= prev) break; + if (abbrev_number < ABBREV_TABLE_SIZE) { + reader->abbrev_table[abbrev_number] = p; + } + prev = abbrev_number; + uleb128(&p); /* tag */ + p++; /* has_children */ + di_skip_die_attributes(&p); + } +} + +static int +di_read_debug_line_cu(DebugInfoReader *reader) +{ + const char *p; + struct LineNumberProgramHeader header; + + p = (const char *)reader->debug_line_cu_end; + if (parse_debug_line_header(reader->obj, &p, &header)) + return -1; + + reader->debug_line_cu_end = (char *)header.cu_end; + reader->debug_line_format = header.format; + reader->debug_line_version = header.version; + reader->debug_line_directories = (char *)header.include_directories; + reader->debug_line_files = (char *)header.filenames; + + return 0; +} + +static void +set_addr_idx_value(DebugInfoValue *v, uint64_t n) +{ + v->as.addr_idx = n; + v->type = VAL_addr; +} + +static void +set_uint_value(DebugInfoValue *v, uint64_t n) +{ + v->as.uint64 = n; + v->type = VAL_uint; +} + +static void +set_int_value(DebugInfoValue *v, int64_t n) +{ + v->as.int64 = n; + v->type = VAL_int; +} + +static void +set_cstr_value(DebugInfoValue *v, const char *s) +{ + v->as.ptr = s; + v->off = 0; + v->type = VAL_cstr; +} + +static void +set_cstrp_value(DebugInfoValue *v, const char *s, uint64_t off) +{ + v->as.ptr = s; + v->off = off; + v->type = VAL_cstr; +} + +static void +set_data_value(DebugInfoValue *v, const char *s) +{ + v->as.ptr = s; + v->type = VAL_data; +} + +static const char * +get_cstr_value(DebugInfoValue *v) +{ + if (v->as.ptr) { + return v->as.ptr + v->off; + } else { + return NULL; + } +} + +static const char * +resolve_strx(DebugInfoReader *reader, uint64_t idx) +{ + const char *p = reader->obj->debug_str_offsets.ptr + reader->current_str_offsets_base; + uint64_t off; + if (reader->format == 4) { + off = ((uint32_t *)p)[idx]; + } + else { + off = ((uint64_t *)p)[idx]; + } + return reader->obj->debug_str.ptr + off; +} + +static void +debug_info_reader_read_addr_value(DebugInfoReader *reader, DebugInfoValue *v) +{ + if (reader->address_size == 4) { + set_uint_value(v, read_uint32(&reader->p)); + } else if (reader->address_size == 8) { + set_uint_value(v, read_uint64(&reader->p)); + } else { + fprintf(stderr,"unknown address_size:%d", reader->address_size); + abort(); + } +} + +static void +debug_info_reader_read_value(DebugInfoReader *reader, uint64_t form, DebugInfoValue *v) +{ + switch (form) { + case DW_FORM_addr: + debug_info_reader_read_addr_value(reader, v); + break; + case DW_FORM_block2: + v->size = read_uint16(&reader->p); + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_block4: + v->size = read_uint32(&reader->p); + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_data2: + set_uint_value(v, read_uint16(&reader->p)); + break; + case DW_FORM_data4: + set_uint_value(v, read_uint32(&reader->p)); + break; + case DW_FORM_data8: + set_uint_value(v, read_uint64(&reader->p)); + break; + case DW_FORM_string: + v->size = strlen(reader->p); + set_cstr_value(v, reader->p); + reader->p += v->size + 1; + break; + case DW_FORM_block: + v->size = uleb128(&reader->p); + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_block1: + v->size = read_uint8(&reader->p); + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_data1: + set_uint_value(v, read_uint8(&reader->p)); + break; + case DW_FORM_flag: + set_uint_value(v, read_uint8(&reader->p)); + break; + case DW_FORM_sdata: + set_int_value(v, read_sleb128(reader)); + break; + case DW_FORM_strp: + set_cstrp_value(v, reader->obj->debug_str.ptr, read_uint(reader)); + break; + case DW_FORM_udata: + set_uint_value(v, read_uleb128(reader)); + break; + case DW_FORM_ref_addr: + if (reader->current_version <= 2) { + // DWARF Version 2 specifies that references have + // the same size as an address on the target system + debug_info_reader_read_addr_value(reader, v); + } else { + if (reader->format == 4) { + set_uint_value(v, read_uint32(&reader->p)); + } else if (reader->format == 8) { + set_uint_value(v, read_uint64(&reader->p)); + } else { + fprintf(stderr,"unknown format:%d", reader->format); + abort(); + } + } + break; + case DW_FORM_ref1: + set_uint_value(v, read_uint8(&reader->p)); + break; + case DW_FORM_ref2: + set_uint_value(v, read_uint16(&reader->p)); + break; + case DW_FORM_ref4: + set_uint_value(v, read_uint32(&reader->p)); + break; + case DW_FORM_ref8: + set_uint_value(v, read_uint64(&reader->p)); + break; + case DW_FORM_ref_udata: + set_uint_value(v, uleb128(&reader->p)); + break; + case DW_FORM_indirect: + /* TODO: read the referred value */ + set_uint_value(v, uleb128(&reader->p)); + break; + case DW_FORM_sec_offset: + set_uint_value(v, read_uint(reader)); /* offset */ + /* addrptr: debug_addr */ + /* lineptr: debug_line */ + /* loclist: debug_loclists */ + /* loclistptr: debug_loclists */ + /* macptr: debug_macro */ + /* rnglist: debug_rnglists */ + /* rnglistptr: debug_rnglists */ + /* stroffsetsptr: debug_str_offsets */ + break; + case DW_FORM_exprloc: + v->size = (size_t)read_uleb128(reader); + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_flag_present: + set_uint_value(v, 1); + break; + case DW_FORM_strx: + set_cstr_value(v, resolve_strx(reader, uleb128(&reader->p))); + break; + case DW_FORM_addrx: + set_addr_idx_value(v, uleb128(&reader->p)); + break; + case DW_FORM_ref_sup4: + set_uint_value(v, read_uint32(&reader->p)); + break; + case DW_FORM_strp_sup: + set_uint_value(v, read_uint(reader)); + /* *p = reader->sup_file + reader->sup_str->sh_offset + ret; */ + break; + case DW_FORM_data16: + v->size = 16; + set_data_value(v, reader->p); + reader->p += v->size; + break; + case DW_FORM_line_strp: + set_cstrp_value(v, reader->obj->debug_line_str.ptr, read_uint(reader)); + break; + case DW_FORM_ref_sig8: + set_uint_value(v, read_uint64(&reader->p)); + break; + case DW_FORM_implicit_const: + set_int_value(v, sleb128(&reader->q)); + break; + case DW_FORM_loclistx: + set_uint_value(v, read_uleb128(reader)); + break; + case DW_FORM_rnglistx: + set_uint_value(v, read_uleb128(reader)); + break; + case DW_FORM_ref_sup8: + set_uint_value(v, read_uint64(&reader->p)); + break; + case DW_FORM_strx1: + set_cstr_value(v, resolve_strx(reader, read_uint8(&reader->p))); + break; + case DW_FORM_strx2: + set_cstr_value(v, resolve_strx(reader, read_uint16(&reader->p))); + break; + case DW_FORM_strx3: + set_cstr_value(v, resolve_strx(reader, read_uint24(&reader->p))); + break; + case DW_FORM_strx4: + set_cstr_value(v, resolve_strx(reader, read_uint32(&reader->p))); + break; + case DW_FORM_addrx1: + set_addr_idx_value(v, read_uint8(&reader->p)); + break; + case DW_FORM_addrx2: + set_addr_idx_value(v, read_uint16(&reader->p)); + break; + case DW_FORM_addrx3: + set_addr_idx_value(v, read_uint24(&reader->p)); + break; + case DW_FORM_addrx4: + set_addr_idx_value(v, read_uint32(&reader->p)); + break; + case 0: + goto fail; + break; + } + return; + + fail: + fprintf(stderr, "%d: unsupported form: %#"PRIx64"\n", __LINE__, form); + exit(1); +} + +/* find abbrev in current compilation unit */ +static const char * +di_find_abbrev(DebugInfoReader *reader, uint64_t abbrev_number) +{ + const char *p; + if (abbrev_number < ABBREV_TABLE_SIZE) { + return reader->abbrev_table[abbrev_number]; + } + p = reader->abbrev_table[ABBREV_TABLE_SIZE-1]; + /* skip 255th record */ + uleb128(&p); /* tag */ + p++; /* has_children */ + di_skip_die_attributes(&p); + for (uint64_t n = uleb128(&p); abbrev_number != n; n = uleb128(&p)) { + if (n == 0) { + fprintf(stderr,"%d: Abbrev Number %"PRId64" not found\n",__LINE__, abbrev_number); + exit(1); + } + uleb128(&p); /* tag */ + p++; /* has_children */ + di_skip_die_attributes(&p); + } + return p; +} + +#if 0 +static void +hexdump0(const unsigned char *p, size_t n) +{ + size_t i; + fprintf(stderr, " 0 1 2 3 4 5 6 7 8 9 A B C D E F\n"); + for (i=0; i < n; i++){ + switch (i & 15) { + case 0: + fprintf(stderr, "%02" PRIdSIZE ": %02X ", i/16, p[i]); + break; + case 15: + fprintf(stderr, "%02X\n", p[i]); + break; + default: + fprintf(stderr, "%02X ", p[i]); + break; + } + } + if ((i & 15) != 15) { + fprintf(stderr, "\n"); + } +} +#define hexdump(p,n) hexdump0((const unsigned char *)p, n) + +static void +div_inspect(DebugInfoValue *v) +{ + switch (v->type) { + case VAL_uint: + fprintf(stderr,"%d: type:%d size:%" PRIxSIZE " v:%"PRIx64"\n",__LINE__,v->type,v->size,v->as.uint64); + break; + case VAL_int: + fprintf(stderr,"%d: type:%d size:%" PRIxSIZE " v:%"PRId64"\n",__LINE__,v->type,v->size,(int64_t)v->as.uint64); + break; + case VAL_cstr: + fprintf(stderr,"%d: type:%d size:%" PRIxSIZE " v:'%s'\n",__LINE__,v->type,v->size,v->as.ptr); + break; + case VAL_data: + fprintf(stderr,"%d: type:%d size:%" PRIxSIZE " v:\n",__LINE__,v->type,v->size); + hexdump(v->as.ptr, 16); + break; + } +} +#endif + +static DIE * +di_read_die(DebugInfoReader *reader, DIE *die) +{ + uint64_t abbrev_number = uleb128(&reader->p); + if (abbrev_number == 0) { + reader->level--; + return NULL; + } + + reader->q = di_find_abbrev(reader, abbrev_number); + + die->pos = reader->p - reader->obj->debug_info.ptr - 1; + die->tag = (int)uleb128(&reader->q); /* tag */ + die->has_children = *reader->q++; /* has_children */ + if (die->has_children) { + reader->level++; + } + return die; +} + +static DebugInfoValue * +di_read_record(DebugInfoReader *reader, DebugInfoValue *vp) +{ + uint64_t at = uleb128(&reader->q); + uint64_t form = uleb128(&reader->q); + if (!at || !form) return NULL; + vp->at = at; + vp->form = form; + debug_info_reader_read_value(reader, form, vp); + return vp; +} + +static void +di_skip_records(DebugInfoReader *reader) +{ + for (;;) { + DebugInfoValue v = {{}}; + uint64_t at = uleb128(&reader->q); + uint64_t form = uleb128(&reader->q); + if (!at || !form) return; + debug_info_reader_read_value(reader, form, &v); + } +} + +typedef struct addr_header { + const char *ptr; + uint64_t unit_length; + uint8_t format; + uint8_t address_size; + /* uint8_t segment_selector_size; */ +} addr_header_t; + +static void +addr_header_init(obj_info_t *obj, addr_header_t *header) { + const char *p = obj->debug_addr.ptr; + + header->ptr = p; + + if (!p) return; + + header->unit_length = *(uint32_t *)p; + p += sizeof(uint32_t); + + header->format = 4; + if (header->unit_length == 0xffffffff) { + header->unit_length = *(uint64_t *)p; + p += sizeof(uint64_t); + header->format = 8; + } + + p += 2; /* version */ + header->address_size = *p++; + p++; /* segment_selector_size */ +} + +static uint64_t +read_addr(addr_header_t *header, uint64_t addr_base, uint64_t idx) { + if (header->address_size == 4) { + return ((uint32_t*)(header->ptr + addr_base))[idx]; + } + else { + return ((uint64_t*)(header->ptr + addr_base))[idx]; + } +} + +typedef struct rnglists_header { + uint64_t unit_length; + uint8_t format; + uint8_t address_size; + uint32_t offset_entry_count; +} rnglists_header_t; + +static void +rnglists_header_init(obj_info_t *obj, rnglists_header_t *header) { + const char *p = obj->debug_rnglists.ptr; + + if (!p) return; + + header->unit_length = *(uint32_t *)p; + p += sizeof(uint32_t); + + header->format = 4; + if (header->unit_length == 0xffffffff) { + header->unit_length = *(uint64_t *)p; + p += sizeof(uint64_t); + header->format = 8; + } + + p += 2; /* version */ + header->address_size = *p++; + p++; /* segment_selector_size */ + header->offset_entry_count = *(uint32_t *)p; +} + +typedef struct { + uint64_t low_pc; + uint64_t high_pc; + uint64_t ranges; + bool low_pc_set; + bool high_pc_set; + bool ranges_set; +} ranges_t; + +static void +ranges_set(ranges_t *ptr, DebugInfoValue *v, addr_header_t *addr_header, uint64_t addr_base) +{ + uint64_t n = 0; + if (v->type == VAL_uint) { + n = v->as.uint64; + } + else if (v->type == VAL_addr) { + n = read_addr(addr_header, addr_base, v->as.addr_idx); + } + switch (v->at) { + case DW_AT_low_pc: + ptr->low_pc = n; + ptr->low_pc_set = true; + break; + case DW_AT_high_pc: + if (v->form == DW_FORM_addr) { + ptr->high_pc = n; + } + else { + ptr->high_pc = ptr->low_pc + n; + } + ptr->high_pc_set = true; + break; + case DW_AT_ranges: + ptr->ranges = n; + ptr->ranges_set = true; + break; + } +} + +static uint64_t +read_dw_form_addr(DebugInfoReader *reader, const char **ptr) +{ + const char *p = *ptr; + *ptr = p + reader->address_size; + if (reader->address_size == 4) { + return read_uint32(&p); + } else if (reader->address_size == 8) { + return read_uint64(&p); + } else { + fprintf(stderr,"unknown address_size:%d", reader->address_size); + abort(); + } +} + +static uintptr_t +ranges_include(DebugInfoReader *reader, ranges_t *ptr, uint64_t addr, rnglists_header_t *rnglists_header) +{ + if (ptr->high_pc_set) { + if (ptr->ranges_set || !ptr->low_pc_set) { + exit(1); + } + if (ptr->low_pc <= addr && addr <= ptr->high_pc) { + return (uintptr_t)ptr->low_pc; + } + } + else if (ptr->ranges_set) { + /* TODO: support base address selection entry */ + const char *p; + uint64_t base = ptr->low_pc_set ? ptr->low_pc : reader->current_low_pc; + bool base_valid = true; + if (reader->current_version >= 5) { + if (rnglists_header->offset_entry_count == 0) { + // DW_FORM_sec_offset + p = reader->obj->debug_rnglists.ptr + ptr->ranges + reader->current_rnglists_base; + } + else { + // DW_FORM_rnglistx + const char *offset_array = reader->obj->debug_rnglists.ptr + reader->current_rnglists_base; + if (rnglists_header->format == 4) { + p = offset_array + ((uint32_t *)offset_array)[ptr->ranges]; + } + else { + p = offset_array + ((uint64_t *)offset_array)[ptr->ranges]; + } + } + for (;;) { + uint8_t rle = read_uint8(&p); + uintptr_t from = 0, to = 0; + if (rle == DW_RLE_end_of_list) break; + switch (rle) { + case DW_RLE_base_addressx: + uleb128(&p); + base_valid = false; /* not supported yet */ + break; + case DW_RLE_startx_endx: + uleb128(&p); + uleb128(&p); + break; + case DW_RLE_startx_length: + uleb128(&p); + uleb128(&p); + break; + case DW_RLE_offset_pair: + if (!base_valid) break; + from = (uintptr_t)base + uleb128(&p); + to = (uintptr_t)base + uleb128(&p); + break; + case DW_RLE_base_address: + base = read_dw_form_addr(reader, &p); + base_valid = true; + break; + case DW_RLE_start_end: + from = (uintptr_t)read_dw_form_addr(reader, &p); + to = (uintptr_t)read_dw_form_addr(reader, &p); + break; + case DW_RLE_start_length: + from = (uintptr_t)read_dw_form_addr(reader, &p); + to = from + uleb128(&p); + break; + } + if (from <= addr && addr < to) { + return from; + } + } + return false; + } + p = reader->obj->debug_ranges.ptr + ptr->ranges; + for (;;) { + uintptr_t from = read_uintptr(&p); + uintptr_t to = read_uintptr(&p); + if (!from && !to) break; + if (from == UINTPTR_MAX) { + /* base address selection entry */ + base = to; + } + else if (base + from <= addr && addr < base + to) { + return (uintptr_t)base + from; + } + } + } + else if (ptr->low_pc_set) { + if (ptr->low_pc == addr) { + return (uintptr_t)ptr->low_pc; + } + } + return false; +} + +#if 0 +static void +ranges_inspect(DebugInfoReader *reader, ranges_t *ptr) +{ + if (ptr->high_pc_set) { + if (ptr->ranges_set || !ptr->low_pc_set) { + fprintf(stderr,"low_pc_set:%d high_pc_set:%d ranges_set:%d\n",ptr->low_pc_set,ptr->high_pc_set,ptr->ranges_set); + exit(1); + } + fprintf(stderr,"low_pc:%"PRIx64" high_pc:%"PRIx64"\n",ptr->low_pc,ptr->high_pc); + } + else if (ptr->ranges_set) { + char *p = reader->obj->debug_ranges.ptr + ptr->ranges; + fprintf(stderr,"low_pc:%"PRIx64" ranges:%"PRIx64" %lx ",ptr->low_pc,ptr->ranges, p-reader->obj->mapped); + for (;;) { + uintptr_t from = read_uintptr(&p); + uintptr_t to = read_uintptr(&p); + if (!from && !to) break; + fprintf(stderr,"%"PRIx64"-%"PRIx64" ",ptr->low_pc+from,ptr->low_pc+to); + } + fprintf(stderr,"\n"); + } + else if (ptr->low_pc_set) { + fprintf(stderr,"low_pc:%"PRIx64"\n",ptr->low_pc); + } + else { + fprintf(stderr,"empty\n"); + } +} +#endif + +static int +di_read_cu(DebugInfoReader *reader) +{ + uint64_t unit_length; + uint16_t version; + uint64_t debug_abbrev_offset; + reader->format = 4; + reader->current_cu = reader->p; + unit_length = read_uint32(&reader->p); + if (unit_length == 0xffffffff) { + unit_length = read_uint64(&reader->p); + reader->format = 8; + } + reader->cu_end = reader->p + unit_length; + version = read_uint16(&reader->p); + reader->current_version = version; + if (version > 5) { + return -1; + } + else if (version == 5) { + /* unit_type = */ read_uint8(&reader->p); + reader->address_size = read_uint8(&reader->p); + debug_abbrev_offset = read_uint(reader); + } + else { + debug_abbrev_offset = read_uint(reader); + reader->address_size = read_uint8(&reader->p); + } + reader->q0 = reader->obj->debug_abbrev.ptr + debug_abbrev_offset; + + reader->level = 0; + di_read_debug_abbrev_cu(reader); + if (di_read_debug_line_cu(reader)) return -1; + +#if defined(__GNUC__) && !defined(__clang__) && !defined(__INTEL_COMPILER_BUILD_DATE) + /* Though DWARF specifies "the applicable base address defaults to the base + address of the compilation unit", but GCC seems to use zero as default */ +#else + do { + DIE die; + + if (!di_read_die(reader, &die)) continue; + + if (die.tag != DW_TAG_compile_unit) { + di_skip_records(reader); + break; + } + + reader->current_str_offsets_base = 0; + reader->current_addr_base = 0; + reader->current_rnglists_base = 0; + + DebugInfoValue low_pc = {{}}; + /* enumerate abbrev */ + for (;;) { + DebugInfoValue v = {{}}; + if (!di_read_record(reader, &v)) break; + switch (v.at) { + case DW_AT_low_pc: + // clang may output DW_AT_addr_base after DW_AT_low_pc. + // We need to resolve the DW_FORM_addr* after DW_AT_addr_base is parsed. + low_pc = v; + break; + case DW_AT_str_offsets_base: + reader->current_str_offsets_base = v.as.uint64; + break; + case DW_AT_addr_base: + reader->current_addr_base = v.as.uint64; + break; + case DW_AT_rnglists_base: + reader->current_rnglists_base = v.as.uint64; + break; + } + } + // Resolve the DW_FORM_addr of DW_AT_low_pc + switch (low_pc.type) { + case VAL_uint: + reader->current_low_pc = low_pc.as.uint64; + break; + case VAL_addr: + { + addr_header_t header; + addr_header_init(reader->obj, &header); + reader->current_low_pc = read_addr(&header, reader->current_addr_base, low_pc.as.addr_idx); + } + break; + } + } while (0); +#endif + return 0; +} + +static void +read_abstract_origin(DebugInfoReader *reader, uint64_t form, uint64_t abstract_origin, line_info_t *line) +{ + const char *p = reader->p; + const char *q = reader->q; + int level = reader->level; + DIE die; + + switch (form) { + case DW_FORM_ref1: + case DW_FORM_ref2: + case DW_FORM_ref4: + case DW_FORM_ref8: + case DW_FORM_ref_udata: + reader->p = reader->current_cu + abstract_origin; + break; + case DW_FORM_ref_addr: + goto finish; /* not supported yet */ + case DW_FORM_ref_sig8: + goto finish; /* not supported yet */ + case DW_FORM_ref_sup4: + case DW_FORM_ref_sup8: + goto finish; /* not supported yet */ + default: + goto finish; + } + if (!di_read_die(reader, &die)) goto finish; + + /* enumerate abbrev */ + for (;;) { + DebugInfoValue v = {{}}; + if (!di_read_record(reader, &v)) break; + switch (v.at) { + case DW_AT_name: + line->sname = get_cstr_value(&v); + break; + } + } + + finish: + reader->p = p; + reader->q = q; + reader->level = level; +} + +static void +debug_info_read(DebugInfoReader *reader, int num_traces, void **traces, + line_info_t *lines, int offset) { + + addr_header_t addr_header = {}; + addr_header_init(reader->obj, &addr_header); + + rnglists_header_t rnglists_header = {}; + rnglists_header_init(reader->obj, &rnglists_header); + + while (reader->p < reader->cu_end) { + DIE die; + ranges_t ranges = {}; + line_info_t line = {}; + + if (!di_read_die(reader, &die)) continue; + /* fprintf(stderr,"%d:%tx: <%d>\n",__LINE__,die.pos,reader->level,die.tag); */ + + if (die.tag != DW_TAG_subprogram && die.tag != DW_TAG_inlined_subroutine) { + skip_die: + di_skip_records(reader); + continue; + } + + /* enumerate abbrev */ + for (;;) { + DebugInfoValue v = {{}}; + /* ptrdiff_t pos = reader->p - reader->p0; */ + if (!di_read_record(reader, &v)) break; + /* fprintf(stderr,"\n%d:%tx: AT:%lx FORM:%lx\n",__LINE__,pos,v.at,v.form); */ + /* div_inspect(&v); */ + switch (v.at) { + case DW_AT_name: + line.sname = get_cstr_value(&v); + break; + case DW_AT_call_file: + fill_filename((int)v.as.uint64, reader->debug_line_format, reader->debug_line_version, reader->debug_line_directories, reader->debug_line_files, &line, reader->obj); + break; + case DW_AT_call_line: + line.line = (int)v.as.uint64; + break; + case DW_AT_low_pc: + case DW_AT_high_pc: + case DW_AT_ranges: + ranges_set(&ranges, &v, &addr_header, reader->current_addr_base); + break; + case DW_AT_declaration: + goto skip_die; + case DW_AT_inline: + /* 1 or 3 */ + break; /* goto skip_die; */ + case DW_AT_abstract_origin: + read_abstract_origin(reader, v.form, v.as.uint64, &line); + break; /* goto skip_die; */ + } + } + /* ranges_inspect(reader, &ranges); */ + /* fprintf(stderr,"%d:%tx: %x ",__LINE__,diepos,die.tag); */ + for (int i=offset; i < num_traces; i++) { + uintptr_t addr = (uintptr_t)traces[i]; + uintptr_t offset = addr - reader->obj->base_addr + reader->obj->vmaddr; + uintptr_t saddr = ranges_include(reader, &ranges, offset, &rnglists_header); + if (saddr) { + /* fprintf(stdout, "%d:%tx: %d %lx->%lx %x %s: %s/%s %d %s %s %s\n",__LINE__,die.pos, i,addr,offset, die.tag,line.sname,line.dirname,line.filename,line.line,reader->obj->path,line.sname,lines[i].sname); */ + if (lines[i].sname) { + line_info_t *lp = malloc(sizeof(line_info_t)); + memcpy(lp, &lines[i], sizeof(line_info_t)); + lines[i].next = lp; + lp->dirname = line.dirname; + lp->filename = line.filename; + lp->line = line.line; + lp->saddr = 0; + } + lines[i].path = reader->obj->path; + lines[i].base_addr = line.base_addr; + lines[i].sname = line.sname; + lines[i].saddr = saddr + reader->obj->base_addr - reader->obj->vmaddr; + } + } + } +} + +// This function parses the following attributes of Line Number Program Header in DWARF 5: +// +// * directory_entry_format_count +// * directory_entry_format +// * directories_count +// * directories +// +// or +// +// * file_name_entry_format_count +// * file_name_entry_format +// * file_names_count +// * file_names +// +// It records DW_LNCT_path and DW_LNCT_directory_index at the index "idx". +static const char * +parse_ver5_debug_line_header(const char *p, int idx, uint8_t format, obj_info_t *obj, const char **out_path, uint64_t *out_directory_index) { + int i, j; + int entry_format_count = *(uint8_t *)p++; + const char *entry_format = p; + + /* skip the part of entry_format */ + for (i = 0; i < entry_format_count * 2; i++) uleb128(&p); + + int entry_count = (int)uleb128(&p); + + DebugInfoReader reader; + debug_info_reader_init(&reader, obj); + reader.format = format; + reader.p = p; + for (j = 0; j < entry_count; j++) { + const char *format = entry_format; + for (i = 0; i < entry_format_count; i++) { + DebugInfoValue v = {{}}; + unsigned long dw_lnct = uleb128(&format); + unsigned long dw_form = uleb128(&format); + debug_info_reader_read_value(&reader, dw_form, &v); + if (dw_lnct == 1 /* DW_LNCT_path */ && v.type == VAL_cstr && out_path) + *out_path = v.as.ptr + v.off; + if (dw_lnct == 2 /* DW_LNCT_directory_index */ && v.type == VAL_uint && out_directory_index) + *out_directory_index = v.as.uint64; + } + if (j == idx) return 0; + } + + return reader.p; +} + +#ifdef USE_ELF +static unsigned long +uncompress_debug_section(ElfW(Shdr) *shdr, char *file, char **ptr) +{ + *ptr = NULL; +#ifdef SUPPORT_COMPRESSED_DEBUG_LINE + ElfW(Chdr) *chdr = (ElfW(Chdr) *)(file + shdr->sh_offset); + unsigned long destsize = chdr->ch_size; + int ret = 0; + + if (chdr->ch_type != ELFCOMPRESS_ZLIB) { + /* unsupported compression type */ + return 0; + } + + *ptr = malloc(destsize); + if (!*ptr) return 0; + ret = uncompress((Bytef *)*ptr, &destsize, + (const Bytef*)chdr + sizeof(ElfW(Chdr)), + shdr->sh_size - sizeof(ElfW(Chdr))); + if (ret != Z_OK) goto fail; + return destsize; + +fail: + free(*ptr); + *ptr = NULL; +#endif + return 0; +} + /* read file and fill lines */ static uintptr_t fill_lines(int num_traces, void **traces, int check_debuglink, @@ -485,10 +2007,10 @@ fill_lines(int num_traces, void **traces, int check_debuglink, { int i, j; char *shstr; - char *section_name; ElfW(Ehdr) *ehdr; ElfW(Shdr) *shdr, *shstr_shdr; - ElfW(Shdr) *debug_line_shdr = NULL, *gnu_debuglink_shdr = NULL; + ElfW(Shdr) *gnu_debuglink_shdr = NULL; + ElfW(Shdr) *note_gnu_build_id = NULL; int fd; off_t filesize; char *file; @@ -496,7 +2018,6 @@ fill_lines(int num_traces, void **traces, int check_debuglink, ElfW(Shdr) *dynsym_shdr = NULL, *dynstr_shdr = NULL; obj_info_t *obj = *objp; uintptr_t dladdr_fbase = 0; - bool compressed_p = false; fd = open(binary_filename, O_RDONLY); if (fd < 0) { @@ -525,6 +2046,7 @@ fill_lines(int num_traces, void **traces, int check_debuglink, kprintf("mmap: %s\n", strerror(e)); goto fail; } + close(fd); ehdr = (ElfW(Ehdr) *)file; if (memcmp(ehdr->e_ident, "\177ELF", 4) != 0) { @@ -532,11 +2054,8 @@ fill_lines(int num_traces, void **traces, int check_debuglink, * Huh? Maybe filename was overridden by setproctitle() and * it match non-elf file. */ - close(fd); goto fail; } - - obj->fd = fd; obj->mapped = file; obj->mapped_size = (size_t)filesize; @@ -546,7 +2065,7 @@ fill_lines(int num_traces, void **traces, int check_debuglink, shstr = file + shstr_shdr->sh_offset; for (i = 0; i < ehdr->e_shnum; i++) { - section_name = shstr + shdr[i].sh_name; + char *section_name = shstr + shdr[i].sh_name; switch (shdr[i].sh_type) { case SHT_STRTAB: if (!strcmp(section_name, ".strtab")) { @@ -564,16 +2083,44 @@ fill_lines(int num_traces, void **traces, int check_debuglink, /* if (!strcmp(section_name, ".dynsym")) */ dynsym_shdr = shdr + i; break; + case SHT_NOTE: + if (!strcmp(section_name, ".note.gnu.build-id")) { + note_gnu_build_id = shdr + i; + } + break; case SHT_PROGBITS: - if (!strcmp(section_name, ".debug_line")) { - if (shdr[i].sh_flags & SHF_COMPRESSED) { - compressed_p = true; - } - debug_line_shdr = shdr + i; - } - else if (!strcmp(section_name, ".gnu_debuglink")) { + if (!strcmp(section_name, ".gnu_debuglink")) { gnu_debuglink_shdr = shdr + i; } + else { + const char *debug_section_names[] = { + ".debug_abbrev", + ".debug_info", + ".debug_line", + ".debug_ranges", + ".debug_str_offsets", + ".debug_addr", + ".debug_rnglists", + ".debug_str", + ".debug_line_str" + }; + + for (j=0; j < DWARF_SECTION_COUNT; j++) { + struct dwarf_section *s = obj_dwarf_section_at(obj, j); + + if (strcmp(section_name, debug_section_names[j]) != 0) + continue; + + s->ptr = file + shdr[i].sh_offset; + s->size = shdr[i].sh_size; + s->flags = shdr[i].sh_flags; + if (s->flags & SHF_COMPRESSED) { + s->size = uncompress_debug_section(&shdr[i], file, &s->ptr); + if (!s->size) goto fail; + } + break; + } + } break; } } @@ -585,20 +2132,22 @@ fill_lines(int num_traces, void **traces, int check_debuglink, char *strtab = file + dynstr_shdr->sh_offset; ElfW(Sym) *symtab = (ElfW(Sym) *)(file + dynsym_shdr->sh_offset); int symtab_count = (int)(dynsym_shdr->sh_size / sizeof(ElfW(Sym))); - for (j = 0; j < symtab_count; j++) { - ElfW(Sym) *sym = &symtab[j]; - Dl_info info; - void *h, *s; - if (ELF_ST_TYPE(sym->st_info) != STT_FUNC || sym->st_size <= 0) continue; - h = dlopen(NULL, RTLD_NOW|RTLD_LOCAL); - if (!h) continue; - s = dlsym(h, strtab + sym->st_name); - if (!s) continue; - if (dladdr(s, &info)) { - dladdr_fbase = (uintptr_t)info.dli_fbase; - break; - } - } + void *handle = dlopen(NULL, RTLD_NOW|RTLD_LOCAL); + if (handle) { + for (j = 0; j < symtab_count; j++) { + ElfW(Sym) *sym = &symtab[j]; + Dl_info info; + void *s; + if (ELF_ST_TYPE(sym->st_info) != STT_FUNC || sym->st_size == 0) continue; + s = dlsym(handle, strtab + sym->st_name); + if (s && dladdr(s, &info)) { + obj->base_addr = dladdr_fbase; + dladdr_fbase = (uintptr_t)info.dli_fbase; + break; + } + } + dlclose(handle); + } if (ehdr->e_type == ET_EXEC) { obj->base_addr = 0; } @@ -609,33 +2158,48 @@ fill_lines(int num_traces, void **traces, int check_debuglink, } } - if (!symtab_shdr) { - symtab_shdr = dynsym_shdr; - strtab_shdr = dynstr_shdr; - } - - if (symtab_shdr && strtab_shdr) { - char *strtab = file + strtab_shdr->sh_offset; - ElfW(Sym) *symtab = (ElfW(Sym) *)(file + symtab_shdr->sh_offset); - int symtab_count = (int)(symtab_shdr->sh_size / sizeof(ElfW(Sym))); - for (j = 0; j < symtab_count; j++) { - ElfW(Sym) *sym = &symtab[j]; - uintptr_t saddr = (uintptr_t)sym->st_value + obj->base_addr; - if (ELF_ST_TYPE(sym->st_info) != STT_FUNC || sym->st_size <= 0) continue; - for (i = offset; i < num_traces; i++) { - uintptr_t d = (uintptr_t)traces[i] - saddr; - if (lines[i].line > 0 || d <= 0 || d > (uintptr_t)sym->st_size) - continue; - /* fill symbol name and addr from .symtab */ - lines[i].sname = strtab + sym->st_name; - lines[i].saddr = saddr; - lines[i].path = obj->path; - lines[i].base_addr = obj->base_addr; - } - } + if (obj->debug_info.ptr && obj->debug_abbrev.ptr) { + DebugInfoReader reader; + debug_info_reader_init(&reader, obj); + i = 0; + while (reader.p < reader.pend) { + /* fprintf(stderr, "%d:%tx: CU[%d]\n", __LINE__, reader.p - reader.obj->debug_info.ptr, i++); */ + if (di_read_cu(&reader)) goto use_symtab; + debug_info_read(&reader, num_traces, traces, lines, offset); + } + } + else { + /* This file doesn't have dwarf, use symtab or dynsym */ +use_symtab: + if (!symtab_shdr) { + /* This file doesn't have symtab, use dynsym instead */ + symtab_shdr = dynsym_shdr; + strtab_shdr = dynstr_shdr; + } + + if (symtab_shdr && strtab_shdr) { + char *strtab = file + strtab_shdr->sh_offset; + ElfW(Sym) *symtab = (ElfW(Sym) *)(file + symtab_shdr->sh_offset); + int symtab_count = (int)(symtab_shdr->sh_size / sizeof(ElfW(Sym))); + for (j = 0; j < symtab_count; j++) { + ElfW(Sym) *sym = &symtab[j]; + uintptr_t saddr = (uintptr_t)sym->st_value + obj->base_addr; + if (ELF_ST_TYPE(sym->st_info) != STT_FUNC) continue; + for (i = offset; i < num_traces; i++) { + uintptr_t d = (uintptr_t)traces[i] - saddr; + if (lines[i].line > 0 || d > (uintptr_t)sym->st_size) + continue; + /* fill symbol name and addr from .symtab */ + if (!lines[i].sname) lines[i].sname = strtab + sym->st_name; + lines[i].saddr = saddr; + lines[i].path = obj->path; + lines[i].base_addr = obj->base_addr; + } + } + } } - if (!debug_line_shdr) { + if (!obj->debug_line.ptr) { /* This file doesn't have .debug_line section, let's check .gnu_debuglink section instead. */ if (gnu_debuglink_shdr && check_debuglink) { @@ -643,23 +2207,251 @@ fill_lines(int num_traces, void **traces, int check_debuglink, num_traces, traces, objp, lines, offset); } + if (note_gnu_build_id && check_debuglink) { + ElfW(Nhdr) *nhdr = (ElfW(Nhdr)*) (file + note_gnu_build_id->sh_offset); + const char *build_id = (char *)(nhdr + 1) + nhdr->n_namesz; + follow_debuglink_build_id(build_id, nhdr->n_descsz, + num_traces, traces, + objp, lines, offset); + } goto finish; } - if (!compressed_p && - parse_debug_line(num_traces, traces, - file + debug_line_shdr->sh_offset, - debug_line_shdr->sh_size, - obj, lines, offset)) - goto fail; + if (parse_debug_line(num_traces, traces, + obj->debug_line.ptr, + obj->debug_line.size, + obj, lines, offset) == -1) + goto fail; + finish: return dladdr_fbase; fail: return (uintptr_t)-1; } +#else /* Mach-O */ +/* read file and fill lines */ +static uintptr_t +fill_lines(int num_traces, void **traces, int check_debuglink, + obj_info_t **objp, line_info_t *lines, int offset) +{ +# ifdef __LP64__ +# define LP(x) x##_64 +# else +# define LP(x) x +# endif + int fd; + off_t filesize; + char *file, *p = NULL; + obj_info_t *obj = *objp; + struct LP(mach_header) *header; + uintptr_t dladdr_fbase = 0; + + { + char *s = binary_filename; + char *base = strrchr(binary_filename, '/')+1; + size_t max = PATH_MAX; + size_t size = strlen(binary_filename); + size_t basesize = size - (base - binary_filename); + s += size; + max -= size; + p = s; + size = strlcpy(s, ".dSYM/Contents/Resources/DWARF/", max); + if (size == 0) goto fail; + s += size; + max -= size; + if (max <= basesize) goto fail; + memcpy(s, base, basesize); + s[basesize] = 0; + + fd = open(binary_filename, O_RDONLY); + if (fd < 0) { + *p = 0; /* binary_filename becomes original file name */ + fd = open(binary_filename, O_RDONLY); + if (fd < 0) { + goto fail; + } + } + } + + filesize = lseek(fd, 0, SEEK_END); + if (filesize < 0) { + int e = errno; + close(fd); + kprintf("lseek: %s\n", strerror(e)); + goto fail; + } +#if SIZEOF_OFF_T > SIZEOF_SIZE_T + if (filesize > (off_t)SIZE_MAX) { + close(fd); + kprintf("Too large file %s\n", binary_filename); + goto fail; + } +#endif + lseek(fd, 0, SEEK_SET); + /* async-signal unsafe */ + file = (char *)mmap(NULL, (size_t)filesize, PROT_READ, MAP_SHARED, fd, 0); + if (file == MAP_FAILED) { + int e = errno; + close(fd); + kprintf("mmap: %s\n", strerror(e)); + goto fail; + } + close(fd); + + obj->mapped = file; + obj->mapped_size = (size_t)filesize; + + header = (struct LP(mach_header) *)file; + if (header->magic == LP(MH_MAGIC)) { + /* non universal binary */ + p = file; + } + else if (header->magic == FAT_CIGAM) { + struct LP(mach_header) *mhp = _NSGetMachExecuteHeader(); + struct fat_header *fat = (struct fat_header *)file; + char *q = file + sizeof(*fat); + uint32_t nfat_arch = __builtin_bswap32(fat->nfat_arch); + /* fprintf(stderr,"%d: fat:%s %d\n",__LINE__, binary_filename,nfat_arch); */ + for (uint32_t i = 0; i < nfat_arch; i++) { + struct fat_arch *arch = (struct fat_arch *)q; + cpu_type_t cputype = __builtin_bswap32(arch->cputype); + cpu_subtype_t cpusubtype = __builtin_bswap32(arch->cpusubtype); + uint32_t offset = __builtin_bswap32(arch->offset); + /* fprintf(stderr,"%d: fat %d %x/%x %x/%x\n",__LINE__, i, mhp->cputype,mhp->cpusubtype, cputype,cpusubtype); */ + if (mhp->cputype == cputype && + (cpu_subtype_t)(mhp->cpusubtype & ~CPU_SUBTYPE_MASK) == cpusubtype) { + p = file + offset; + file = p; + header = (struct LP(mach_header) *)p; + if (header->magic == LP(MH_MAGIC)) { + goto found_mach_header; + } + break; + } + q += sizeof(*arch); + } + kprintf("'%s' is not a Mach-O universal binary file!\n",binary_filename); + close(fd); + goto fail; + } + else { + kprintf("'%s' is not a " +# ifdef __LP64__ + "64" +# else + "32" +# endif + "-bit Mach-O file!\n",binary_filename); + close(fd); + goto fail; + } +found_mach_header: + p += sizeof(*header); + + for (uint32_t i = 0; i < (uint32_t)header->ncmds; i++) { + struct load_command *lcmd = (struct load_command *)p; + switch (lcmd->cmd) { + case LP(LC_SEGMENT): + { + static const char *debug_section_names[] = { + "__debug_abbrev", + "__debug_info", + "__debug_line", + "__debug_ranges", + "__debug_str_offsets", + "__debug_addr", + "__debug_rnglists", + "__debug_str", + "__debug_line_str", + }; + struct LP(segment_command) *scmd = (struct LP(segment_command) *)lcmd; + if (strcmp(scmd->segname, "__TEXT") == 0) { + obj->vmaddr = scmd->vmaddr; + } + else if (strcmp(scmd->segname, "__DWARF") == 0) { + p += sizeof(struct LP(segment_command)); + for (uint64_t i = 0; i < scmd->nsects; i++) { + struct LP(section) *sect = (struct LP(section) *)p; + p += sizeof(struct LP(section)); + for (int j=0; j < DWARF_SECTION_COUNT; j++) { + struct dwarf_section *s = obj_dwarf_section_at(obj, j); + + if (strcmp(sect->sectname, debug_section_names[j]) != 0) + continue; + + s->ptr = file + sect->offset; + s->size = sect->size; + s->flags = sect->flags; + if (s->flags & SHF_COMPRESSED) { + goto fail; + } + break; + } + } + } + } + break; + + case LC_SYMTAB: + { + struct symtab_command *cmd = (struct symtab_command *)lcmd; + struct LP(nlist) *nl = (struct LP(nlist) *)(file + cmd->symoff); + char *strtab = file + cmd->stroff, *sname = 0; + uint32_t j; + uintptr_t saddr = 0; + /* kprintf("[%2d]: %x/symtab %p\n", i, cmd->cmd, (void *)p); */ + for (j = 0; j < cmd->nsyms; j++) { + uintptr_t symsize, d; + struct LP(nlist) *e = &nl[j]; + /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */ + if (e->n_type != N_FUN) continue; + if (e->n_sect) { + saddr = (uintptr_t)e->n_value + obj->base_addr - obj->vmaddr; + sname = strtab + e->n_un.n_strx; + /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */ + continue; + } + for (int k = offset; k < num_traces; k++) { + d = (uintptr_t)traces[k] - saddr; + symsize = e->n_value; + /* kprintf("%lx %lx %lx\n",saddr,symsize,traces[k]); */ + if (lines[k].line > 0 || d > (uintptr_t)symsize) + continue; + /* fill symbol name and addr from .symtab */ + if (!lines[k].sname) lines[k].sname = sname; + lines[k].saddr = saddr; + lines[k].path = obj->path; + lines[k].base_addr = obj->base_addr; + } + } + } + } + p += lcmd->cmdsize; + } + + if (obj->debug_info.ptr && obj->debug_abbrev.ptr) { + DebugInfoReader reader; + debug_info_reader_init(&reader, obj); + while (reader.p < reader.pend) { + if (di_read_cu(&reader)) goto fail; + debug_info_read(&reader, num_traces, traces, lines, offset); + } + } + + if (parse_debug_line(num_traces, traces, + obj->debug_line.ptr, + obj->debug_line.size, + obj, lines, offset) == -1) + goto fail; + + return dladdr_fbase; +fail: + return (uintptr_t)-1; +} +#endif #define HAVE_MAIN_EXE_PATH -#if defined(__FreeBSD__) +#if defined(__FreeBSD__) || defined(__DragonFly__) # include <sys/sysctl.h> #endif /* ssize_t main_exe_path(void) @@ -668,16 +2460,21 @@ fail: * and returns strlen(binary_filename). * it is NUL terminated. */ -#if defined(__linux__) +#if defined(__linux__) || defined(__NetBSD__) static ssize_t main_exe_path(void) { -# define PROC_SELF_EXE "/proc/self/exe" +# if defined(__linux__) +# define PROC_SELF_EXE "/proc/self/exe" +# elif defined(__NetBSD__) +# define PROC_SELF_EXE "/proc/curproc/exe" +# endif ssize_t len = readlink(PROC_SELF_EXE, binary_filename, PATH_MAX); + if (len < 0) return 0; binary_filename[len] = 0; return len; } -#elif defined(__FreeBSD__) +#elif defined(__FreeBSD__) || defined(__DragonFly__) static ssize_t main_exe_path(void) { @@ -691,10 +2488,69 @@ main_exe_path(void) len--; /* sysctl sets strlen+1 */ return len; } +#elif defined(HAVE_LIBPROC_H) +static ssize_t +main_exe_path(void) +{ + int len = proc_pidpath(getpid(), binary_filename, PATH_MAX); + if (len == 0) return 0; + binary_filename[len] = 0; + return len; +} #else #undef HAVE_MAIN_EXE_PATH #endif +static void +print_line0(line_info_t *line, void *address) +{ + uintptr_t addr = (uintptr_t)address; + uintptr_t d = addr - line->saddr; + if (!address) { + /* inlined */ + if (line->dirname && line->dirname[0]) { + kprintf("%s(%s) %s/%s:%d\n", line->path, line->sname, line->dirname, line->filename, line->line); + } + else { + kprintf("%s(%s) %s:%d\n", line->path, line->sname, line->filename, line->line); + } + } + else if (!line->path) { + kprintf("[0x%"PRIxPTR"]\n", addr); + } + else if (!line->sname) { + kprintf("%s(0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, addr-line->base_addr, addr); + } + else if (!line->saddr) { + kprintf("%s(%s) [0x%"PRIxPTR"]\n", line->path, line->sname, addr); + } + else if (line->line <= 0) { + kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, line->sname, + d, addr); + } + else if (!line->filename) { + kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] ???:%d\n", line->path, line->sname, + d, addr, line->line); + } + else if (line->dirname && line->dirname[0]) { + kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s/%s:%d\n", line->path, line->sname, + d, addr, line->dirname, line->filename, line->line); + } + else { + kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s:%d\n", line->path, line->sname, + d, addr, line->filename, line->line); + } +} + +static void +print_line(line_info_t *line, void *address) +{ + print_line0(line, address); + if (line->next) { + print_line(line->next, NULL); + } +} + void rb_dump_backtrace_with_lines(int num_traces, void **traces) { @@ -704,6 +2560,7 @@ rb_dump_backtrace_with_lines(int num_traces, void **traces) obj_info_t *obj = NULL; /* 2 is NULL + main executable */ void **dladdr_fbases = (void **)calloc(num_traces+2, sizeof(void *)); + #ifdef HAVE_MAIN_EXE_PATH char *main_path = NULL; /* used on printing backtrace */ ssize_t len; @@ -734,8 +2591,8 @@ rb_dump_backtrace_with_lines(int num_traces, void **traces) /* if the binary is strip-ed, this may effect */ for (p=dladdr_fbases; *p; p++) { if (*p == info.dli_fbase) { - lines[i].path = info.dli_fname; - lines[i].sname = info.dli_sname; + if (info.dli_fname) lines[i].path = info.dli_fname; + if (info.dli_sname) lines[i].sname = info.dli_sname; goto next_line; } } @@ -745,7 +2602,11 @@ rb_dump_backtrace_with_lines(int num_traces, void **traces) obj->base_addr = (uintptr_t)info.dli_fbase; path = info.dli_fname; obj->path = path; - lines[i].path = path; + if (path) lines[i].path = path; + if (info.dli_sname) { + lines[i].sname = info.dli_sname; + lines[i].saddr = (uintptr_t)info.dli_saddr; + } strlcpy(binary_filename, path, PATH_MAX); if (fill_lines(num_traces, traces, 1, &obj, lines, i) == (uintptr_t)-1) break; @@ -756,46 +2617,36 @@ next_line: /* output */ for (i = 0; i < num_traces; i++) { - line_info_t *line = &lines[i]; - uintptr_t addr = (uintptr_t)traces[i]; - uintptr_t d = addr - line->saddr; - if (!line->path) { - kprintf("[0x%lx]\n", addr); - } - else if (!line->saddr || !line->sname) { - kprintf("%s(0x%lx) [0x%lx]\n", line->path, addr-line->base_addr, addr); - } - else if (line->line <= 0) { - kprintf("%s(%s+0x%lx) [0x%lx]\n", line->path, line->sname, - d, addr); - } - else if (!line->filename) { - kprintf("%s(%s+0x%lx) [0x%lx] ???:%d\n", line->path, line->sname, - d, addr, line->line); - } - else if (line->dirname && line->dirname[0]) { - kprintf("%s(%s+0x%lx) [0x%lx] %s/%s:%d\n", line->path, line->sname, - d, addr, line->dirname, line->filename, line->line); - } - else { - kprintf("%s(%s+0x%lx) [0x%lx] %s:%d\n", line->path, line->sname, - d, addr, line->filename, line->line); - } + print_line(&lines[i], traces[i]); + /* FreeBSD's backtrace may show _start and so on */ - if (line->sname && strcmp("main", line->sname) == 0) + if (lines[i].sname && strcmp("main", lines[i].sname) == 0) break; } /* free */ while (obj) { obj_info_t *o = obj; - obj = o->next; - if (o->fd) { - munmap(o->mapped, o->mapped_size); - close(o->fd); + for (i=0; i < DWARF_SECTION_COUNT; i++) { + struct dwarf_section *s = obj_dwarf_section_at(obj, i); + if (s->flags & SHF_COMPRESSED) { + free(s->ptr); + } + } + if (obj->mapped_size) { + munmap(obj->mapped, obj->mapped_size); } + obj = o->next; free(o); } + for (i = 0; i < num_traces; i++) { + line_info_t *line = lines[i].next; + while (line) { + line_info_t *l = line; + line = line->next; + free(l); + } + } free(lines); free(dladdr_fbases); } @@ -841,7 +2692,7 @@ next_line: #define MAXNBUF (sizeof(intmax_t) * CHAR_BIT + 1) static inline int toupper(int c) { return ('A' <= c && c <= 'Z') ? (c&0x5f) : c; } #define hex2ascii(hex) (hex2ascii_data[hex]) -char const hex2ascii_data[] = "0123456789abcdefghijklmnopqrstuvwxyz"; +static const char hex2ascii_data[] = "0123456789abcdefghijklmnopqrstuvwxyz"; static inline int imax(int a, int b) { return (a > b ? a : b); } static int kvprintf(char const *fmt, void (*func)(int), void *arg, int radix, va_list ap); @@ -855,7 +2706,7 @@ static void putce(int c) (void)ret; } -int +static int kprintf(const char *fmt, ...) { va_list ap; diff --git a/addr2line.h b/addr2line.h index d99f010934..f09b665800 100644 --- a/addr2line.h +++ b/addr2line.h @@ -1,3 +1,5 @@ +#ifndef RUBY_ADDR2LINE_H +#define RUBY_ADDR2LINE_H /********************************************************************** addr2line.h - @@ -8,10 +10,7 @@ **********************************************************************/ -#ifndef RUBY_ADDR2LINE_H -#define RUBY_ADDR2LINE_H - -#ifdef USE_ELF +#if (defined(USE_ELF) || defined(HAVE_MACH_O_LOADER_H)) void rb_dump_backtrace_with_lines(int num_traces, void **traces); diff --git a/appveyor.yml b/appveyor.yml deleted file mode 100644 index 22c7bdce6a..0000000000 --- a/appveyor.yml +++ /dev/null @@ -1,55 +0,0 @@ ---- -shallow_clone: true -platform: - - x64 -environment: - ruby_version: "24-%Platform%" - zlib_version: "1.2.11" - matrix: - - vs: "120" -install: - - chcp - - SET BITS=%Platform:x86=32% - - SET BITS=%BITS:x=% - - SET OPENSSL_DIR=c:\OpenSSL-Win%BITS% - - CALL SET vcvars=%%^VS%VS%COMNTOOLS^%%..\..\VC\vcvarsall.bat - - SET vcvars - - '"%vcvars%" %Platform:x64=amd64%' - - SET ruby_path=C:\Ruby%ruby_version:-x86=% - - SET PATH=\usr\local\bin;%ruby_path%\bin;%PATH%;C:\msys64\mingw64\bin;C:\msys64\usr\bin - - ruby --version - - 'cl' - - SET - - echo> Makefile srcdir=. - - echo>> Makefile MSC_VER=0 - - echo>> Makefile RT=none - - echo>> Makefile RT_VER=0 - - echo>> Makefile BUILTIN_ENCOBJS=nul - - type win32\Makefile.sub >> Makefile - - nmake %mflags% touch-unicode-files - - nmake %mflags% up incs UNICODE_FILES=. - - del Makefile - - mkdir \usr\local\bin - - mkdir \usr\local\include - - mkdir \usr\local\lib - - appveyor DownloadFile https://zlib.net/zlib%zlib_version:.=%.zip - - 7z x -o%APPVEYOR_BUILD_FOLDER%\ext\zlib zlib%zlib_version:.=%.zip - - for %%I in (%OPENSSL_DIR%\*.dll) do mklink /h \usr\local\bin\%%~nxI %%I - - mkdir %Platform%-mswin_%vs% - - ps: Get-ChildItem "win32" -Recurse | foreach {$_.Attributes = 'Readonly'} - - ps: Get-Item $env:Platform"-mswin_"$env:vs | foreach {$_.Attributes = 'Normal'} -build_script: - - cd %APPVEYOR_BUILD_FOLDER% - - cd %Platform%-mswin_%vs% - - ..\win32\configure.bat --without-ext=+,dbm,gdbm,readline --with-opt-dir=/usr/local --with-openssl-dir=%OPENSSL_DIR:\=/% - - nmake -l - - nmake install-nodoc - - \usr\bin\ruby -v -e "p :locale => Encoding.find('locale'), :filesystem => Encoding.find('filesystem')" -test_script: - - set /a JOBS=%NUMBER_OF_PROCESSORS% - - nmake -l "TESTOPTS=-v -q" btest - - nmake -l "TESTOPTS=-v -q" test-basic - - nmake -l "TESTOPTS=-q -j%JOBS%" test-all - - nmake -l test-spec -matrix: - fast_finish: true @@ -11,21 +11,65 @@ **********************************************************************/ -#include "ruby/encoding.h" -#include "ruby/util.h" -#include "ruby/st.h" +#include "debug_counter.h" +#include "id.h" #include "internal.h" +#include "internal/array.h" +#include "internal/compar.h" +#include "internal/enum.h" +#include "internal/gc.h" +#include "internal/hash.h" +#include "internal/numeric.h" +#include "internal/object.h" +#include "internal/proc.h" +#include "internal/rational.h" +#include "internal/vm.h" #include "probes.h" -#include "id.h" -#include "debug_counter.h" +#include "ruby/encoding.h" +#include "ruby/st.h" +#include "ruby/util.h" +#include "transient_heap.h" +#include "builtin.h" -#ifndef ARRAY_DEBUG +#if !ARRAY_DEBUG +# undef NDEBUG # define NDEBUG #endif #include "ruby_assert.h" VALUE rb_cArray; +/* Flags of RArray + * + * 1: RARRAY_EMBED_FLAG + * The array is embedded (its contents follow the header, rather than + * being on a separately allocated buffer). + * 2: RARRAY_SHARED_FLAG (equal to ELTS_SHARED) + * The array is shared. The buffer this array points to is owned by + * another array (the shared root). + * if USE_RVARGC + * 3-9: RARRAY_EMBED_LEN + * The length of the array when RARRAY_EMBED_FLAG is set. + * else + * 3-4: RARRAY_EMBED_LEN + * The length of the array when RARRAY_EMBED_FLAG is set. + * endif + * 12: RARRAY_SHARED_ROOT_FLAG + * The array is a shared root that does reference counting. The buffer + * this array points to is owned by this array but may be pointed to + * by other arrays. + * Note: Frozen arrays may be a shared root without this flag being + * set. Frozen arrays do not have reference counting because + * they cannot be modified. Not updating the reference count + * improves copy-on-write performance. Their reference count is + * assumed to be infinity. + * 13: RARRAY_TRANSIENT_FLAG + * The buffer of the array is allocated on the transient heap. + * 14: RARRAY_PTR_IN_USE_FLAG + * The buffer of the array is in use. This is only used during + * debugging. + */ + /* for OPTIMIZED_CMP: */ #define id_cmp idCmp @@ -33,33 +77,41 @@ VALUE rb_cArray; #define ARY_MAX_SIZE (LONG_MAX / (int)sizeof(VALUE)) #define SMALL_ARRAY_LEN 16 -# define ARY_SHARED_P(ary) \ - (assert(!FL_TEST((ary), ELTS_SHARED) || !FL_TEST((ary), RARRAY_EMBED_FLAG)), \ - FL_TEST((ary),ELTS_SHARED)!=0) -# define ARY_EMBED_P(ary) \ - (assert(!FL_TEST((ary), ELTS_SHARED) || !FL_TEST((ary), RARRAY_EMBED_FLAG)), \ - FL_TEST((ary), RARRAY_EMBED_FLAG)!=0) +RBIMPL_ATTR_MAYBE_UNUSED() +static int +should_be_T_ARRAY(VALUE ary) +{ + return RB_TYPE_P(ary, T_ARRAY); +} #define ARY_HEAP_PTR(a) (assert(!ARY_EMBED_P(a)), RARRAY(a)->as.heap.ptr) #define ARY_HEAP_LEN(a) (assert(!ARY_EMBED_P(a)), RARRAY(a)->as.heap.len) +#define ARY_HEAP_CAPA(a) (assert(!ARY_EMBED_P(a)), assert(!ARY_SHARED_ROOT_P(a)), \ + RARRAY(a)->as.heap.aux.capa) + #define ARY_EMBED_PTR(a) (assert(ARY_EMBED_P(a)), RARRAY(a)->as.ary) #define ARY_EMBED_LEN(a) \ (assert(ARY_EMBED_P(a)), \ (long)((RBASIC(a)->flags >> RARRAY_EMBED_LEN_SHIFT) & \ - (RARRAY_EMBED_LEN_MASK >> RARRAY_EMBED_LEN_SHIFT))) -#define ARY_HEAP_SIZE(a) (assert(!ARY_EMBED_P(a)), assert(ARY_OWNS_HEAP_P(a)), RARRAY(a)->as.heap.aux.capa * sizeof(VALUE)) + (RARRAY_EMBED_LEN_MASK >> RARRAY_EMBED_LEN_SHIFT))) +#define ARY_HEAP_SIZE(a) (assert(!ARY_EMBED_P(a)), assert(ARY_OWNS_HEAP_P(a)), ARY_CAPA(a) * sizeof(VALUE)) + +#define ARY_OWNS_HEAP_P(a) (assert(should_be_T_ARRAY((VALUE)(a))), \ + !FL_TEST_RAW((a), RARRAY_SHARED_FLAG|RARRAY_EMBED_FLAG)) -#define ARY_OWNS_HEAP_P(a) (!FL_TEST((a), ELTS_SHARED|RARRAY_EMBED_FLAG)) #define FL_SET_EMBED(a) do { \ assert(!ARY_SHARED_P(a)); \ FL_SET((a), RARRAY_EMBED_FLAG); \ + RARY_TRANSIENT_UNSET(a); \ + ary_verify(a); \ } while (0) + #define FL_UNSET_EMBED(ary) FL_UNSET((ary), RARRAY_EMBED_FLAG|RARRAY_EMBED_LEN_MASK) #define FL_SET_SHARED(ary) do { \ assert(!ARY_EMBED_P(ary)); \ - FL_SET((ary), ELTS_SHARED); \ + FL_SET((ary), RARRAY_SHARED_FLAG); \ } while (0) -#define FL_UNSET_SHARED(ary) FL_UNSET((ary), ELTS_SHARED) +#define FL_UNSET_SHARED(ary) FL_UNSET((ary), RARRAY_SHARED_FLAG) #define ARY_SET_PTR(ary, p) do { \ assert(!ARY_EMBED_P(ary)); \ @@ -69,7 +121,6 @@ VALUE rb_cArray; #define ARY_SET_EMBED_LEN(ary, n) do { \ long tmp_n = (n); \ assert(ARY_EMBED_P(ary)); \ - assert(!OBJ_FROZEN(ary)); \ RBASIC(ary)->flags &= ~RARRAY_EMBED_LEN_MASK; \ RBASIC(ary)->flags |= (tmp_n) << RARRAY_EMBED_LEN_SHIFT; \ } while (0) @@ -101,8 +152,8 @@ VALUE rb_cArray; } \ } while (0) -#define ARY_CAPA(ary) (ARY_EMBED_P(ary) ? RARRAY_EMBED_LEN_MAX : \ - ARY_SHARED_ROOT_P(ary) ? RARRAY_LEN(ary) : RARRAY(ary)->as.heap.aux.capa) +#define ARY_CAPA(ary) (ARY_EMBED_P(ary) ? ary_embed_capa(ary) : \ + ARY_SHARED_ROOT_P(ary) ? RARRAY_LEN(ary) : ARY_HEAP_CAPA(ary)) #define ARY_SET_CAPA(ary, n) do { \ assert(!ARY_EMBED_P(ary)); \ assert(!ARY_SHARED_P(ary)); \ @@ -110,44 +161,185 @@ VALUE rb_cArray; RARRAY(ary)->as.heap.aux.capa = (n); \ } while (0) -#define ARY_SHARED(ary) (assert(ARY_SHARED_P(ary)), RARRAY(ary)->as.heap.aux.shared) #define ARY_SET_SHARED(ary, value) do { \ const VALUE _ary_ = (ary); \ const VALUE _value_ = (value); \ assert(!ARY_EMBED_P(_ary_)); \ assert(ARY_SHARED_P(_ary_)); \ - assert(ARY_SHARED_ROOT_P(_value_)); \ - RB_OBJ_WRITE(_ary_, &RARRAY(_ary_)->as.heap.aux.shared, _value_); \ + assert(!OBJ_FROZEN(_ary_)); \ + assert(ARY_SHARED_ROOT_P(_value_) || OBJ_FROZEN(_value_)); \ + RB_OBJ_WRITE(_ary_, &RARRAY(_ary_)->as.heap.aux.shared_root, _value_); \ } while (0) -#define RARRAY_SHARED_ROOT_FLAG FL_USER5 -#define ARY_SHARED_ROOT_P(ary) (FL_TEST((ary), RARRAY_SHARED_ROOT_FLAG)) -#define ARY_SHARED_NUM(ary) \ - (assert(ARY_SHARED_ROOT_P(ary)), RARRAY(ary)->as.heap.aux.capa) -#define ARY_SHARED_OCCUPIED(ary) (ARY_SHARED_NUM(ary) == 1) -#define ARY_SET_SHARED_NUM(ary, value) do { \ + +#define ARY_SHARED_ROOT_OCCUPIED(ary) (!OBJ_FROZEN(ary) && ARY_SHARED_ROOT_REFCNT(ary) == 1) +#define ARY_SET_SHARED_ROOT_REFCNT(ary, value) do { \ assert(ARY_SHARED_ROOT_P(ary)); \ + assert(!OBJ_FROZEN(ary)); \ + assert((value) >= 0); \ RARRAY(ary)->as.heap.aux.capa = (value); \ } while (0) #define FL_SET_SHARED_ROOT(ary) do { \ + assert(!OBJ_FROZEN(ary)); \ assert(!ARY_EMBED_P(ary)); \ + assert(!RARRAY_TRANSIENT_P(ary)); \ FL_SET((ary), RARRAY_SHARED_ROOT_FLAG); \ } while (0) -#define ARY_SET(a, i, v) RARRAY_ASET((assert(!ARY_SHARED_P(a)), (a)), (i), (v)) +static inline void +ARY_SET(VALUE a, long i, VALUE v) +{ + assert(!ARY_SHARED_P(a)); + assert(!OBJ_FROZEN(a)); + + RARRAY_ASET(a, i, v); +} +#undef RARRAY_ASET + +static long +ary_embed_capa(VALUE ary) +{ +#if USE_RVARGC + size_t size = rb_gc_obj_slot_size(ary) - offsetof(struct RArray, as.ary); + assert(size % sizeof(VALUE) == 0); + return size / sizeof(VALUE); +#else + return RARRAY_EMBED_LEN_MAX; +#endif +} + +static size_t +ary_embed_size(long capa) +{ + return offsetof(struct RArray, as.ary) + (sizeof(VALUE) * capa); +} + +static bool +ary_embeddable_p(long capa) +{ +#if USE_RVARGC + return rb_gc_size_allocatable_p(ary_embed_size(capa)); +#else + return capa <= RARRAY_EMBED_LEN_MAX; +#endif +} + +bool +rb_ary_embeddable_p(VALUE ary) +{ + /* An array cannot be turned embeddable when the array is: + * - Shared root: other objects may point to the buffer of this array + * so we cannot make it embedded. + * - Frozen: this array may also be a shared root without the shared root + * flag. + * - Shared: we don't want to re-embed an array that points to a shared + * root (to save memory). + */ + return !(ARY_SHARED_ROOT_P(ary) || OBJ_FROZEN(ary) || ARY_SHARED_P(ary)); +} + +size_t +rb_ary_size_as_embedded(VALUE ary) +{ + size_t real_size; + + if (ARY_EMBED_P(ary)) { + real_size = ary_embed_size(ARY_EMBED_LEN(ary)); + } + else if (rb_ary_embeddable_p(ary)) { + real_size = ary_embed_size(ARY_HEAP_CAPA(ary)); + } + else { + real_size = sizeof(struct RArray); + } + return real_size; +} + + +#if ARRAY_DEBUG +#define ary_verify(ary) ary_verify_(ary, __FILE__, __LINE__) + +static VALUE +ary_verify_(VALUE ary, const char *file, int line) +{ + assert(RB_TYPE_P(ary, T_ARRAY)); + + if (ARY_SHARED_P(ary)) { + VALUE root = ARY_SHARED_ROOT(ary); + const VALUE *ptr = ARY_HEAP_PTR(ary); + const VALUE *root_ptr = RARRAY_CONST_PTR_TRANSIENT(root); + long len = ARY_HEAP_LEN(ary), root_len = RARRAY_LEN(root); + assert(ARY_SHARED_ROOT_P(root) || OBJ_FROZEN(root)); + assert(root_ptr <= ptr && ptr + len <= root_ptr + root_len); + ary_verify(root); + } + else if (ARY_EMBED_P(ary)) { + assert(!RARRAY_TRANSIENT_P(ary)); + assert(!ARY_SHARED_P(ary)); + assert(RARRAY_LEN(ary) <= ary_embed_capa(ary)); + } + else { +#if 1 + const VALUE *ptr = RARRAY_CONST_PTR_TRANSIENT(ary); + long i, len = RARRAY_LEN(ary); + volatile VALUE v; + if (len > 1) len = 1; /* check only HEAD */ + for (i=0; i<len; i++) { + v = ptr[i]; /* access check */ + } + v = v; +#endif + } + +#if USE_TRANSIENT_HEAP + if (RARRAY_TRANSIENT_P(ary)) { + assert(rb_transient_heap_managed_ptr_p(RARRAY_CONST_PTR_TRANSIENT(ary))); + } +#endif + + rb_transient_heap_verify(); + + return ary; +} void -rb_mem_clear(register VALUE *mem, register long size) +rb_ary_verify(VALUE ary) +{ + ary_verify(ary); +} +#else +#define ary_verify(ary) ((void)0) +#endif + +VALUE * +rb_ary_ptr_use_start(VALUE ary) +{ +#if ARRAY_DEBUG + FL_SET_RAW(ary, RARRAY_PTR_IN_USE_FLAG); +#endif + return (VALUE *)RARRAY_CONST_PTR_TRANSIENT(ary); +} + +void +rb_ary_ptr_use_end(VALUE ary) +{ +#if ARRAY_DEBUG + FL_UNSET_RAW(ary, RARRAY_PTR_IN_USE_FLAG); +#endif +} + +void +rb_mem_clear(VALUE *mem, long size) { while (size--) { - *mem++ = Qnil; + *mem++ = Qnil; } } static void ary_mem_clear(VALUE ary, long beg, long size) { - RARRAY_PTR_USE(ary, ptr, { - rb_mem_clear(ptr + beg, size); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + rb_mem_clear(ptr + beg, size); }); } @@ -155,44 +347,38 @@ static inline void memfill(register VALUE *mem, register long size, register VALUE val) { while (size--) { - *mem++ = val; + *mem++ = val; } } static void ary_memfill(VALUE ary, long beg, long size, VALUE val) { - RARRAY_PTR_USE(ary, ptr, { - memfill(ptr + beg, size, val); - RB_OBJ_WRITTEN(ary, Qundef, val); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + memfill(ptr + beg, size, val); + RB_OBJ_WRITTEN(ary, Qundef, val); }); } static void ary_memcpy0(VALUE ary, long beg, long argc, const VALUE *argv, VALUE buff_owner_ary) { -#if 1 assert(!ARY_SHARED_P(buff_owner_ary)); if (argc > (int)(128/sizeof(VALUE)) /* is magic number (cache line size) */) { - rb_gc_writebarrier_remember(buff_owner_ary); - RARRAY_PTR_USE(ary, ptr, { - MEMCPY(ptr+beg, argv, VALUE, argc); - }); + rb_gc_writebarrier_remember(buff_owner_ary); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMCPY(ptr+beg, argv, VALUE, argc); + }); } else { - int i; - RARRAY_PTR_USE(ary, ptr, { - for (i=0; i<argc; i++) { - RB_OBJ_WRITE(buff_owner_ary, &ptr[i+beg], argv[i]); - } - }); + int i; + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + for (i=0; i<argc; i++) { + RB_OBJ_WRITE(buff_owner_ary, &ptr[i+beg], argv[i]); + } + }); } -#else - /* giveup write barrier (traditional way) */ - RARRAY_PTR(buff_owner_ary); - MEMCPY(RARRAY_PTR(ary)+beg, argv, VALUE, argc); -#endif } static void @@ -201,49 +387,198 @@ ary_memcpy(VALUE ary, long beg, long argc, const VALUE *argv) ary_memcpy0(ary, beg, argc, argv, ary); } +static VALUE * +ary_heap_alloc(VALUE ary, size_t capa) +{ + VALUE *ptr = rb_transient_heap_alloc(ary, sizeof(VALUE) * capa); + + if (ptr != NULL) { + RARY_TRANSIENT_SET(ary); + } + else { + RARY_TRANSIENT_UNSET(ary); + ptr = ALLOC_N(VALUE, capa); + } + + return ptr; +} + +static void +ary_heap_free_ptr(VALUE ary, const VALUE *ptr, long size) +{ + if (RARRAY_TRANSIENT_P(ary)) { + /* ignore it */ + } + else { + ruby_sized_xfree((void *)ptr, size); + } +} + +static void +ary_heap_free(VALUE ary) +{ + if (RARRAY_TRANSIENT_P(ary)) { + RARY_TRANSIENT_UNSET(ary); + } + else { + ary_heap_free_ptr(ary, ARY_HEAP_PTR(ary), ARY_HEAP_SIZE(ary)); + } +} + +static size_t +ary_heap_realloc(VALUE ary, size_t new_capa) +{ + size_t alloc_capa = new_capa; + size_t old_capa = ARY_HEAP_CAPA(ary); + + if (RARRAY_TRANSIENT_P(ary)) { + if (new_capa <= old_capa) { + /* do nothing */ + alloc_capa = old_capa; + } + else { + VALUE *new_ptr = rb_transient_heap_alloc(ary, sizeof(VALUE) * new_capa); + + if (new_ptr == NULL) { + new_ptr = ALLOC_N(VALUE, new_capa); + RARY_TRANSIENT_UNSET(ary); + } + + MEMCPY(new_ptr, ARY_HEAP_PTR(ary), VALUE, old_capa); + ARY_SET_PTR(ary, new_ptr); + } + } + else { + SIZED_REALLOC_N(RARRAY(ary)->as.heap.ptr, VALUE, new_capa, old_capa); + } + ary_verify(ary); + + return alloc_capa; +} + +#if USE_TRANSIENT_HEAP +static inline void +rb_ary_transient_heap_evacuate_(VALUE ary, int transient, int promote) +{ + if (transient) { + assert(!ARY_SHARED_ROOT_P(ary)); + + VALUE *new_ptr; + const VALUE *old_ptr = ARY_HEAP_PTR(ary); + long capa = ARY_HEAP_CAPA(ary); + + assert(ARY_OWNS_HEAP_P(ary)); + assert(RARRAY_TRANSIENT_P(ary)); + assert(!ARY_PTR_USING_P(ary)); + + if (promote) { + new_ptr = ALLOC_N(VALUE, capa); + RARY_TRANSIENT_UNSET(ary); + } + else { + new_ptr = ary_heap_alloc(ary, capa); + } + + MEMCPY(new_ptr, old_ptr, VALUE, capa); + /* do not use ARY_SET_PTR() because they assert !frozen */ + RARRAY(ary)->as.heap.ptr = new_ptr; + } + + ary_verify(ary); +} + +void +rb_ary_transient_heap_evacuate(VALUE ary, int promote) +{ + rb_ary_transient_heap_evacuate_(ary, RARRAY_TRANSIENT_P(ary), promote); +} + +void +rb_ary_detransient(VALUE ary) +{ + assert(RARRAY_TRANSIENT_P(ary)); + rb_ary_transient_heap_evacuate_(ary, TRUE, TRUE); +} +#else +void +rb_ary_detransient(VALUE ary) +{ + /* do nothing */ +} +#endif + +void +rb_ary_make_embedded(VALUE ary) +{ + assert(rb_ary_embeddable_p(ary)); + if (!ARY_EMBED_P(ary)) { + const VALUE *buf = ARY_HEAP_PTR(ary); + long len = ARY_HEAP_LEN(ary); + bool was_transient = RARRAY_TRANSIENT_P(ary); + + // FL_SET_EMBED also unsets the transient flag + FL_SET_EMBED(ary); + ARY_SET_EMBED_LEN(ary, len); + + MEMCPY((void *)ARY_EMBED_PTR(ary), (void *)buf, VALUE, len); + + if (!was_transient) { + ary_heap_free_ptr(ary, buf, len * sizeof(VALUE)); + } + } +} + static void ary_resize_capa(VALUE ary, long capacity) { assert(RARRAY_LEN(ary) <= capacity); assert(!OBJ_FROZEN(ary)); assert(!ARY_SHARED_P(ary)); - if (capacity > RARRAY_EMBED_LEN_MAX) { + + if (capacity > ary_embed_capa(ary)) { + size_t new_capa = capacity; if (ARY_EMBED_P(ary)) { long len = ARY_EMBED_LEN(ary); - VALUE *ptr = ALLOC_N(VALUE, (capacity)); + VALUE *ptr = ary_heap_alloc(ary, capacity); + MEMCPY(ptr, ARY_EMBED_PTR(ary), VALUE, len); FL_UNSET_EMBED(ary); ARY_SET_PTR(ary, ptr); ARY_SET_HEAP_LEN(ary, len); } else { - SIZED_REALLOC_N(RARRAY(ary)->as.heap.ptr, VALUE, capacity, RARRAY(ary)->as.heap.aux.capa); + new_capa = ary_heap_realloc(ary, capacity); } - ARY_SET_CAPA(ary, (capacity)); + ARY_SET_CAPA(ary, new_capa); } else { if (!ARY_EMBED_P(ary)) { - long len = RARRAY_LEN(ary); - const VALUE *ptr = RARRAY_CONST_PTR(ary); + long len = ARY_HEAP_LEN(ary); + long old_capa = ARY_HEAP_CAPA(ary); + const VALUE *ptr = ARY_HEAP_PTR(ary); - if (len > capacity) len = capacity; + if (len > capacity) len = capacity; MEMCPY((VALUE *)RARRAY(ary)->as.ary, ptr, VALUE, len); + ary_heap_free_ptr(ary, ptr, old_capa); + FL_SET_EMBED(ary); ARY_SET_LEN(ary, len); - ruby_xfree((VALUE *)ptr); } } + + ary_verify(ary); } static inline void ary_shrink_capa(VALUE ary) { long capacity = ARY_HEAP_LEN(ary); - long old_capa = RARRAY(ary)->as.heap.aux.capa; + long old_capa = ARY_HEAP_CAPA(ary); assert(!ARY_SHARED_P(ary)); assert(old_capa >= capacity); - if (old_capa > capacity) - REALLOC_N(RARRAY(ary)->as.heap.ptr, VALUE, capacity); + if (old_capa > capacity) ary_heap_realloc(ary, capacity); + + ary_verify(ary); } static void @@ -252,106 +587,121 @@ ary_double_capa(VALUE ary, long min) long new_capa = ARY_CAPA(ary) / 2; if (new_capa < ARY_DEFAULT_SIZE) { - new_capa = ARY_DEFAULT_SIZE; + new_capa = ARY_DEFAULT_SIZE; } if (new_capa >= ARY_MAX_SIZE - min) { - new_capa = (ARY_MAX_SIZE - min) / 2; + new_capa = (ARY_MAX_SIZE - min) / 2; } new_capa += min; ary_resize_capa(ary, new_capa); + + ary_verify(ary); } static void -rb_ary_decrement_share(VALUE shared) +rb_ary_decrement_share(VALUE shared_root) { - if (shared) { - long num = ARY_SHARED_NUM(shared) - 1; - if (num == 0) { - rb_ary_free(shared); - rb_gc_force_recycle(shared); - } - else if (num > 0) { - ARY_SET_SHARED_NUM(shared, num); - } + if (!OBJ_FROZEN(shared_root)) { + long num = ARY_SHARED_ROOT_REFCNT(shared_root); + ARY_SET_SHARED_ROOT_REFCNT(shared_root, num - 1); } } static void rb_ary_unshare(VALUE ary) { - VALUE shared = RARRAY(ary)->as.heap.aux.shared; - rb_ary_decrement_share(shared); + VALUE shared_root = ARY_SHARED_ROOT(ary); + rb_ary_decrement_share(shared_root); FL_UNSET_SHARED(ary); } -static inline void -rb_ary_unshare_safe(VALUE ary) +static void +rb_ary_reset(VALUE ary) { - if (ARY_SHARED_P(ary) && !ARY_EMBED_P(ary)) { - rb_ary_unshare(ary); + if (ARY_OWNS_HEAP_P(ary)) { + ary_heap_free(ary); + } + else if (ARY_SHARED_P(ary)) { + rb_ary_unshare(ary); } + + FL_SET_EMBED(ary); + ARY_SET_EMBED_LEN(ary, 0); } static VALUE -rb_ary_increment_share(VALUE shared) +rb_ary_increment_share(VALUE shared_root) { - long num = ARY_SHARED_NUM(shared); - if (num >= 0) { - ARY_SET_SHARED_NUM(shared, num + 1); + if (!OBJ_FROZEN(shared_root)) { + long num = ARY_SHARED_ROOT_REFCNT(shared_root); + assert(num >= 0); + ARY_SET_SHARED_ROOT_REFCNT(shared_root, num + 1); } - return shared; + return shared_root; } static void -rb_ary_set_shared(VALUE ary, VALUE shared) +rb_ary_set_shared(VALUE ary, VALUE shared_root) { - rb_ary_increment_share(shared); + rb_ary_increment_share(shared_root); FL_SET_SHARED(ary); - ARY_SET_SHARED(ary, shared); + RB_DEBUG_COUNTER_INC(obj_ary_shared_create); + ARY_SET_SHARED(ary, shared_root); } static inline void rb_ary_modify_check(VALUE ary) { rb_check_frozen(ary); + ary_verify(ary); } void -rb_ary_modify(VALUE ary) +rb_ary_cancel_sharing(VALUE ary) { - rb_ary_modify_check(ary); if (ARY_SHARED_P(ary)) { - long shared_len, len = RARRAY_LEN(ary); - VALUE shared = ARY_SHARED(ary); - if (len <= RARRAY_EMBED_LEN_MAX) { - const VALUE *ptr = ARY_HEAP_PTR(ary); + long shared_len, len = RARRAY_LEN(ary); + VALUE shared_root = ARY_SHARED_ROOT(ary); + + ary_verify(shared_root); + + if (len <= ary_embed_capa(ary)) { + const VALUE *ptr = ARY_HEAP_PTR(ary); FL_UNSET_SHARED(ary); FL_SET_EMBED(ary); - MEMCPY((VALUE *)ARY_EMBED_PTR(ary), ptr, VALUE, len); - rb_ary_decrement_share(shared); + MEMCPY((VALUE *)ARY_EMBED_PTR(ary), ptr, VALUE, len); + rb_ary_decrement_share(shared_root); ARY_SET_EMBED_LEN(ary, len); } - else if (ARY_SHARED_OCCUPIED(shared) && len > ((shared_len = RARRAY_LEN(shared))>>1)) { - long shift = RARRAY_CONST_PTR(ary) - RARRAY_CONST_PTR(shared); - FL_UNSET_SHARED(ary); - ARY_SET_PTR(ary, RARRAY_CONST_PTR(shared)); - ARY_SET_CAPA(ary, shared_len); - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr, ptr+shift, VALUE, len); - }); - FL_SET_EMBED(shared); - rb_ary_decrement_share(shared); - } + else if (ARY_SHARED_ROOT_OCCUPIED(shared_root) && len > ((shared_len = RARRAY_LEN(shared_root))>>1)) { + long shift = RARRAY_CONST_PTR_TRANSIENT(ary) - RARRAY_CONST_PTR_TRANSIENT(shared_root); + FL_UNSET_SHARED(ary); + ARY_SET_PTR(ary, RARRAY_CONST_PTR_TRANSIENT(shared_root)); + ARY_SET_CAPA(ary, shared_len); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMMOVE(ptr, ptr+shift, VALUE, len); + }); + FL_SET_EMBED(shared_root); + rb_ary_decrement_share(shared_root); + } else { - VALUE *ptr = ALLOC_N(VALUE, len); - MEMCPY(ptr, RARRAY_CONST_PTR(ary), VALUE, len); + VALUE *ptr = ary_heap_alloc(ary, len); + MEMCPY(ptr, ARY_HEAP_PTR(ary), VALUE, len); rb_ary_unshare(ary); ARY_SET_CAPA(ary, len); ARY_SET_PTR(ary, ptr); } - rb_gc_writebarrier_remember(ary); + rb_gc_writebarrier_remember(ary); } + ary_verify(ary); +} + +void +rb_ary_modify(VALUE ary) +{ + rb_ary_modify_check(ary); + rb_ary_cancel_sharing(ary); } static VALUE @@ -362,48 +712,58 @@ ary_ensure_room_for_push(VALUE ary, long add_len) long capa; if (old_len > ARY_MAX_SIZE - add_len) { - rb_raise(rb_eIndexError, "index %ld too big", new_len); + rb_raise(rb_eIndexError, "index %ld too big", new_len); } if (ARY_SHARED_P(ary)) { - if (new_len > RARRAY_EMBED_LEN_MAX) { - VALUE shared = ARY_SHARED(ary); - if (ARY_SHARED_OCCUPIED(shared)) { - if (RARRAY_CONST_PTR(ary) - RARRAY_CONST_PTR(shared) + new_len <= RARRAY_LEN(shared)) { - rb_ary_modify_check(ary); - return shared; - } - else { - /* if array is shared, then it is likely it participate in push/shift pattern */ - rb_ary_modify(ary); - capa = ARY_CAPA(ary); - if (new_len > capa - (capa >> 6)) { - ary_double_capa(ary, new_len); - } - return ary; - } - } - } - rb_ary_modify(ary); + if (new_len > ary_embed_capa(ary)) { + VALUE shared_root = ARY_SHARED_ROOT(ary); + if (ARY_SHARED_ROOT_OCCUPIED(shared_root)) { + if (ARY_HEAP_PTR(ary) - RARRAY_CONST_PTR_TRANSIENT(shared_root) + new_len <= RARRAY_LEN(shared_root)) { + rb_ary_modify_check(ary); + + ary_verify(ary); + ary_verify(shared_root); + return shared_root; + } + else { + /* if array is shared, then it is likely it participate in push/shift pattern */ + rb_ary_modify(ary); + capa = ARY_CAPA(ary); + if (new_len > capa - (capa >> 6)) { + ary_double_capa(ary, new_len); + } + ary_verify(ary); + return ary; + } + } + } + ary_verify(ary); + rb_ary_modify(ary); } else { - rb_ary_modify_check(ary); + rb_ary_modify_check(ary); } capa = ARY_CAPA(ary); if (new_len > capa) { - ary_double_capa(ary, new_len); + ary_double_capa(ary, new_len); } + ary_verify(ary); return ary; } /* * call-seq: - * ary.freeze -> ary + * array.freeze -> self + * + * Freezes +self+; returns +self+: * - * Calls Object#freeze on +ary+ to prevent any further - * modification. A RuntimeError will be raised if a modification - * attempt is made. + * a = [] + * a.frozen? # => false + * a.freeze + * a.frozen? # => true * + * An attempt to modify a frozen \Array raises FrozenError. */ VALUE @@ -412,21 +772,6 @@ rb_ary_freeze(VALUE ary) return rb_obj_freeze(ary); } -/* - * call-seq: - * ary.frozen? -> true or false - * - * Return +true+ if this array is frozen (or temporarily frozen - * while being sorted). See also Object#frozen? - */ - -static VALUE -rb_ary_frozen_p(VALUE ary) -{ - if (OBJ_FROZEN(ary)) return Qtrue; - return Qfalse; -} - /* This can be used to take a snapshot of an array (with e.g. rb_ary_replace) and check later whether the array has been modified from the snapshot. The snapshot is cheap, though if @@ -438,18 +783,25 @@ VALUE rb_ary_shared_with_p(VALUE ary1, VALUE ary2) { if (!ARY_EMBED_P(ary1) && ARY_SHARED_P(ary1) && - !ARY_EMBED_P(ary2) && ARY_SHARED_P(ary2) && - RARRAY(ary1)->as.heap.aux.shared == RARRAY(ary2)->as.heap.aux.shared && - RARRAY(ary1)->as.heap.len == RARRAY(ary2)->as.heap.len) { - return Qtrue; + !ARY_EMBED_P(ary2) && ARY_SHARED_P(ary2) && + ARY_SHARED_ROOT(ary1) == ARY_SHARED_ROOT(ary2) && + ARY_HEAP_LEN(ary1) == ARY_HEAP_LEN(ary2)) { + return Qtrue; } return Qfalse; } static VALUE -ary_alloc(VALUE klass) +ary_alloc_embed(VALUE klass, long capa) { - NEWOBJ_OF(ary, struct RArray, klass, T_ARRAY | RARRAY_EMBED_FLAG | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0)); + size_t size = ary_embed_size(capa); + assert(rb_gc_size_allocatable_p(size)); +#if !USE_RVARGC + assert(size <= sizeof(struct RArray)); +#endif + RVARGC_NEWOBJ_OF(ary, struct RArray, klass, + T_ARRAY | RARRAY_EMBED_FLAG | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0), + size); /* Created array is: * FL_SET_EMBED((VALUE)ary); * ARY_SET_EMBED_LEN((VALUE)ary, 0); @@ -458,10 +810,19 @@ ary_alloc(VALUE klass) } static VALUE +ary_alloc_heap(VALUE klass) +{ + RVARGC_NEWOBJ_OF(ary, struct RArray, klass, + T_ARRAY | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0), + sizeof(struct RArray)); + return (VALUE)ary; +} + +static VALUE empty_ary_alloc(VALUE klass) { RUBY_DTRACE_CREATE_HOOK(ARRAY, 0); - return ary_alloc(klass); + return ary_alloc_embed(klass, 0); } static VALUE @@ -470,20 +831,24 @@ ary_new(VALUE klass, long capa) VALUE ary,*ptr; if (capa < 0) { - rb_raise(rb_eArgError, "negative array size (or size too big)"); + rb_raise(rb_eArgError, "negative array size (or size too big)"); } if (capa > ARY_MAX_SIZE) { - rb_raise(rb_eArgError, "array size too big"); + rb_raise(rb_eArgError, "array size too big"); } RUBY_DTRACE_CREATE_HOOK(ARRAY, capa); - ary = ary_alloc(klass); - if (capa > RARRAY_EMBED_LEN_MAX) { - ptr = ALLOC_N(VALUE, capa); - FL_UNSET_EMBED(ary); - ARY_SET_PTR(ary, ptr); + if (ary_embeddable_p(capa)) { + ary = ary_alloc_embed(klass, capa); + } + else { + ary = ary_alloc_heap(klass); ARY_SET_CAPA(ary, capa); + assert(!ARY_EMBED_P(ary)); + + ptr = ary_heap_alloc(ary, capa); + ARY_SET_PTR(ary, ptr); ARY_SET_HEAP_LEN(ary, 0); } @@ -499,7 +864,7 @@ rb_ary_new_capa(long capa) VALUE rb_ary_new(void) { - return rb_ary_new2(RARRAY_EMBED_LEN_MAX); + return rb_ary_new_capa(0); } VALUE @@ -513,7 +878,7 @@ VALUE va_start(ar, n); for (i=0; i<n; i++) { - ARY_SET(ary, i, va_arg(ar, VALUE)); + ARY_SET(ary, i, va_arg(ar, VALUE)); } va_end(ar); @@ -521,15 +886,15 @@ VALUE return ary; } -VALUE +MJIT_FUNC_EXPORTED VALUE rb_ary_tmp_new_from_values(VALUE klass, long n, const VALUE *elts) { VALUE ary; ary = ary_new(klass, n); if (n > 0 && elts) { - ary_memcpy(ary, 0, n, elts); - ARY_SET_LEN(ary, n); + ary_memcpy(ary, 0, n, elts); + ARY_SET_LEN(ary, n); } return ary; @@ -541,16 +906,89 @@ rb_ary_new_from_values(long n, const VALUE *elts) return rb_ary_tmp_new_from_values(rb_cArray, n, elts); } +static VALUE +ec_ary_alloc_embed(rb_execution_context_t *ec, VALUE klass, long capa) +{ + size_t size = ary_embed_size(capa); + assert(rb_gc_size_allocatable_p(size)); +#if !USE_RVARGC + assert(size <= sizeof(struct RArray)); +#endif + RB_RVARGC_EC_NEWOBJ_OF(ec, ary, struct RArray, klass, + T_ARRAY | RARRAY_EMBED_FLAG | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0), + size); + /* Created array is: + * FL_SET_EMBED((VALUE)ary); + * ARY_SET_EMBED_LEN((VALUE)ary, 0); + */ + return (VALUE)ary; +} + +static VALUE +ec_ary_alloc_heap(rb_execution_context_t *ec, VALUE klass) +{ + RB_RVARGC_EC_NEWOBJ_OF(ec, ary, struct RArray, klass, + T_ARRAY | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0), + sizeof(struct RArray)); + return (VALUE)ary; +} + +static VALUE +ec_ary_new(rb_execution_context_t *ec, VALUE klass, long capa) +{ + VALUE ary,*ptr; + + if (capa < 0) { + rb_raise(rb_eArgError, "negative array size (or size too big)"); + } + if (capa > ARY_MAX_SIZE) { + rb_raise(rb_eArgError, "array size too big"); + } + + RUBY_DTRACE_CREATE_HOOK(ARRAY, capa); + + if (ary_embeddable_p(capa)) { + ary = ec_ary_alloc_embed(ec, klass, capa); + } + else { + ary = ec_ary_alloc_heap(ec, klass); + ARY_SET_CAPA(ary, capa); + assert(!ARY_EMBED_P(ary)); + + ptr = ary_heap_alloc(ary, capa); + ARY_SET_PTR(ary, ptr); + ARY_SET_HEAP_LEN(ary, 0); + } + + return ary; +} + VALUE -rb_ary_tmp_new(long capa) +rb_ec_ary_new_from_values(rb_execution_context_t *ec, long n, const VALUE *elts) { - return ary_new(0, capa); + VALUE ary; + + ary = ec_ary_new(ec, rb_cArray, n); + if (n > 0 && elts) { + ary_memcpy(ary, 0, n, elts); + ARY_SET_LEN(ary, n); + } + + return ary; } VALUE -rb_ary_tmp_new_fill(long capa) +rb_ary_hidden_new(long capa) { VALUE ary = ary_new(0, capa); + rb_ary_transient_heap_evacuate(ary, TRUE); + return ary; +} + +VALUE +rb_ary_hidden_new_fill(long capa) +{ + VALUE ary = rb_ary_hidden_new(capa); ary_memfill(ary, 0, capa, Qnil); ARY_SET_LEN(ary, capa); return ary; @@ -560,11 +998,29 @@ void rb_ary_free(VALUE ary) { if (ARY_OWNS_HEAP_P(ary)) { - RB_DEBUG_COUNTER_INC(obj_ary_ptr); - ruby_sized_xfree((void *)ARY_HEAP_PTR(ary), ARY_HEAP_SIZE(ary)); + if (USE_DEBUG_COUNTER && + !ARY_SHARED_ROOT_P(ary) && + ARY_HEAP_CAPA(ary) > RARRAY_LEN(ary)) { + RB_DEBUG_COUNTER_INC(obj_ary_extracapa); + } + + if (RARRAY_TRANSIENT_P(ary)) { + RB_DEBUG_COUNTER_INC(obj_ary_transient); + } + else { + RB_DEBUG_COUNTER_INC(obj_ary_ptr); + ary_heap_free(ary); + } } else { - RB_DEBUG_COUNTER_INC(obj_ary_embed); + RB_DEBUG_COUNTER_INC(obj_ary_embed); + } + + if (ARY_SHARED_P(ary)) { + RB_DEBUG_COUNTER_INC(obj_ary_shared); + } + if (ARY_SHARED_ROOT_P(ary) && ARY_SHARED_ROOT_OCCUPIED(ary)) { + RB_DEBUG_COUNTER_INC(obj_ary_shared_root_occupied); } } @@ -572,51 +1028,69 @@ RUBY_FUNC_EXPORTED size_t rb_ary_memsize(VALUE ary) { if (ARY_OWNS_HEAP_P(ary)) { - return ARY_CAPA(ary) * sizeof(VALUE); + return ARY_CAPA(ary) * sizeof(VALUE); } else { - return 0; + return 0; } } -static inline void -ary_discard(VALUE ary) -{ - rb_ary_free(ary); - RBASIC(ary)->flags |= RARRAY_EMBED_FLAG; - RBASIC(ary)->flags &= ~RARRAY_EMBED_LEN_MASK; -} - static VALUE ary_make_shared(VALUE ary) { - assert(!ARY_EMBED_P(ary)); + assert(USE_RVARGC || !ARY_EMBED_P(ary)); + ary_verify(ary); + if (ARY_SHARED_P(ary)) { - return ARY_SHARED(ary); + return ARY_SHARED_ROOT(ary); } else if (ARY_SHARED_ROOT_P(ary)) { - return ary; + return ary; } else if (OBJ_FROZEN(ary)) { - ary_shrink_capa(ary); - FL_SET_SHARED_ROOT(ary); - ARY_SET_SHARED_NUM(ary, 1); - return ary; + if (!ARY_EMBED_P(ary)) { + rb_ary_transient_heap_evacuate(ary, TRUE); + ary_shrink_capa(ary); + } + return ary; } else { - long capa = ARY_CAPA(ary), len = RARRAY_LEN(ary); - NEWOBJ_OF(shared, struct RArray, 0, T_ARRAY | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0)); - FL_UNSET_EMBED(shared); + rb_ary_transient_heap_evacuate(ary, TRUE); + + long capa = ARY_CAPA(ary); + long len = RARRAY_LEN(ary); + + /* Shared roots cannot be embedded because the reference count + * (refcnt) is stored in as.heap.aux.capa. */ + VALUE shared = ary_alloc_heap(0); + FL_SET_SHARED_ROOT(shared); + + if (ARY_EMBED_P(ary)) { + /* Cannot use ary_heap_alloc because we don't want to allocate + * on the transient heap. */ + VALUE *ptr = ALLOC_N(VALUE, capa); + ARY_SET_PTR(shared, ptr); + ary_memcpy(shared, 0, len, RARRAY_PTR(ary)); + + FL_UNSET_EMBED(ary); + ARY_SET_HEAP_LEN(ary, len); + ARY_SET_PTR(ary, ptr); + } + else { + ARY_SET_PTR(shared, RARRAY_PTR(ary)); + } + + ARY_SET_LEN(shared, capa); + ary_mem_clear(shared, len, capa - len); + ARY_SET_SHARED_ROOT_REFCNT(shared, 1); + FL_SET_SHARED(ary); + RB_DEBUG_COUNTER_INC(obj_ary_shared_create); + ARY_SET_SHARED(ary, shared); - ARY_SET_LEN((VALUE)shared, capa); - ARY_SET_PTR((VALUE)shared, RARRAY_CONST_PTR(ary)); - ary_mem_clear((VALUE)shared, len, capa - len); - FL_SET_SHARED_ROOT(shared); - ARY_SET_SHARED_NUM((VALUE)shared, 1); - FL_SET_SHARED(ary); - ARY_SET_SHARED(ary, (VALUE)shared); - OBJ_FREEZE(shared); - return (VALUE)shared; + ary_verify(shared); + ary_verify(ary); + + return shared; } } @@ -625,9 +1099,11 @@ ary_make_substitution(VALUE ary) { long len = RARRAY_LEN(ary); - if (len <= RARRAY_EMBED_LEN_MAX) { - VALUE subst = rb_ary_new2(len); - ary_memcpy(subst, 0, len, RARRAY_CONST_PTR(ary)); + if (ary_embeddable_p(len)) { + VALUE subst = rb_ary_new_capa(len); + assert(ARY_EMBED_P(subst)); + + ary_memcpy(subst, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary)); ARY_SET_EMBED_LEN(subst, len); return subst; } @@ -655,23 +1131,30 @@ rb_check_array_type(VALUE ary) return rb_check_convert_type_with_id(ary, T_ARRAY, "Array", idTo_ary); } +MJIT_FUNC_EXPORTED VALUE +rb_check_to_array(VALUE ary) +{ + return rb_check_convert_type_with_id(ary, T_ARRAY, "Array", idTo_a); +} + +VALUE +rb_to_array(VALUE ary) +{ + return rb_convert_type_with_id(ary, T_ARRAY, "Array", idTo_a); +} + /* * call-seq: - * Array.try_convert(obj) -> array or nil + * Array.try_convert(object) -> object, new_array, or nil * - * Tries to convert +obj+ into an array, using +to_ary+ method. Returns the - * converted array or +nil+ if +obj+ cannot be converted for any reason. - * This method can be used to check if an argument is an array. + * If +object+ is an \Array object, returns +object+. * - * Array.try_convert([1]) #=> [1] - * Array.try_convert("1") #=> nil + * Otherwise if +object+ responds to <tt>:to_ary</tt>, + * calls <tt>object.to_ary</tt> and returns the result. * - * if tmp = Array.try_convert(arg) - * # the argument is an array - * elsif tmp = String.try_convert(arg) - * # the argument is a string - * end + * Returns +nil+ if +object+ does not respond to <tt>:to_ary</tt> * + * Raises an exception unless <tt>object.to_ary</tt> returns an \Array object. */ static VALUE @@ -680,60 +1163,76 @@ rb_ary_s_try_convert(VALUE dummy, VALUE ary) return rb_check_array_type(ary); } +/* :nodoc: */ +static VALUE +rb_ary_s_new(int argc, VALUE *argv, VALUE klass) +{ + VALUE ary; + + if (klass == rb_cArray) { + long size = 0; + if (argc > 0 && FIXNUM_P(argv[0])) { + size = FIX2LONG(argv[0]); + if (size < 0) size = 0; + } + + ary = ary_new(klass, size); + + rb_obj_call_init_kw(ary, argc, argv, RB_PASS_CALLED_KEYWORDS); + } + else { + ary = rb_class_new_instance_pass_kw(argc, argv, klass); + } + + return ary; +} + /* * call-seq: - * Array.new(size=0, default=nil) - * Array.new(array) - * Array.new(size) {|index| block } - * - * Returns a new array. - * - * In the first form, if no arguments are sent, the new array will be empty. - * When a +size+ and an optional +default+ are sent, an array is created with - * +size+ copies of +default+. Take notice that all elements will reference the - * same object +default+. + * Array.new -> new_empty_array + * Array.new(array) -> new_array + * Array.new(size) -> new_array + * Array.new(size, default_value) -> new_array + * Array.new(size) {|index| ... } -> new_array * - * The second form creates a copy of the array passed as a parameter (the - * array is generated by calling to_ary on the parameter). + * Returns a new \Array. * - * first_array = ["Matz", "Guido"] + * With no block and no arguments, returns a new empty \Array object. * - * second_array = Array.new(first_array) #=> ["Matz", "Guido"] + * With no block and a single \Array argument +array+, + * returns a new \Array formed from +array+: * - * first_array.equal? second_array #=> false + * a = Array.new([:foo, 'bar', 2]) + * a.class # => Array + * a # => [:foo, "bar", 2] * - * In the last form, an array of the given size is created. Each element in - * this array is created by passing the element's index to the given block - * and storing the return value. + * With no block and a single \Integer argument +size+, + * returns a new \Array of the given size + * whose elements are all +nil+: * - * Array.new(3){ |index| index ** 2 } - * # => [0, 1, 4] + * a = Array.new(3) + * a # => [nil, nil, nil] * - * == Common gotchas + * With no block and arguments +size+ and +default_value+, + * returns an \Array of the given size; + * each element is that same +default_value+: * - * When sending the second parameter, the same object will be used as the - * value for all the array elements: + * a = Array.new(3, 'x') + * a # => ['x', 'x', 'x'] * - * a = Array.new(2, Hash.new) - * # => [{}, {}] + * With a block and argument +size+, + * returns an \Array of the given size; + * the block is called with each successive integer +index+; + * the element for that +index+ is the return value from the block: * - * a[0]['cat'] = 'feline' - * a # => [{"cat"=>"feline"}, {"cat"=>"feline"}] + * a = Array.new(3) {|index| "Element #{index}" } + * a # => ["Element 0", "Element 1", "Element 2"] * - * a[1]['cat'] = 'Felix' - * a # => [{"cat"=>"Felix"}, {"cat"=>"Felix"}] - * - * Since all the Array elements store the same hash, changes to one of them - * will affect them all. - * - * If multiple copies are what you want, you should use the block - * version which uses the result of that block each time an element - * of the array needs to be initialized: - * - * a = Array.new(2) { Hash.new } - * a[0]['cat'] = 'feline' - * a # => [{"cat"=>"feline"}, {}] + * Raises ArgumentError if +size+ is negative. * + * With a block and no argument, + * or a single argument +0+, + * ignores the block and returns a new empty \Array. */ static VALUE @@ -744,51 +1243,48 @@ rb_ary_initialize(int argc, VALUE *argv, VALUE ary) rb_ary_modify(ary); if (argc == 0) { - if (ARY_OWNS_HEAP_P(ary) && RARRAY_CONST_PTR(ary) != 0) { - ruby_sized_xfree((void *)RARRAY_CONST_PTR(ary), ARY_HEAP_SIZE(ary)); - } - rb_ary_unshare_safe(ary); - FL_SET_EMBED(ary); - ARY_SET_EMBED_LEN(ary, 0); - if (rb_block_given_p()) { - rb_warning("given block not used"); - } - return ary; + rb_ary_reset(ary); + assert(ARY_EMBED_P(ary)); + assert(ARY_EMBED_LEN(ary) == 0); + if (rb_block_given_p()) { + rb_warning("given block not used"); + } + return ary; } rb_scan_args(argc, argv, "02", &size, &val); if (argc == 1 && !FIXNUM_P(size)) { - val = rb_check_array_type(size); - if (!NIL_P(val)) { - rb_ary_replace(ary, val); - return ary; - } + val = rb_check_array_type(size); + if (!NIL_P(val)) { + rb_ary_replace(ary, val); + return ary; + } } len = NUM2LONG(size); /* NUM2LONG() may call size.to_int, ary can be frozen, modified, etc */ if (len < 0) { - rb_raise(rb_eArgError, "negative array size"); + rb_raise(rb_eArgError, "negative array size"); } if (len > ARY_MAX_SIZE) { - rb_raise(rb_eArgError, "array size too big"); + rb_raise(rb_eArgError, "array size too big"); } /* recheck after argument conversion */ rb_ary_modify(ary); ary_resize_capa(ary, len); if (rb_block_given_p()) { - long i; + long i; - if (argc == 2) { - rb_warn("block supersedes default value argument"); - } - for (i=0; i<len; i++) { - rb_ary_store(ary, i, rb_yield(LONG2NUM(i))); - ARY_SET_LEN(ary, i + 1); - } + if (argc == 2) { + rb_warn("block supersedes default value argument"); + } + for (i=0; i<len; i++) { + rb_ary_store(ary, i, rb_yield(LONG2NUM(i))); + ARY_SET_LEN(ary, i + 1); + } } else { - ary_memfill(ary, 0, len, val); - ARY_SET_LEN(ary, len); + ary_memfill(ary, 0, len, val); + ARY_SET_LEN(ary, len); } return ary; } @@ -796,7 +1292,7 @@ rb_ary_initialize(int argc, VALUE *argv, VALUE ary) /* * Returns a new array populated with the given objects. * - * Array.[]( 1, 'a', /^A/ ) # => [1, "a", /^A/] + * Array.[]( 1, 'a', /^A/) # => [1, "a", /^A/] * Array[ 1, 'a', /^A/ ] # => [1, "a", /^A/] * [ 1, 'a', /^A/ ] # => [1, "a", /^A/] */ @@ -819,26 +1315,26 @@ rb_ary_store(VALUE ary, long idx, VALUE val) long len = RARRAY_LEN(ary); if (idx < 0) { - idx += len; - if (idx < 0) { - rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", - idx - len, -len); - } + idx += len; + if (idx < 0) { + rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", + idx - len, -len); + } } else if (idx >= ARY_MAX_SIZE) { - rb_raise(rb_eIndexError, "index %ld too big", idx); + rb_raise(rb_eIndexError, "index %ld too big", idx); } rb_ary_modify(ary); if (idx >= ARY_CAPA(ary)) { - ary_double_capa(ary, idx); + ary_double_capa(ary, idx); } if (idx > len) { - ary_mem_clear(ary, len, idx - len + 1); + ary_mem_clear(ary, len, idx - len + 1); } if (idx >= len) { - ARY_SET_LEN(ary, idx + 1); + ARY_SET_LEN(ary, idx + 1); } ARY_SET(ary, idx, val); } @@ -850,31 +1346,87 @@ ary_make_partial(VALUE ary, VALUE klass, long offset, long len) assert(len >= 0); assert(offset+len <= RARRAY_LEN(ary)); - if (len <= RARRAY_EMBED_LEN_MAX) { - VALUE result = ary_alloc(klass); - ary_memcpy(result, 0, len, RARRAY_CONST_PTR(ary) + offset); + const size_t rarray_embed_capa_max = (sizeof(struct RArray) - offsetof(struct RArray, as.ary)) / sizeof(VALUE); + + if ((size_t)len <= rarray_embed_capa_max && ary_embeddable_p(len)) { + VALUE result = ary_alloc_embed(klass, len); + ary_memcpy(result, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary) + offset); ARY_SET_EMBED_LEN(result, len); return result; } else { - VALUE shared, result = ary_alloc(klass); - FL_UNSET_EMBED(result); + VALUE shared = ary_make_shared(ary); + + VALUE result = ary_alloc_heap(klass); + assert(!ARY_EMBED_P(result)); - shared = ary_make_shared(ary); - ARY_SET_PTR(result, RARRAY_CONST_PTR(ary)); + ARY_SET_PTR(result, RARRAY_CONST_PTR_TRANSIENT(ary)); ARY_SET_LEN(result, RARRAY_LEN(ary)); rb_ary_set_shared(result, shared); ARY_INCREASE_PTR(result, offset); ARY_SET_LEN(result, len); + + ary_verify(shared); + ary_verify(result); return result; } } static VALUE +ary_make_partial_step(VALUE ary, VALUE klass, long offset, long len, long step) +{ + assert(offset >= 0); + assert(len >= 0); + assert(offset+len <= RARRAY_LEN(ary)); + assert(step != 0); + + const VALUE *values = RARRAY_CONST_PTR_TRANSIENT(ary); + const long orig_len = len; + + if (step > 0 && step >= len) { + VALUE result = ary_new(klass, 1); + VALUE *ptr = (VALUE *)ARY_EMBED_PTR(result); + RB_OBJ_WRITE(result, ptr, values[offset]); + ARY_SET_EMBED_LEN(result, 1); + return result; + } + else if (step < 0 && step < -len) { + step = -len; + } + + long ustep = (step < 0) ? -step : step; + len = roomof(len, ustep); + + long i; + long j = offset + ((step > 0) ? 0 : (orig_len - 1)); + + VALUE result = ary_new(klass, len); + if (ARY_EMBED_P(result)) { + VALUE *ptr = (VALUE *)ARY_EMBED_PTR(result); + for (i = 0; i < len; ++i) { + RB_OBJ_WRITE(result, ptr+i, values[j]); + j += step; + } + ARY_SET_EMBED_LEN(result, len); + } + else { + RARRAY_PTR_USE_TRANSIENT(result, ptr, { + for (i = 0; i < len; ++i) { + RB_OBJ_WRITE(result, ptr+i, values[j]); + j += step; + } + }); + ARY_SET_LEN(result, len); + } + + return result; +} + +static VALUE ary_make_shared_copy(VALUE ary) { - return ary_make_partial(ary, rb_obj_class(ary), 0, RARRAY_LEN(ary)); + return ary_make_partial(ary, rb_cArray, 0, RARRAY_LEN(ary)); } enum ary_take_pos_flags @@ -886,51 +1438,57 @@ enum ary_take_pos_flags static VALUE ary_take_first_or_last(int argc, const VALUE *argv, VALUE ary, enum ary_take_pos_flags last) { - VALUE nv; long n; long len; long offset = 0; - rb_scan_args(argc, argv, "1", &nv); - n = NUM2LONG(nv); + argc = rb_check_arity(argc, 0, 1); + /* the case optional argument is omitted should be handled in + * callers of this function. if another arity case is added, + * this arity check needs to rewrite. */ + RUBY_ASSERT_ALWAYS(argc == 1); + + n = NUM2LONG(argv[0]); len = RARRAY_LEN(ary); if (n > len) { - n = len; + n = len; } else if (n < 0) { - rb_raise(rb_eArgError, "negative array size"); + rb_raise(rb_eArgError, "negative array size"); } if (last) { - offset = len - n; + offset = len - n; } return ary_make_partial(ary, rb_cArray, offset, n); } /* * call-seq: - * ary << obj -> ary + * array << object -> self + * + * Appends +object+ to +self+; returns +self+: * - * Append---Pushes the given object on to the end of this array. This - * expression returns the array itself, so several appends - * may be chained together. + * a = [:foo, 'bar', 2] + * a << :baz # => [:foo, "bar", 2, :baz] * - * a = [ 1, 2 ] - * a << "c" << "d" << [ 3, 4 ] - * #=> [ 1, 2, "c", "d", [ 3, 4 ] ] - * a - * #=> [ 1, 2, "c", "d", [ 3, 4 ] ] + * Appends +object+ as one element, even if it is another \Array: + * + * a = [:foo, 'bar', 2] + * a1 = a << [3, 4] + * a1 # => [:foo, "bar", 2, [3, 4]] * */ VALUE rb_ary_push(VALUE ary, VALUE item) { - long idx = RARRAY_LEN(ary); + long idx = RARRAY_LEN((ary_verify(ary), ary)); VALUE target_ary = ary_ensure_room_for_push(ary, 1); - RARRAY_PTR_USE(ary, ptr, { - RB_OBJ_WRITE(target_ary, &ptr[idx], item); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + RB_OBJ_WRITE(target_ary, &ptr[idx], item); }); ARY_SET_LEN(ary, idx + 1); + ary_verify(ary); return ary; } @@ -946,19 +1504,24 @@ rb_ary_cat(VALUE ary, const VALUE *argv, long len) /* * call-seq: - * ary.push(obj, ... ) -> ary - * ary.append(obj, ... ) -> ary - * - * Append --- Pushes the given object(s) on to the end of this array. This - * expression returns the array itself, so several appends - * may be chained together. See also Array#pop for the opposite - * effect. - * - * a = [ "a", "b", "c" ] - * a.push("d", "e", "f") - * #=> ["a", "b", "c", "d", "e", "f"] - * [1, 2, 3].push(4).push(5) - * #=> [1, 2, 3, 4, 5] + * array.push(*objects) -> self + * + * Appends trailing elements. + * + * Appends each argument in +objects+ to +self+; returns +self+: + * + * a = [:foo, 'bar', 2] + * a.push(:baz, :bat) # => [:foo, "bar", 2, :baz, :bat] + * + * Appends each argument as one element, even if it is another \Array: + * + * a = [:foo, 'bar', 2] + * a1 = a.push([:baz, :bat], [:bam, :bad]) + * a1 # => [:foo, "bar", 2, [:baz, :bat], [:bam, :bad]] + * + * Array#append is an alias for Array#push. + * + * Related: #pop, #shift, #unshift. */ static VALUE @@ -975,32 +1538,46 @@ rb_ary_pop(VALUE ary) n = RARRAY_LEN(ary); if (n == 0) return Qnil; if (ARY_OWNS_HEAP_P(ary) && - n * 3 < ARY_CAPA(ary) && - ARY_CAPA(ary) > ARY_DEFAULT_SIZE) + n * 3 < ARY_CAPA(ary) && + ARY_CAPA(ary) > ARY_DEFAULT_SIZE) { - ary_resize_capa(ary, n * 2); + ary_resize_capa(ary, n * 2); } --n; ARY_SET_LEN(ary, n); + ary_verify(ary); return RARRAY_AREF(ary, n); } /* * call-seq: - * ary.pop -> obj or nil - * ary.pop(n) -> new_ary + * array.pop -> object or nil + * array.pop(n) -> new_array + * + * Removes and returns trailing elements. + * + * When no argument is given and +self+ is not empty, + * removes and returns the last element: * - * Removes the last element from +self+ and returns it, or - * +nil+ if the array is empty. + * a = [:foo, 'bar', 2] + * a.pop # => 2 + * a # => [:foo, "bar"] * - * If a number +n+ is given, returns an array of the last +n+ elements - * (or less) just like <code>array.slice!(-n, n)</code> does. See also - * Array#push for the opposite effect. + * Returns +nil+ if the array is empty. * - * a = [ "a", "b", "c", "d" ] - * a.pop #=> "d" - * a.pop(2) #=> ["b", "c"] - * a #=> ["a"] + * When a non-negative \Integer argument +n+ is given and is in range, + * + * removes and returns the last +n+ elements in a new \Array: + * a = [:foo, 'bar', 2] + * a.pop(2) # => ["bar", 2] + * + * If +n+ is positive and out of range, + * removes and returns all elements: + * + * a = [:foo, 'bar', 2] + * a.pop(50) # => [:foo, "bar", 2] + * + * Related: #push, #shift, #unshift. */ static VALUE @@ -1009,12 +1586,13 @@ rb_ary_pop_m(int argc, VALUE *argv, VALUE ary) VALUE result; if (argc == 0) { - return rb_ary_pop(ary); + return rb_ary_pop(ary); } rb_ary_modify_check(ary); result = ary_take_first_or_last(argc, argv, ary, ARY_TAKE_LAST); ARY_INCREASE_LEN(ary, -RARRAY_LEN(result)); + ary_verify(ary); return result; } @@ -1024,52 +1602,49 @@ rb_ary_shift(VALUE ary) VALUE top; long len = RARRAY_LEN(ary); - rb_ary_modify_check(ary); - if (len == 0) return Qnil; + if (len == 0) { + rb_ary_modify_check(ary); + return Qnil; + } + top = RARRAY_AREF(ary, 0); - if (!ARY_SHARED_P(ary)) { - if (len < ARY_DEFAULT_SIZE) { - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr, ptr+1, VALUE, len-1); - }); /* WB: no new reference */ - ARY_INCREASE_LEN(ary, -1); - return top; - } - assert(!ARY_EMBED_P(ary)); /* ARY_EMBED_LEN_MAX < ARY_DEFAULT_SIZE */ - - ARY_SET(ary, 0, Qnil); - ary_make_shared(ary); - } - else if (ARY_SHARED_OCCUPIED(ARY_SHARED(ary))) { - RARRAY_PTR_USE(ary, ptr, ptr[0] = Qnil); - } - ARY_INCREASE_PTR(ary, 1); /* shift ptr */ - ARY_INCREASE_LEN(ary, -1); + + rb_ary_behead(ary, 1); return top; } /* * call-seq: - * ary.shift -> obj or nil - * ary.shift(n) -> new_ary + * array.shift -> object or nil + * array.shift(n) -> new_array * - * Removes the first element of +self+ and returns it (shifting all - * other elements down by one). Returns +nil+ if the array - * is empty. + * Removes and returns leading elements. * - * If a number +n+ is given, returns an array of the first +n+ elements - * (or less) just like <code>array.slice!(0, n)</code> does. With +ary+ - * containing only the remainder elements, not including what was shifted to - * +new_ary+. See also Array#unshift for the opposite effect. + * When no argument is given, removes and returns the first element: * - * args = [ "-m", "-q", "filename" ] - * args.shift #=> "-m" - * args #=> ["-q", "filename"] + * a = [:foo, 'bar', 2] + * a.shift # => :foo + * a # => ['bar', 2] * - * args = [ "-m", "-q", "filename" ] - * args.shift(2) #=> ["-m", "-q"] - * args #=> ["filename"] + * Returns +nil+ if +self+ is empty. + * + * When positive \Integer argument +n+ is given, removes the first +n+ elements; + * returns those elements in a new \Array: + * + * a = [:foo, 'bar', 2] + * a.shift(2) # => [:foo, 'bar'] + * a # => [2] + * + * If +n+ is as large as or larger than <tt>self.length</tt>, + * removes all elements; returns those elements in a new \Array: + * + * a = [:foo, 'bar', 2] + * a.shift(3) # => [:foo, 'bar', 2] + * + * If +n+ is zero, returns a new empty \Array; +self+ is unmodified. + * + * Related: #push, #pop, #unshift. */ static VALUE @@ -1079,105 +1654,147 @@ rb_ary_shift_m(int argc, VALUE *argv, VALUE ary) long n; if (argc == 0) { - return rb_ary_shift(ary); + return rb_ary_shift(ary); } rb_ary_modify_check(ary); result = ary_take_first_or_last(argc, argv, ary, ARY_TAKE_FIRST); n = RARRAY_LEN(result); - if (ARY_SHARED_P(ary)) { - if (ARY_SHARED_OCCUPIED(ARY_SHARED(ary))) { - setup_occupied_shared: - ary_mem_clear(ary, 0, n); - } - ARY_INCREASE_PTR(ary, n); + rb_ary_behead(ary,n); + + return result; +} + +MJIT_FUNC_EXPORTED VALUE +rb_ary_behead(VALUE ary, long n) +{ + if (n <= 0) { + return ary; } - else { - if (RARRAY_LEN(ary) < ARY_DEFAULT_SIZE) { - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr, ptr+n, VALUE, RARRAY_LEN(ary)-n); - }); /* WB: no new reference */ - } - else { - ary_make_shared(ary); - goto setup_occupied_shared; - } + + rb_ary_modify_check(ary); + + if (!ARY_SHARED_P(ary)) { + if (ARY_EMBED_P(ary) || RARRAY_LEN(ary) < ARY_DEFAULT_SIZE) { + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMMOVE(ptr, ptr + n, VALUE, RARRAY_LEN(ary) - n); + }); /* WB: no new reference */ + ARY_INCREASE_LEN(ary, -n); + ary_verify(ary); + return ary; + } + + ary_mem_clear(ary, 0, n); + ary_make_shared(ary); + } + else if (ARY_SHARED_ROOT_OCCUPIED(ARY_SHARED_ROOT(ary))) { + ary_mem_clear(ary, 0, n); } + + ARY_INCREASE_PTR(ary, n); ARY_INCREASE_LEN(ary, -n); + ary_verify(ary); - return result; + return ary; } static VALUE -ary_ensure_room_for_unshift(VALUE ary, int argc) +make_room_for_unshift(VALUE ary, const VALUE *head, VALUE *sharedp, int argc, long capa, long len) +{ + if (head - sharedp < argc) { + long room = capa - len - argc; + + room -= room >> 4; + MEMMOVE((VALUE *)sharedp + argc + room, head, VALUE, len); + head = sharedp + argc + room; + } + ARY_SET_PTR(ary, head - argc); + assert(ARY_SHARED_ROOT_OCCUPIED(ARY_SHARED_ROOT(ary))); + + ary_verify(ary); + return ARY_SHARED_ROOT(ary); +} + +static VALUE +ary_modify_for_unshift(VALUE ary, int argc) { long len = RARRAY_LEN(ary); long new_len = len + argc; long capa; const VALUE *head, *sharedp; - if (len > ARY_MAX_SIZE - argc) { - rb_raise(rb_eIndexError, "index %ld too big", new_len); - } - - if (ARY_SHARED_P(ary)) { - VALUE shared = ARY_SHARED(ary); - capa = RARRAY_LEN(shared); - if (ARY_SHARED_OCCUPIED(shared) && capa > new_len) { - head = RARRAY_CONST_PTR(ary); - sharedp = RARRAY_CONST_PTR(shared); - goto makeroom_if_need; - } - } - rb_ary_modify(ary); capa = ARY_CAPA(ary); if (capa - (capa >> 6) <= new_len) { - ary_double_capa(ary, new_len); + ary_double_capa(ary, new_len); } /* use shared array for big "queues" */ - if (new_len > ARY_DEFAULT_SIZE * 4) { - /* make a room for unshifted items */ - capa = ARY_CAPA(ary); - ary_make_shared(ary); - - head = sharedp = RARRAY_CONST_PTR(ary); - goto makeroom; - makeroom_if_need: - if (head - sharedp < argc) { - long room; - makeroom: - room = capa - new_len; - room -= room >> 4; - MEMMOVE((VALUE *)sharedp + argc + room, head, VALUE, len); - head = sharedp + argc + room; - } - ARY_SET_PTR(ary, head - argc); - assert(ARY_SHARED_OCCUPIED(ARY_SHARED(ary))); - return ARY_SHARED(ary); + if (new_len > ARY_DEFAULT_SIZE * 4 && !ARY_EMBED_P(ary)) { + ary_verify(ary); + + /* make a room for unshifted items */ + capa = ARY_CAPA(ary); + ary_make_shared(ary); + + head = sharedp = RARRAY_CONST_PTR_TRANSIENT(ary); + return make_room_for_unshift(ary, head, (void *)sharedp, argc, capa, len); + } + else { + /* sliding items */ + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMMOVE(ptr + argc, ptr, VALUE, len); + }); + + ary_verify(ary); + return ary; + } +} + +static VALUE +ary_ensure_room_for_unshift(VALUE ary, int argc) +{ + long len = RARRAY_LEN(ary); + long new_len = len + argc; + + if (len > ARY_MAX_SIZE - argc) { + rb_raise(rb_eIndexError, "index %ld too big", new_len); + } + else if (! ARY_SHARED_P(ary)) { + return ary_modify_for_unshift(ary, argc); } else { - /* sliding items */ - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr + argc, ptr, VALUE, len); - }); + VALUE shared_root = ARY_SHARED_ROOT(ary); + long capa = RARRAY_LEN(shared_root); - return ary; + if (! ARY_SHARED_ROOT_OCCUPIED(shared_root)) { + return ary_modify_for_unshift(ary, argc); + } + else if (new_len > capa) { + return ary_modify_for_unshift(ary, argc); + } + else { + const VALUE * head = RARRAY_CONST_PTR_TRANSIENT(ary); + void *sharedp = (void *)RARRAY_CONST_PTR_TRANSIENT(shared_root); + + rb_ary_modify_check(ary); + return make_room_for_unshift(ary, head, sharedp, argc, capa, len); + } } } /* * call-seq: - * ary.unshift(obj, ...) -> ary - * ary.prepend(obj, ...) -> ary + * array.unshift(*objects) -> self * - * Prepends objects to the front of +self+, moving other elements upwards. - * See also Array#shift for the opposite effect. + * Prepends the given +objects+ to +self+: * - * a = [ "b", "c", "d" ] - * a.unshift("a") #=> ["a", "b", "c", "d"] - * a.unshift(1, 2) #=> [ 1, 2, "a", "b", "c", "d"] + * a = [:foo, 'bar', 2] + * a.unshift(:bam, :bat) # => [:bam, :bat, :foo, "bar", 2] + * + * Array#prepend is an alias for Array#unshift. + * + * Related: #push, #pop, #shift. */ static VALUE @@ -1187,8 +1804,8 @@ rb_ary_unshift_m(int argc, VALUE *argv, VALUE ary) VALUE target_ary; if (argc == 0) { - rb_ary_modify_check(ary); - return ary; + rb_ary_modify_check(ary); + return ary; } target_ary = ary_ensure_room_for_unshift(ary, argc); @@ -1210,7 +1827,7 @@ rb_ary_elt(VALUE ary, long offset) long len = RARRAY_LEN(ary); if (len == 0) return Qnil; if (offset < 0 || len <= offset) { - return Qnil; + return Qnil; } return RARRAY_AREF(ary, offset); } @@ -1218,21 +1835,11 @@ rb_ary_elt(VALUE ary, long offset) VALUE rb_ary_entry(VALUE ary, long offset) { - long len = RARRAY_LEN(ary); - const VALUE *ptr = RARRAY_CONST_PTR(ary); - if (len == 0) return Qnil; - if (offset < 0) { - offset += len; - if (offset < 0) return Qnil; - } - else if (len <= offset) { - return Qnil; - } - return ptr[offset]; + return rb_ary_entry_internal(ary, offset); } VALUE -rb_ary_subseq(VALUE ary, long beg, long len) +rb_ary_subseq_step(VALUE ary, long beg, long len, long step) { VALUE klass; long alen = RARRAY_LEN(ary); @@ -1241,48 +1848,129 @@ rb_ary_subseq(VALUE ary, long beg, long len) if (beg < 0 || len < 0) return Qnil; if (alen < len || alen < beg + len) { - len = alen - beg; + len = alen - beg; } - klass = rb_obj_class(ary); + klass = rb_cArray; if (len == 0) return ary_new(klass, 0); + if (step == 0) + rb_raise(rb_eArgError, "slice step cannot be zero"); + if (step == 1) + return ary_make_partial(ary, klass, beg, len); + else + return ary_make_partial_step(ary, klass, beg, len, step); +} - return ary_make_partial(ary, klass, beg, len); +VALUE +rb_ary_subseq(VALUE ary, long beg, long len) +{ + return rb_ary_subseq_step(ary, beg, len, 1); } +static VALUE rb_ary_aref2(VALUE ary, VALUE b, VALUE e); + /* * call-seq: - * ary[index] -> obj or nil - * ary[start, length] -> new_ary or nil - * ary[range] -> new_ary or nil - * ary.slice(index) -> obj or nil - * ary.slice(start, length) -> new_ary or nil - * ary.slice(range) -> new_ary or nil - * - * Element Reference --- Returns the element at +index+, or returns a - * subarray starting at the +start+ index and continuing for +length+ - * elements, or returns a subarray specified by +range+ of indices. - * - * Negative indices count backward from the end of the array (-1 is the last - * element). For +start+ and +range+ cases the starting index is just before - * an element. Additionally, an empty array is returned when the starting - * index for an element range is at the end of the array. - * - * Returns +nil+ if the index (or starting index) are out of range. - * - * a = [ "a", "b", "c", "d", "e" ] - * a[2] + a[0] + a[1] #=> "cab" - * a[6] #=> nil - * a[1, 2] #=> [ "b", "c" ] - * a[1..3] #=> [ "b", "c", "d" ] - * a[4..7] #=> [ "e" ] - * a[6..10] #=> nil - * a[-3, 3] #=> [ "c", "d", "e" ] - * # special cases - * a[5] #=> nil - * a[6, 1] #=> nil - * a[5, 1] #=> [] - * a[5..10] #=> [] + * array[index] -> object or nil + * array[start, length] -> object or nil + * array[range] -> object or nil + * array[aseq] -> object or nil + * array.slice(index) -> object or nil + * array.slice(start, length) -> object or nil + * array.slice(range) -> object or nil + * array.slice(aseq) -> object or nil * + * Returns elements from +self+; does not modify +self+. + * + * When a single \Integer argument +index+ is given, returns the element at offset +index+: + * + * a = [:foo, 'bar', 2] + * a[0] # => :foo + * a[2] # => 2 + * a # => [:foo, "bar", 2] + * + * If +index+ is negative, counts relative to the end of +self+: + * + * a = [:foo, 'bar', 2] + * a[-1] # => 2 + * a[-2] # => "bar" + * + * If +index+ is out of range, returns +nil+. + * + * When two \Integer arguments +start+ and +length+ are given, + * returns a new \Array of size +length+ containing successive elements beginning at offset +start+: + * + * a = [:foo, 'bar', 2] + * a[0, 2] # => [:foo, "bar"] + * a[1, 2] # => ["bar", 2] + * + * If <tt>start + length</tt> is greater than <tt>self.length</tt>, + * returns all elements from offset +start+ to the end: + * + * a = [:foo, 'bar', 2] + * a[0, 4] # => [:foo, "bar", 2] + * a[1, 3] # => ["bar", 2] + * a[2, 2] # => [2] + * + * If <tt>start == self.size</tt> and <tt>length >= 0</tt>, + * returns a new empty \Array. + * + * If +length+ is negative, returns +nil+. + * + * When a single \Range argument +range+ is given, + * treats <tt>range.min</tt> as +start+ above + * and <tt>range.size</tt> as +length+ above: + * + * a = [:foo, 'bar', 2] + * a[0..1] # => [:foo, "bar"] + * a[1..2] # => ["bar", 2] + * + * Special case: If <tt>range.start == a.size</tt>, returns a new empty \Array. + * + * If <tt>range.end</tt> is negative, calculates the end index from the end: + * + * a = [:foo, 'bar', 2] + * a[0..-1] # => [:foo, "bar", 2] + * a[0..-2] # => [:foo, "bar"] + * a[0..-3] # => [:foo] + * + * If <tt>range.start</tt> is negative, calculates the start index from the end: + * + * a = [:foo, 'bar', 2] + * a[-1..2] # => [2] + * a[-2..2] # => ["bar", 2] + * a[-3..2] # => [:foo, "bar", 2] + * + * If <tt>range.start</tt> is larger than the array size, returns +nil+. + * + * a = [:foo, 'bar', 2] + * a[4..1] # => nil + * a[4..0] # => nil + * a[4..-1] # => nil + * + * When a single Enumerator::ArithmeticSequence argument +aseq+ is given, + * returns an \Array of elements corresponding to the indexes produced by + * the sequence. + * + * a = ['--', 'data1', '--', 'data2', '--', 'data3'] + * a[(1..).step(2)] # => ["data1", "data2", "data3"] + * + * Unlike slicing with range, if the start or the end of the arithmetic sequence + * is larger than array size, throws RangeError. + * + * a = ['--', 'data1', '--', 'data2', '--', 'data3'] + * a[(1..11).step(2)] + * # RangeError (((1..11).step(2)) out of range) + * a[(7..).step(2)] + * # RangeError (((7..).step(2)) out of range) + * + * If given a single argument, and its type is not one of the listed, tries to + * convert it to Integer, and raises if it is impossible: + * + * a = [:foo, 'bar', 2] + * # Raises TypeError (no implicit conversion of Symbol into Integer): + * a[:foo] + * + * Array#slice is an alias for Array#[]. */ VALUE @@ -1290,54 +1978,53 @@ rb_ary_aref(int argc, const VALUE *argv, VALUE ary) { rb_check_arity(argc, 1, 2); if (argc == 2) { - return rb_ary_aref2(ary, argv[0], argv[1]); + return rb_ary_aref2(ary, argv[0], argv[1]); } return rb_ary_aref1(ary, argv[0]); } -VALUE +static VALUE rb_ary_aref2(VALUE ary, VALUE b, VALUE e) { long beg = NUM2LONG(b); long len = NUM2LONG(e); if (beg < 0) { - beg += RARRAY_LEN(ary); + beg += RARRAY_LEN(ary); } return rb_ary_subseq(ary, beg, len); } -VALUE +MJIT_FUNC_EXPORTED VALUE rb_ary_aref1(VALUE ary, VALUE arg) { - long beg, len; + long beg, len, step; /* special case - speeding up */ if (FIXNUM_P(arg)) { - return rb_ary_entry(ary, FIX2LONG(arg)); + return rb_ary_entry(ary, FIX2LONG(arg)); } - /* check if idx is Range */ - switch (rb_range_beg_len(arg, &beg, &len, RARRAY_LEN(ary), 0)) { + /* check if idx is Range or ArithmeticSequence */ + switch (rb_arithmetic_sequence_beg_len_step(arg, &beg, &len, &step, RARRAY_LEN(ary), 0)) { case Qfalse: - break; + break; case Qnil: - return Qnil; + return Qnil; default: - return rb_ary_subseq(ary, beg, len); + return rb_ary_subseq_step(ary, beg, len, step); } + return rb_ary_entry(ary, NUM2LONG(arg)); } /* * call-seq: - * ary.at(index) -> obj or nil + * array.at(index) -> object * - * Returns the element at +index+. A negative index counts from the end of - * +self+. Returns +nil+ if the index is out of range. See also - * Array#[]. + * Returns the element at \Integer offset +index+; does not modify +self+. + * a = [:foo, 'bar', 2] + * a.at(0) # => :foo + * a.at(2) # => 2 * - * a = [ "a", "b", "c", "d", "e" ] - * a.at(0) #=> "a" - * a.at(-1) #=> "e" */ VALUE @@ -1348,81 +2035,131 @@ rb_ary_at(VALUE ary, VALUE pos) /* * call-seq: - * ary.first -> obj or nil - * ary.first(n) -> new_ary + * array.first -> object or nil + * array.first(n) -> new_array + * + * Returns elements from +self+; does not modify +self+. + * + * When no argument is given, returns the first element: + * + * a = [:foo, 'bar', 2] + * a.first # => :foo + * a # => [:foo, "bar", 2] + * + * If +self+ is empty, returns +nil+. + * + * When non-negative \Integer argument +n+ is given, + * returns the first +n+ elements in a new \Array: * - * Returns the first element, or the first +n+ elements, of the array. - * If the array is empty, the first form returns +nil+, and the - * second form returns an empty array. See also Array#last for - * the opposite effect. + * a = [:foo, 'bar', 2] + * a.first(2) # => [:foo, "bar"] * - * a = [ "q", "r", "s", "t" ] - * a.first #=> "q" - * a.first(2) #=> ["q", "r"] + * If <tt>n >= array.size</tt>, returns all elements: + * + * a = [:foo, 'bar', 2] + * a.first(50) # => [:foo, "bar", 2] + * + * If <tt>n == 0</tt> returns an new empty \Array: + * + * a = [:foo, 'bar', 2] + * a.first(0) # [] + * + * Related: #last. */ - static VALUE rb_ary_first(int argc, VALUE *argv, VALUE ary) { if (argc == 0) { - if (RARRAY_LEN(ary) == 0) return Qnil; - return RARRAY_AREF(ary, 0); + if (RARRAY_LEN(ary) == 0) return Qnil; + return RARRAY_AREF(ary, 0); } else { - return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_FIRST); + return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_FIRST); } } /* * call-seq: - * ary.last -> obj or nil - * ary.last(n) -> new_ary + * array.last -> object or nil + * array.last(n) -> new_array + * + * Returns elements from +self+; +self+ is not modified. + * + * When no argument is given, returns the last element: + * + * a = [:foo, 'bar', 2] + * a.last # => 2 + * a # => [:foo, "bar", 2] + * + * If +self+ is empty, returns +nil+. + * + * When non-negative \Integer argument +n+ is given, + * returns the last +n+ elements in a new \Array: + * + * a = [:foo, 'bar', 2] + * a.last(2) # => ["bar", 2] + * + * If <tt>n >= array.size</tt>, returns all elements: * - * Returns the last element(s) of +self+. If the array is empty, - * the first form returns +nil+. + * a = [:foo, 'bar', 2] + * a.last(50) # => [:foo, "bar", 2] * - * See also Array#first for the opposite effect. + * If <tt>n == 0</tt>, returns an new empty \Array: * - * a = [ "w", "x", "y", "z" ] - * a.last #=> "z" - * a.last(2) #=> ["y", "z"] + * a = [:foo, 'bar', 2] + * a.last(0) # [] + * + * Related: #first. */ VALUE rb_ary_last(int argc, const VALUE *argv, VALUE ary) { if (argc == 0) { - long len = RARRAY_LEN(ary); - if (len == 0) return Qnil; - return RARRAY_AREF(ary, len-1); + long len = RARRAY_LEN(ary); + if (len == 0) return Qnil; + return RARRAY_AREF(ary, len-1); } else { - return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_LAST); + return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_LAST); } } /* * call-seq: - * ary.fetch(index) -> obj - * ary.fetch(index, default) -> obj - * ary.fetch(index) { |index| block } -> obj - * - * Tries to return the element at position +index+, but throws an IndexError - * exception if the referenced +index+ lies outside of the array bounds. This - * error can be prevented by supplying a second argument, which will act as a - * +default+ value. - * - * Alternatively, if a block is given it will only be executed when an - * invalid +index+ is referenced. - * - * Negative values of +index+ count from the end of the array. - * - * a = [ 11, 22, 33, 44 ] - * a.fetch(1) #=> 22 - * a.fetch(-1) #=> 44 - * a.fetch(4, 'cat') #=> "cat" - * a.fetch(100) { |i| puts "#{i} is out of bounds" } - * #=> "100 is out of bounds" + * array.fetch(index) -> element + * array.fetch(index, default_value) -> element + * array.fetch(index) {|index| ... } -> element + * + * Returns the element at offset +index+. + * + * With the single \Integer argument +index+, + * returns the element at offset +index+: + * + * a = [:foo, 'bar', 2] + * a.fetch(1) # => "bar" + * + * If +index+ is negative, counts from the end of the array: + * + * a = [:foo, 'bar', 2] + * a.fetch(-1) # => 2 + * a.fetch(-2) # => "bar" + * + * With arguments +index+ and +default_value+, + * returns the element at offset +index+ if index is in range, + * otherwise returns +default_value+: + * + * a = [:foo, 'bar', 2] + * a.fetch(1, nil) # => "bar" + * + * With argument +index+ and a block, + * returns the element at offset +index+ if index is in range + * (and the block is not called); otherwise calls the block with index and returns its return value: + * + * a = [:foo, 'bar', 2] + * a.fetch(1) {|index| raise 'Cannot happen' } # => "bar" + * a.fetch(50) {|index| "Value for #{index}" } # => "Value for 50" + * */ static VALUE @@ -1435,48 +2172,60 @@ rb_ary_fetch(int argc, VALUE *argv, VALUE ary) rb_scan_args(argc, argv, "11", &pos, &ifnone); block_given = rb_block_given_p(); if (block_given && argc == 2) { - rb_warn("block supersedes default value argument"); + rb_warn("block supersedes default value argument"); } idx = NUM2LONG(pos); if (idx < 0) { - idx += RARRAY_LEN(ary); + idx += RARRAY_LEN(ary); } if (idx < 0 || RARRAY_LEN(ary) <= idx) { - if (block_given) return rb_yield(pos); - if (argc == 1) { - rb_raise(rb_eIndexError, "index %ld outside of array bounds: %ld...%ld", - idx - (idx < 0 ? RARRAY_LEN(ary) : 0), -RARRAY_LEN(ary), RARRAY_LEN(ary)); - } - return ifnone; + if (block_given) return rb_yield(pos); + if (argc == 1) { + rb_raise(rb_eIndexError, "index %ld outside of array bounds: %ld...%ld", + idx - (idx < 0 ? RARRAY_LEN(ary) : 0), -RARRAY_LEN(ary), RARRAY_LEN(ary)); + } + return ifnone; } return RARRAY_AREF(ary, idx); } /* * call-seq: - * ary.find_index(obj) -> int or nil - * ary.find_index { |item| block } -> int or nil - * ary.find_index -> Enumerator - * ary.index(obj) -> int or nil - * ary.index { |item| block } -> int or nil - * ary.index -> Enumerator + * array.index(object) -> integer or nil + * array.index {|element| ... } -> integer or nil + * array.index -> new_enumerator + * + * Returns the index of a specified element. * - * Returns the _index_ of the first object in +ary+ such that the object is - * <code>==</code> to +obj+. + * When argument +object+ is given but no block, + * returns the index of the first element +element+ + * for which <tt>object == element</tt>: * - * If a block is given instead of an argument, returns the _index_ of the - * first object for which the block returns +true+. Returns +nil+ if no - * match is found. + * a = [:foo, 'bar', 2, 'bar'] + * a.index('bar') # => 1 * - * See also Array#rindex. + * Returns +nil+ if no such element found. * - * An Enumerator is returned if neither a block nor argument is given. + * When both argument +object+ and a block are given, + * calls the block with each successive element; + * returns the index of the first element for which the block returns a truthy value: * - * a = [ "a", "b", "c" ] - * a.index("b") #=> 1 - * a.index("z") #=> nil - * a.index { |x| x == "b" } #=> 1 + * a = [:foo, 'bar', 2, 'bar'] + * a.index {|element| element == 'bar' } # => 1 + * + * Returns +nil+ if the block never returns a truthy value. + * + * When neither an argument nor a block is given, returns a new Enumerator: + * + * a = [:foo, 'bar', 2] + * e = a.index + * e # => #<Enumerator: [:foo, "bar", 2]:index> + * e.each {|element| element == 'bar' } # => 1 + * + * Array#find_index is an alias for Array#index. + * + * Related: #rindex. */ static VALUE @@ -1486,49 +2235,58 @@ rb_ary_index(int argc, VALUE *argv, VALUE ary) long i; if (argc == 0) { - RETURN_ENUMERATOR(ary, 0, 0); - for (i=0; i<RARRAY_LEN(ary); i++) { - if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) { - return LONG2NUM(i); - } - } - return Qnil; + RETURN_ENUMERATOR(ary, 0, 0); + for (i=0; i<RARRAY_LEN(ary); i++) { + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) { + return LONG2NUM(i); + } + } + return Qnil; } rb_check_arity(argc, 0, 1); val = argv[0]; if (rb_block_given_p()) - rb_warn("given block not used"); + rb_warn("given block not used"); for (i=0; i<RARRAY_LEN(ary); i++) { - VALUE e = RARRAY_AREF(ary, i); - if (rb_equal(e, val)) { - return LONG2NUM(i); - } + VALUE e = RARRAY_AREF(ary, i); + if (rb_equal(e, val)) { + return LONG2NUM(i); + } } return Qnil; } /* * call-seq: - * ary.rindex(obj) -> int or nil - * ary.rindex { |item| block } -> int or nil - * ary.rindex -> Enumerator + * array.rindex(object) -> integer or nil + * array.rindex {|element| ... } -> integer or nil + * array.rindex -> new_enumerator + * + * Returns the index of the last element for which <tt>object == element</tt>. + * + * When argument +object+ is given but no block, returns the index of the last such element found: + * + * a = [:foo, 'bar', 2, 'bar'] + * a.rindex('bar') # => 3 * - * Returns the _index_ of the last object in +self+ <code>==</code> to +obj+. + * Returns +nil+ if no such object found. * - * If a block is given instead of an argument, returns the _index_ of the - * first object for which the block returns +true+, starting from the last - * object. + * When a block is given but no argument, calls the block with each successive element; + * returns the index of the last element for which the block returns a truthy value: * - * Returns +nil+ if no match is found. + * a = [:foo, 'bar', 2, 'bar'] + * a.rindex {|element| element == 'bar' } # => 3 * - * See also Array#index. + * Returns +nil+ if the block never returns a truthy value. * - * If neither block nor argument is given, an Enumerator is returned instead. + * When neither an argument nor a block is given, returns a new \Enumerator: * - * a = [ "a", "b", "b", "b", "c" ] - * a.rindex("b") #=> 3 - * a.rindex("z") #=> nil - * a.rindex { |x| x == "b" } #=> 3 + * a = [:foo, 'bar', 2, 'bar'] + * e = a.rindex + * e # => #<Enumerator: [:foo, "bar", 2, "bar"]:rindex> + * e.each {|element| element == 'bar' } # => 3 + * + * Related: #index. */ static VALUE @@ -1538,25 +2296,28 @@ rb_ary_rindex(int argc, VALUE *argv, VALUE ary) long i = RARRAY_LEN(ary), len; if (argc == 0) { - RETURN_ENUMERATOR(ary, 0, 0); - while (i--) { - if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) - return LONG2NUM(i); - if (i > (len = RARRAY_LEN(ary))) { - i = len; - } - } - return Qnil; + RETURN_ENUMERATOR(ary, 0, 0); + while (i--) { + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) + return LONG2NUM(i); + if (i > (len = RARRAY_LEN(ary))) { + i = len; + } + } + return Qnil; } rb_check_arity(argc, 0, 1); val = argv[0]; if (rb_block_given_p()) - rb_warn("given block not used"); + rb_warn("given block not used"); while (i--) { - VALUE e = RARRAY_AREF(ary, i); - if (rb_equal(e, val)) { - return LONG2NUM(i); - } + VALUE e = RARRAY_AREF(ary, i); + if (rb_equal(e, val)) { + return LONG2NUM(i); + } + if (i > RARRAY_LEN(ary)) { + break; + } } return Qnil; } @@ -1579,57 +2340,64 @@ rb_ary_splice(VALUE ary, long beg, long len, const VALUE *rptr, long rlen) if (len < 0) rb_raise(rb_eIndexError, "negative length (%ld)", len); olen = RARRAY_LEN(ary); if (beg < 0) { - beg += olen; - if (beg < 0) { - rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", - beg - olen, -olen); - } + beg += olen; + if (beg < 0) { + rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", + beg - olen, -olen); + } } if (olen < len || olen < beg + len) { - len = olen - beg; + len = olen - beg; } { - const VALUE *optr = RARRAY_CONST_PTR(ary); - rofs = (rptr >= optr && rptr < optr + olen) ? rptr - optr : -1; + const VALUE *optr = RARRAY_CONST_PTR_TRANSIENT(ary); + rofs = (rptr >= optr && rptr < optr + olen) ? rptr - optr : -1; } if (beg >= olen) { - VALUE target_ary; - if (beg > ARY_MAX_SIZE - rlen) { - rb_raise(rb_eIndexError, "index %ld too big", beg); - } - target_ary = ary_ensure_room_for_push(ary, rlen-len); /* len is 0 or negative */ - len = beg + rlen; - ary_mem_clear(ary, olen, beg - olen); - if (rlen > 0) { - if (rofs != -1) rptr = RARRAY_CONST_PTR(ary) + rofs; - ary_memcpy0(ary, beg, rlen, rptr, target_ary); - } - ARY_SET_LEN(ary, len); + VALUE target_ary; + if (beg > ARY_MAX_SIZE - rlen) { + rb_raise(rb_eIndexError, "index %ld too big", beg); + } + target_ary = ary_ensure_room_for_push(ary, rlen-len); /* len is 0 or negative */ + len = beg + rlen; + ary_mem_clear(ary, olen, beg - olen); + if (rlen > 0) { + if (rofs != -1) rptr = RARRAY_CONST_PTR_TRANSIENT(ary) + rofs; + ary_memcpy0(ary, beg, rlen, rptr, target_ary); + } + ARY_SET_LEN(ary, len); } else { - long alen; - - if (olen - len > ARY_MAX_SIZE - rlen) { - rb_raise(rb_eIndexError, "index %ld too big", olen + rlen - len); - } - rb_ary_modify(ary); - alen = olen + rlen - len; - if (alen >= ARY_CAPA(ary)) { - ary_double_capa(ary, alen); - } - - if (len != rlen) { - RARRAY_PTR_USE(ary, ptr, - MEMMOVE(ptr + beg + rlen, ptr + beg + len, - VALUE, olen - (beg + len))); - ARY_SET_LEN(ary, alen); - } - if (rlen > 0) { - if (rofs != -1) rptr = RARRAY_CONST_PTR(ary) + rofs; - MEMMOVE(RARRAY_PTR(ary) + beg, rptr, VALUE, rlen); - } + long alen; + + if (olen - len > ARY_MAX_SIZE - rlen) { + rb_raise(rb_eIndexError, "index %ld too big", olen + rlen - len); + } + rb_ary_modify(ary); + alen = olen + rlen - len; + if (alen >= ARY_CAPA(ary)) { + ary_double_capa(ary, alen); + } + + if (len != rlen) { + RARRAY_PTR_USE_TRANSIENT(ary, ptr, + MEMMOVE(ptr + beg + rlen, ptr + beg + len, + VALUE, olen - (beg + len))); + ARY_SET_LEN(ary, alen); + } + if (rlen > 0) { + if (rofs != -1) rptr = RARRAY_CONST_PTR_TRANSIENT(ary) + rofs; + /* give up wb-protected ary */ + RB_OBJ_WB_UNPROTECT_FOR(ARRAY, ary); + + /* do not use RARRAY_PTR() because it can causes GC. + * ary can contain T_NONE object because it is not cleared. + */ + RARRAY_PTR_USE_TRANSIENT(ary, ptr, + MEMMOVE(ptr + beg, rptr, VALUE, rlen)); + } } } @@ -1640,22 +2408,14 @@ rb_ary_set_len(VALUE ary, long len) rb_ary_modify_check(ary); if (ARY_SHARED_P(ary)) { - rb_raise(rb_eRuntimeError, "can't set length of shared "); + rb_raise(rb_eRuntimeError, "can't set length of shared "); } if (len > (capa = (long)ARY_CAPA(ary))) { - rb_bug("probable buffer overflow: %ld for %ld", len, capa); + rb_bug("probable buffer overflow: %ld for %ld", len, capa); } ARY_SET_LEN(ary, len); } -/*! - * expands or shrinks \a ary to \a len elements. - * expanded region will be filled with Qnil. - * \param ary an array - * \param len new size - * \return \a ary - * \post the size of \a ary is \a len. - */ VALUE rb_ary_resize(VALUE ary, long len) { @@ -1665,117 +2425,231 @@ rb_ary_resize(VALUE ary, long len) olen = RARRAY_LEN(ary); if (len == olen) return ary; if (len > ARY_MAX_SIZE) { - rb_raise(rb_eIndexError, "index %ld too big", len); + rb_raise(rb_eIndexError, "index %ld too big", len); } if (len > olen) { - if (len >= ARY_CAPA(ary)) { - ary_double_capa(ary, len); - } - ary_mem_clear(ary, olen, len - olen); - ARY_SET_LEN(ary, len); + if (len >= ARY_CAPA(ary)) { + ary_double_capa(ary, len); + } + ary_mem_clear(ary, olen, len - olen); + ARY_SET_LEN(ary, len); } else if (ARY_EMBED_P(ary)) { ARY_SET_EMBED_LEN(ary, len); } - else if (len <= RARRAY_EMBED_LEN_MAX) { - VALUE tmp[RARRAY_EMBED_LEN_MAX]; - MEMCPY(tmp, ARY_HEAP_PTR(ary), VALUE, len); - ary_discard(ary); - MEMCPY((VALUE *)ARY_EMBED_PTR(ary), tmp, VALUE, len); /* WB: no new reference */ + else if (len <= ary_embed_capa(ary)) { + const VALUE *ptr = ARY_HEAP_PTR(ary); + long ptr_capa = ARY_HEAP_SIZE(ary); + bool is_malloc_ptr = !ARY_SHARED_P(ary) && !RARRAY_TRANSIENT_P(ary); + + FL_UNSET(ary, RARRAY_TRANSIENT_FLAG); + FL_SET_EMBED(ary); + + MEMCPY((VALUE *)ARY_EMBED_PTR(ary), ptr, VALUE, len); /* WB: no new reference */ ARY_SET_EMBED_LEN(ary, len); + + if (is_malloc_ptr) ruby_sized_xfree((void *)ptr, ptr_capa); } else { - if (olen > len + ARY_DEFAULT_SIZE) { - SIZED_REALLOC_N(RARRAY(ary)->as.heap.ptr, VALUE, len, RARRAY(ary)->as.heap.aux.capa); - ARY_SET_CAPA(ary, len); - } - ARY_SET_HEAP_LEN(ary, len); + if (olen > len + ARY_DEFAULT_SIZE) { + size_t new_capa = ary_heap_realloc(ary, len); + ARY_SET_CAPA(ary, new_capa); + } + ARY_SET_HEAP_LEN(ary, len); } + ary_verify(ary); return ary; } +static VALUE +ary_aset_by_rb_ary_store(VALUE ary, long key, VALUE val) +{ + rb_ary_store(ary, key, val); + return val; +} + +static VALUE +ary_aset_by_rb_ary_splice(VALUE ary, long beg, long len, VALUE val) +{ + VALUE rpl = rb_ary_to_ary(val); + rb_ary_splice(ary, beg, len, RARRAY_CONST_PTR_TRANSIENT(rpl), RARRAY_LEN(rpl)); + RB_GC_GUARD(rpl); + return val; +} + /* * call-seq: - * ary[index] = obj -> obj - * ary[start, length] = obj or other_ary or nil -> obj or other_ary or nil - * ary[range] = obj or other_ary or nil -> obj or other_ary or nil - * - * Element Assignment --- Sets the element at +index+, or replaces a subarray - * from the +start+ index for +length+ elements, or replaces a subarray - * specified by the +range+ of indices. - * - * If indices are greater than the current capacity of the array, the array - * grows automatically. Elements are inserted into the array at +start+ if - * +length+ is zero. - * - * Negative indices will count backward from the end of the array. For - * +start+ and +range+ cases the starting index is just before an element. - * - * An IndexError is raised if a negative index points past the beginning of - * the array. - * - * See also Array#push, and Array#unshift. - * - * a = Array.new - * a[4] = "4"; #=> [nil, nil, nil, nil, "4"] - * a[0, 3] = [ 'a', 'b', 'c' ] #=> ["a", "b", "c", nil, "4"] - * a[1..2] = [ 1, 2 ] #=> ["a", 1, 2, nil, "4"] - * a[0, 2] = "?" #=> ["?", 2, nil, "4"] - * a[0..2] = "A" #=> ["A", "4"] - * a[-1] = "Z" #=> ["A", "Z"] - * a[1..-1] = nil #=> ["A", nil] - * a[1..-1] = [] #=> ["A"] - * a[0, 0] = [ 1, 2 ] #=> [1, 2, "A"] - * a[3, 0] = "B" #=> [1, 2, "A", "B"] + * array[index] = object -> object + * array[start, length] = object -> object + * array[range] = object -> object + * + * Assigns elements in +self+; returns the given +object+. + * + * When \Integer argument +index+ is given, assigns +object+ to an element in +self+. + * + * If +index+ is non-negative, assigns +object+ the element at offset +index+: + * + * a = [:foo, 'bar', 2] + * a[0] = 'foo' # => "foo" + * a # => ["foo", "bar", 2] + * + * If +index+ is greater than <tt>self.length</tt>, extends the array: + * + * a = [:foo, 'bar', 2] + * a[7] = 'foo' # => "foo" + * a # => [:foo, "bar", 2, nil, nil, nil, nil, "foo"] + * + * If +index+ is negative, counts backwards from the end of the array: + * + * a = [:foo, 'bar', 2] + * a[-1] = 'two' # => "two" + * a # => [:foo, "bar", "two"] + * + * When \Integer arguments +start+ and +length+ are given and +object+ is not an \Array, + * removes <tt>length - 1</tt> elements beginning at offset +start+, + * and assigns +object+ at offset +start+: + * + * a = [:foo, 'bar', 2] + * a[0, 2] = 'foo' # => "foo" + * a # => ["foo", 2] + * + * If +start+ is negative, counts backwards from the end of the array: + * + * a = [:foo, 'bar', 2] + * a[-2, 2] = 'foo' # => "foo" + * a # => [:foo, "foo"] + * + * If +start+ is non-negative and outside the array (<tt> >= self.size</tt>), + * extends the array with +nil+, assigns +object+ at offset +start+, + * and ignores +length+: + * + * a = [:foo, 'bar', 2] + * a[6, 50] = 'foo' # => "foo" + * a # => [:foo, "bar", 2, nil, nil, nil, "foo"] + * + * If +length+ is zero, shifts elements at and following offset +start+ + * and assigns +object+ at offset +start+: + * + * a = [:foo, 'bar', 2] + * a[1, 0] = 'foo' # => "foo" + * a # => [:foo, "foo", "bar", 2] + * + * If +length+ is too large for the existing array, does not extend the array: + * + * a = [:foo, 'bar', 2] + * a[1, 5] = 'foo' # => "foo" + * a # => [:foo, "foo"] + * + * When \Range argument +range+ is given and +object+ is an \Array, + * removes <tt>length - 1</tt> elements beginning at offset +start+, + * and assigns +object+ at offset +start+: + * + * a = [:foo, 'bar', 2] + * a[0..1] = 'foo' # => "foo" + * a # => ["foo", 2] + * + * if <tt>range.begin</tt> is negative, counts backwards from the end of the array: + * + * a = [:foo, 'bar', 2] + * a[-2..2] = 'foo' # => "foo" + * a # => [:foo, "foo"] + * + * If the array length is less than <tt>range.begin</tt>, + * assigns +object+ at offset <tt>range.begin</tt>, and ignores +length+: + * + * a = [:foo, 'bar', 2] + * a[6..50] = 'foo' # => "foo" + * a # => [:foo, "bar", 2, nil, nil, nil, "foo"] + * + * If <tt>range.end</tt> is zero, shifts elements at and following offset +start+ + * and assigns +object+ at offset +start+: + * + * a = [:foo, 'bar', 2] + * a[1..0] = 'foo' # => "foo" + * a # => [:foo, "foo", "bar", 2] + * + * If <tt>range.end</tt> is negative, assigns +object+ at offset +start+, + * retains <tt>range.end.abs -1</tt> elements past that, and removes those beyond: + * + * a = [:foo, 'bar', 2] + * a[1..-1] = 'foo' # => "foo" + * a # => [:foo, "foo"] + * a = [:foo, 'bar', 2] + * a[1..-2] = 'foo' # => "foo" + * a # => [:foo, "foo", 2] + * a = [:foo, 'bar', 2] + * a[1..-3] = 'foo' # => "foo" + * a # => [:foo, "foo", "bar", 2] + * a = [:foo, 'bar', 2] + * + * If <tt>range.end</tt> is too large for the existing array, + * replaces array elements, but does not extend the array with +nil+ values: + * + * a = [:foo, 'bar', 2] + * a[1..5] = 'foo' # => "foo" + * a # => [:foo, "foo"] + * */ static VALUE rb_ary_aset(int argc, VALUE *argv, VALUE ary) { long offset, beg, len; - VALUE rpl; + rb_check_arity(argc, 2, 3); + rb_ary_modify_check(ary); if (argc == 3) { - rb_ary_modify_check(ary); - beg = NUM2LONG(argv[0]); - len = NUM2LONG(argv[1]); - goto range; + beg = NUM2LONG(argv[0]); + len = NUM2LONG(argv[1]); + return ary_aset_by_rb_ary_splice(ary, beg, len, argv[2]); } - rb_check_arity(argc, 2, 2); - rb_ary_modify_check(ary); if (FIXNUM_P(argv[0])) { - offset = FIX2LONG(argv[0]); - goto fixnum; + offset = FIX2LONG(argv[0]); + return ary_aset_by_rb_ary_store(ary, offset, argv[1]); } if (rb_range_beg_len(argv[0], &beg, &len, RARRAY_LEN(ary), 1)) { - /* check if idx is Range */ - range: - rpl = rb_ary_to_ary(argv[argc-1]); - rb_ary_splice(ary, beg, len, RARRAY_CONST_PTR(rpl), RARRAY_LEN(rpl)); - RB_GC_GUARD(rpl); - return argv[argc-1]; + /* check if idx is Range */ + return ary_aset_by_rb_ary_splice(ary, beg, len, argv[1]); } offset = NUM2LONG(argv[0]); -fixnum: - rb_ary_store(ary, offset, argv[1]); - return argv[1]; + return ary_aset_by_rb_ary_store(ary, offset, argv[1]); } /* * call-seq: - * ary.insert(index, obj...) -> ary + * array.insert(index, *objects) -> self + * + * Inserts given +objects+ before or after the element at \Integer index +offset+; + * returns +self+. * - * Inserts the given values before the element with the given +index+. + * When +index+ is non-negative, inserts all given +objects+ + * before the element at offset +index+: * - * Negative indices count backwards from the end of the array, where +-1+ is - * the last element. If a negative index is used, the given values will be - * inserted after that element, so using an index of +-1+ will insert the - * values at the end of the array. + * a = [:foo, 'bar', 2] + * a.insert(1, :bat, :bam) # => [:foo, :bat, :bam, "bar", 2] + * + * Extends the array if +index+ is beyond the array (<tt>index >= self.size</tt>): + * + * a = [:foo, 'bar', 2] + * a.insert(5, :bat, :bam) + * a # => [:foo, "bar", 2, nil, nil, :bat, :bam] + * + * Does nothing if no objects given: + * + * a = [:foo, 'bar', 2] + * a.insert(1) + * a.insert(50) + * a.insert(-50) + * a # => [:foo, "bar", 2] + * + * When +index+ is negative, inserts all given +objects+ + * _after_ the element at offset <tt>index+self.size</tt>: + * + * a = [:foo, 'bar', 2] + * a.insert(-2, :bat, :bam) + * a # => [:foo, "bar", :bat, :bam, 2] * - * a = %w{ a b c d } - * a.insert(2, 99) #=> ["a", "b", 99, "c", "d"] - * a.insert(-2, 1, 2, 3) #=> ["a", "b", 99, "c", 1, 2, 3, "d"] */ static VALUE @@ -1788,15 +2662,15 @@ rb_ary_insert(int argc, VALUE *argv, VALUE ary) pos = NUM2LONG(argv[0]); if (argc == 1) return ary; if (pos == -1) { - pos = RARRAY_LEN(ary); + pos = RARRAY_LEN(ary); } else if (pos < 0) { - long minpos = -RARRAY_LEN(ary) - 1; - if (pos < minpos) { - rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", - pos, minpos); - } - pos++; + long minpos = -RARRAY_LEN(ary) - 1; + if (pos < minpos) { + rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld", + pos, minpos); + } + pos++; } rb_ary_splice(ary, pos, 0, argv + 1, argc - 1); return ary; @@ -1813,50 +2687,104 @@ ary_enum_length(VALUE ary, VALUE args, VALUE eobj) /* * call-seq: - * ary.each { |item| block } -> ary - * ary.each -> Enumerator + * array.each {|element| ... } -> self + * array.each -> Enumerator + * + * Iterates over array elements. + * + * When a block given, passes each successive array element to the block; + * returns +self+: + * + * a = [:foo, 'bar', 2] + * a.each {|element| puts "#{element.class} #{element}" } + * + * Output: + * + * Symbol foo + * String bar + * Integer 2 * - * Calls the given block once for each element in +self+, passing that element - * as a parameter. Returns the array itself. + * Allows the array to be modified during iteration: * - * If no block is given, an Enumerator is returned. + * a = [:foo, 'bar', 2] + * a.each {|element| puts element; a.clear if element.to_s.start_with?('b') } * - * a = [ "a", "b", "c" ] - * a.each {|x| print x, " -- " } + * Output: * - * produces: + * foo + * bar * - * a -- b -- c -- + * When no block given, returns a new \Enumerator: + * a = [:foo, 'bar', 2] + * + * e = a.each + * e # => #<Enumerator: [:foo, "bar", 2]:each> + * a1 = e.each {|element| puts "#{element.class} #{element}" } + * + * Output: + * + * Symbol foo + * String bar + * Integer 2 + * + * Related: #each_index, #reverse_each. */ VALUE rb_ary_each(VALUE ary) { long i; - + ary_verify(ary); RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); for (i=0; i<RARRAY_LEN(ary); i++) { - rb_yield(RARRAY_AREF(ary, i)); + rb_yield(RARRAY_AREF(ary, i)); } return ary; } /* * call-seq: - * ary.each_index { |index| block } -> ary - * ary.each_index -> Enumerator + * array.each_index {|index| ... } -> self + * array.each_index -> Enumerator + * + * Iterates over array indexes. + * + * When a block given, passes each successive array index to the block; + * returns +self+: + * + * a = [:foo, 'bar', 2] + * a.each_index {|index| puts "#{index} #{a[index]}" } + * + * Output: + * + * 0 foo + * 1 bar + * 2 2 + * + * Allows the array to be modified during iteration: + * + * a = [:foo, 'bar', 2] + * a.each_index {|index| puts index; a.clear if index > 0 } + * + * Output: * - * Same as Array#each, but passes the +index+ of the element instead of the - * element itself. + * 0 + * 1 * - * An Enumerator is returned if no block is given. + * When no block given, returns a new \Enumerator: * - * a = [ "a", "b", "c" ] - * a.each_index {|x| print x, " -- " } + * a = [:foo, 'bar', 2] + * e = a.each_index + * e # => #<Enumerator: [:foo, "bar", 2]:each_index> + * a1 = e.each {|index| puts "#{index} #{a[index]}"} * - * produces: + * Output: * - * 0 -- 1 -- 2 -- + * 0 foo + * 1 bar + * 2 2 + * + * Related: #each, #reverse_each. */ static VALUE @@ -1866,24 +2794,54 @@ rb_ary_each_index(VALUE ary) RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); for (i=0; i<RARRAY_LEN(ary); i++) { - rb_yield(LONG2NUM(i)); + rb_yield(LONG2NUM(i)); } return ary; } /* * call-seq: - * ary.reverse_each { |item| block } -> ary - * ary.reverse_each -> Enumerator + * array.reverse_each {|element| ... } -> self + * array.reverse_each -> Enumerator + * + * Iterates backwards over array elements. + * + * When a block given, passes, in reverse order, each element to the block; + * returns +self+: + * + * a = [:foo, 'bar', 2] + * a.reverse_each {|element| puts "#{element.class} #{element}" } + * + * Output: + * + * Integer 2 + * String bar + * Symbol foo + * + * Allows the array to be modified during iteration: * - * Same as Array#each, but traverses +self+ in reverse order. + * a = [:foo, 'bar', 2] + * a.reverse_each {|element| puts element; a.clear if element.to_s.start_with?('b') } * - * a = [ "a", "b", "c" ] - * a.reverse_each {|x| print x, " " } + * Output: * - * produces: + * 2 + * bar * - * c b a + * When no block given, returns a new \Enumerator: + * + * a = [:foo, 'bar', 2] + * e = a.reverse_each + * e # => #<Enumerator: [:foo, "bar", 2]:reverse_each> + * a1 = e.each {|element| puts "#{element.class} #{element}" } + * + * Output: + * + * Integer 2 + * String bar + * Symbol foo + * + * Related: #each, #each_index. */ static VALUE @@ -1894,24 +2852,21 @@ rb_ary_reverse_each(VALUE ary) RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); len = RARRAY_LEN(ary); while (len--) { - long nlen; - rb_yield(RARRAY_AREF(ary, len)); - nlen = RARRAY_LEN(ary); - if (nlen < len) { - len = nlen; - } + long nlen; + rb_yield(RARRAY_AREF(ary, len)); + nlen = RARRAY_LEN(ary); + if (nlen < len) { + len = nlen; + } } return ary; } /* * call-seq: - * ary.length -> int - * - * Returns the number of elements in +self+. May be zero. + * array.length -> an_integer * - * [ 1, 2, 3, 4, 5 ].length #=> 5 - * [].length #=> 0 + * Returns the count of elements in +self+. */ static VALUE @@ -1923,19 +2878,16 @@ rb_ary_length(VALUE ary) /* * call-seq: - * ary.empty? -> true or false + * array.empty? -> true or false * - * Returns +true+ if +self+ contains no elements. - * - * [].empty? #=> true + * Returns +true+ if the count of elements in +self+ is zero, + * +false+ otherwise. */ static VALUE rb_ary_empty_p(VALUE ary) { - if (RARRAY_LEN(ary) == 0) - return Qtrue; - return Qfalse; + return RBOOL(RARRAY_LEN(ary) == 0); } VALUE @@ -1943,15 +2895,18 @@ rb_ary_dup(VALUE ary) { long len = RARRAY_LEN(ary); VALUE dup = rb_ary_new2(len); - ary_memcpy(dup, 0, len, RARRAY_CONST_PTR(ary)); + ary_memcpy(dup, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary)); ARY_SET_LEN(dup, len); + + ary_verify(ary); + ary_verify(dup); return dup; } VALUE rb_ary_resurrect(VALUE ary) { - return rb_ary_new4(RARRAY_LEN(ary), RARRAY_CONST_PTR(ary)); + return ary_make_partial(ary, rb_cArray, 0, RARRAY_LEN(ary)); } extern VALUE rb_output_fs; @@ -1968,15 +2923,15 @@ recursive_join(VALUE obj, VALUE argp, int recur) int *first = (int *)arg[3]; if (recur) { - rb_raise(rb_eArgError, "recursive array join"); + rb_raise(rb_eArgError, "recursive array join"); } else { - ary_join_1(obj, ary, sep, 0, result, first); + ary_join_1(obj, ary, sep, 0, result, first); } return Qnil; } -static void +static long ary_join_0(VALUE ary, VALUE sep, long max, VALUE result) { long i; @@ -1984,11 +2939,40 @@ ary_join_0(VALUE ary, VALUE sep, long max, VALUE result) if (max > 0) rb_enc_copy(result, RARRAY_AREF(ary, 0)); for (i=0; i<max; i++) { - val = RARRAY_AREF(ary, i); - if (i > 0 && !NIL_P(sep)) - rb_str_buf_append(result, sep); - rb_str_buf_append(result, val); - if (OBJ_TAINTED(val)) OBJ_TAINT(result); + val = RARRAY_AREF(ary, i); + if (!RB_TYPE_P(val, T_STRING)) break; + if (i > 0 && !NIL_P(sep)) + rb_str_buf_append(result, sep); + rb_str_buf_append(result, val); + } + return i; +} + +static void +ary_join_1_str(VALUE dst, VALUE src, int *first) +{ + rb_str_buf_append(dst, src); + if (*first) { + rb_enc_copy(dst, src); + *first = FALSE; + } +} + +static void +ary_join_1_ary(VALUE obj, VALUE ary, VALUE sep, VALUE result, VALUE val, int *first) +{ + if (val == ary) { + rb_raise(rb_eArgError, "recursive array join"); + } + else { + VALUE args[4]; + + *first = FALSE; + args[0] = val; + args[1] = sep; + args[2] = result; + args[3] = (VALUE)first; + rb_exec_recursive(recursive_join, obj, (VALUE)args); } } @@ -1998,50 +2982,25 @@ ary_join_1(VALUE obj, VALUE ary, VALUE sep, long i, VALUE result, int *first) VALUE val, tmp; for (; i<RARRAY_LEN(ary); i++) { - if (i > 0 && !NIL_P(sep)) - rb_str_buf_append(result, sep); - - val = RARRAY_AREF(ary, i); - if (RB_TYPE_P(val, T_STRING)) { - str_join: - rb_str_buf_append(result, val); - if (*first) { - rb_enc_copy(result, val); - *first = FALSE; - } - } - else if (RB_TYPE_P(val, T_ARRAY)) { - obj = val; - ary_join: - if (val == ary) { - rb_raise(rb_eArgError, "recursive array join"); - } - else { - VALUE args[4]; - - *first = FALSE; - args[0] = val; - args[1] = sep; - args[2] = result; - args[3] = (VALUE)first; - rb_exec_recursive(recursive_join, obj, (VALUE)args); - } - } - else { - tmp = rb_check_string_type(val); - if (!NIL_P(tmp)) { - val = tmp; - goto str_join; - } - tmp = rb_check_array_type(val); - if (!NIL_P(tmp)) { - obj = val; - val = tmp; - goto ary_join; - } - val = rb_obj_as_string(val); - goto str_join; - } + if (i > 0 && !NIL_P(sep)) + rb_str_buf_append(result, sep); + + val = RARRAY_AREF(ary, i); + if (RB_TYPE_P(val, T_STRING)) { + ary_join_1_str(result, val, first); + } + else if (RB_TYPE_P(val, T_ARRAY)) { + ary_join_1_ary(val, ary, sep, result, val, first); + } + else if (!NIL_P(tmp = rb_check_string_type(val))) { + ary_join_1_str(result, tmp, first); + } + else if (!NIL_P(tmp = rb_check_array_type(val))) { + ary_join_1_ary(val, ary, sep, result, tmp, first); + } + else { + ary_join_1_str(result, rb_obj_as_string(val), first); + } } } @@ -2049,36 +3008,36 @@ VALUE rb_ary_join(VALUE ary, VALUE sep) { long len = 1, i; - int taint = FALSE; VALUE val, tmp, result; if (RARRAY_LEN(ary) == 0) return rb_usascii_str_new(0, 0); - if (OBJ_TAINTED(ary)) taint = TRUE; if (!NIL_P(sep)) { - StringValue(sep); - len += RSTRING_LEN(sep) * (RARRAY_LEN(ary) - 1); + StringValue(sep); + len += RSTRING_LEN(sep) * (RARRAY_LEN(ary) - 1); } for (i=0; i<RARRAY_LEN(ary); i++) { - val = RARRAY_AREF(ary, i); - tmp = rb_check_string_type(val); - - if (NIL_P(tmp) || tmp != val) { - int first; - result = rb_str_buf_new(len + (RARRAY_LEN(ary)-i)*10); - rb_enc_associate(result, rb_usascii_encoding()); - if (taint) OBJ_TAINT(result); - ary_join_0(ary, sep, i, result); - first = i == 0; - ary_join_1(ary, ary, sep, i, result, &first); - return result; - } + val = RARRAY_AREF(ary, i); + tmp = rb_check_string_type(val); + + if (NIL_P(tmp) || tmp != val) { + int first; + long n = RARRAY_LEN(ary); + if (i > n) i = n; + result = rb_str_buf_new(len + (n-i)*10); + rb_enc_associate(result, rb_usascii_encoding()); + i = ary_join_0(ary, sep, i, result); + first = i == 0; + ary_join_1(ary, ary, sep, i, result, &first); + return result; + } - len += RSTRING_LEN(tmp); + len += RSTRING_LEN(tmp); } - result = rb_str_buf_new(len); - if (taint) OBJ_TAINT(result); + result = rb_str_new(0, len); + rb_str_set_len(result, 0); + ary_join_0(ary, sep, RARRAY_LEN(ary), result); return result; @@ -2086,29 +3045,43 @@ rb_ary_join(VALUE ary, VALUE sep) /* * call-seq: - * ary.join(separator=$,) -> str + * array.join ->new_string + * array.join(separator = $,) -> new_string + * + * Returns the new \String formed by joining the array elements after conversion. + * For each element +element+: + * + * - Uses <tt>element.to_s</tt> if +element+ is not a <tt>kind_of?(Array)</tt>. + * - Uses recursive <tt>element.join(separator)</tt> if +element+ is a <tt>kind_of?(Array)</tt>. + * + * With no argument, joins using the output field separator, <tt>$,</tt>: + * + * a = [:foo, 'bar', 2] + * $, # => nil + * a.join # => "foobar2" + * + * With \string argument +separator+, joins using that separator: * - * Returns a string created by converting each element of the array to - * a string, separated by the given +separator+. - * If the +separator+ is +nil+, it uses current <code>$,</code>. - * If both the +separator+ and <code>$,</code> are +nil+, - * it uses an empty string. + * a = [:foo, 'bar', 2] + * a.join("\n") # => "foo\nbar\n2" * - * [ "a", "b", "c" ].join #=> "abc" - * [ "a", "b", "c" ].join("-") #=> "a-b-c" + * Joins recursively for nested Arrays: * - * For nested arrays, join is applied recursively: + * a = [:foo, [:bar, [:baz, :bat]]] + * a.join # => "foobarbazbat" * - * [ "a", [1, 2, [:x, :y]], "b" ].join("-") #=> "a-1-2-x-y-b" */ - static VALUE rb_ary_join_m(int argc, VALUE *argv, VALUE ary) { VALUE sep; - rb_scan_args(argc, argv, "01", &sep); - if (NIL_P(sep)) sep = rb_output_fs; + if (rb_check_arity(argc, 0, 1) == 0 || NIL_P(sep = argv[0])) { + sep = rb_output_fs; + if (!NIL_P(sep)) { + rb_category_warn(RB_WARN_CATEGORY_DEPRECATED, "$, is set to non-nil value"); + } + } return rb_ary_join(ary, sep); } @@ -2116,32 +3089,32 @@ rb_ary_join_m(int argc, VALUE *argv, VALUE ary) static VALUE inspect_ary(VALUE ary, VALUE dummy, int recur) { - int tainted = OBJ_TAINTED(ary); long i; VALUE s, str; if (recur) return rb_usascii_str_new_cstr("[...]"); str = rb_str_buf_new2("["); for (i=0; i<RARRAY_LEN(ary); i++) { - s = rb_inspect(RARRAY_AREF(ary, i)); - if (OBJ_TAINTED(s)) tainted = TRUE; - if (i > 0) rb_str_buf_cat2(str, ", "); - else rb_enc_copy(str, s); - rb_str_buf_append(str, s); + s = rb_inspect(RARRAY_AREF(ary, i)); + if (i > 0) rb_str_buf_cat2(str, ", "); + else rb_enc_copy(str, s); + rb_str_buf_append(str, s); } rb_str_buf_cat2(str, "]"); - if (tainted) OBJ_TAINT(str); return str; } /* * call-seq: - * ary.inspect -> string - * ary.to_s -> string + * array.inspect -> new_string * - * Creates a string representation of +self+. + * Returns the new \String formed by calling method <tt>#inspect</tt> + * on each array element: * - * [ "a", "b", "c" ].to_s #=> "[\"a\", \"b\", \"c\"]" + * a = [:foo, 'bar', 2] + * a.inspect # => "[:foo, \"bar\", 2]" + * + * Array#to_s is an alias for Array#inspect. */ static VALUE @@ -2159,33 +3132,59 @@ rb_ary_to_s(VALUE ary) /* * call-seq: - * ary.to_a -> ary + * to_a -> self or new_array * - * Returns +self+. + * When +self+ is an instance of \Array, returns +self+: + * + * a = [:foo, 'bar', 2] + * a.to_a # => [:foo, "bar", 2] + * + * Otherwise, returns a new \Array containing the elements of +self+: + * + * class MyArray < Array; end + * a = MyArray.new(['foo', 'bar', 'two']) + * a.instance_of?(Array) # => false + * a.kind_of?(Array) # => true + * a1 = a.to_a + * a1 # => ["foo", "bar", "two"] + * a1.class # => Array # Not MyArray * - * If called on a subclass of Array, converts the receiver to an Array object. */ static VALUE rb_ary_to_a(VALUE ary) { if (rb_obj_class(ary) != rb_cArray) { - VALUE dup = rb_ary_new2(RARRAY_LEN(ary)); - rb_ary_replace(dup, ary); - return dup; + VALUE dup = rb_ary_new2(RARRAY_LEN(ary)); + rb_ary_replace(dup, ary); + return dup; } return ary; } /* * call-seq: - * ary.to_h -> hash + * array.to_h -> new_hash + * array.to_h {|item| ... } -> new_hash + * + * Returns a new \Hash formed from +self+. + * + * When a block is given, calls the block with each array element; + * the block must return a 2-element \Array whose two elements + * form a key-value pair in the returned \Hash: * - * Returns the result of interpreting <i>ary</i> as an array of - * <tt>[key, value]</tt> pairs. + * a = ['foo', :bar, 1, [2, 3], {baz: 4}] + * h = a.to_h {|item| [item, item] } + * h # => {"foo"=>"foo", :bar=>:bar, 1=>1, [2, 3]=>[2, 3], {:baz=>4}=>{:baz=>4}} + * + * When no block is given, +self+ must be an \Array of 2-element sub-arrays, + * each sub-array is formed into a key-value pair in the new \Hash: + * + * [].to_h # => {} + * a = [['foo', 'zero'], ['bar', 'one'], ['baz', 'two']] + * h = a.to_h + * h # => {"foo"=>"zero", "bar"=>"one", "baz"=>"two"} * - * [[:foo, :bar], [1, 2]].to_h - * # => {:foo => :bar, 1 => 2} */ static VALUE @@ -2193,25 +3192,28 @@ rb_ary_to_h(VALUE ary) { long i; VALUE hash = rb_hash_new_with_size(RARRAY_LEN(ary)); + int block_given = rb_block_given_p(); + for (i=0; i<RARRAY_LEN(ary); i++) { - const VALUE elt = rb_ary_elt(ary, i); - const VALUE key_value_pair = rb_check_array_type(elt); - if (NIL_P(key_value_pair)) { - rb_raise(rb_eTypeError, "wrong element type %"PRIsVALUE" at %ld (expected array)", - rb_obj_class(elt), i); - } - if (RARRAY_LEN(key_value_pair) != 2) { - rb_raise(rb_eArgError, "wrong array length at %ld (expected 2, was %ld)", - i, RARRAY_LEN(key_value_pair)); - } - rb_hash_aset(hash, RARRAY_AREF(key_value_pair, 0), RARRAY_AREF(key_value_pair, 1)); + const VALUE e = rb_ary_elt(ary, i); + const VALUE elt = block_given ? rb_yield_force_blockarg(e) : e; + const VALUE key_value_pair = rb_check_array_type(elt); + if (NIL_P(key_value_pair)) { + rb_raise(rb_eTypeError, "wrong element type %"PRIsVALUE" at %ld (expected array)", + rb_obj_class(elt), i); + } + if (RARRAY_LEN(key_value_pair) != 2) { + rb_raise(rb_eArgError, "wrong array length at %ld (expected 2, was %ld)", + i, RARRAY_LEN(key_value_pair)); + } + rb_hash_aset(hash, RARRAY_AREF(key_value_pair, 0), RARRAY_AREF(key_value_pair, 1)); } return hash; } /* * call-seq: - * ary.to_ary -> ary + * array.to_ary -> self * * Returns +self+. */ @@ -2226,9 +3228,9 @@ static void ary_reverse(VALUE *p1, VALUE *p2) { while (p1 < p2) { - VALUE tmp = *p1; - *p1++ = *p2; - *p2-- = tmp; + VALUE tmp = *p1; + *p1++ = *p2; + *p2-- = tmp; } } @@ -2240,23 +3242,23 @@ rb_ary_reverse(VALUE ary) rb_ary_modify(ary); if (len > 1) { - RARRAY_PTR_USE(ary, p1, { - p2 = p1 + len - 1; /* points last item */ - ary_reverse(p1, p2); - }); /* WB: no new reference */ + RARRAY_PTR_USE_TRANSIENT(ary, p1, { + p2 = p1 + len - 1; /* points last item */ + ary_reverse(p1, p2); + }); /* WB: no new reference */ } return ary; } /* * call-seq: - * ary.reverse! -> ary + * array.reverse! -> self + * + * Reverses +self+ in place: * - * Reverses +self+ in place. + * a = ['foo', 'bar', 'two'] + * a.reverse! # => ["two", "bar", "foo"] * - * a = [ "a", "b", "c" ] - * a.reverse! #=> ["c", "b", "a"] - * a #=> ["c", "b", "a"] */ static VALUE @@ -2267,12 +3269,14 @@ rb_ary_reverse_bang(VALUE ary) /* * call-seq: - * ary.reverse -> new_ary + * array.reverse -> new_array * - * Returns a new array containing +self+'s elements in reverse order. + * Returns a new \Array with the elements of +self+ in reverse order: + * + * a = ['foo', 'bar', 'two'] + * a1 = a.reverse + * a1 # => ["two", "bar", "foo"] * - * [ "a", "b", "c" ].reverse #=> ["c", "b", "a"] - * [ 1 ].reverse #=> [1] */ static VALUE @@ -2282,9 +3286,9 @@ rb_ary_reverse_m(VALUE ary) VALUE dup = rb_ary_new2(len); if (len > 0) { - const VALUE *p1 = RARRAY_CONST_PTR(ary); - VALUE *p2 = (VALUE *)RARRAY_CONST_PTR(dup) + len - 1; - do *p2-- = *p1++; while (--len > 0); + const VALUE *p1 = RARRAY_CONST_PTR_TRANSIENT(ary); + VALUE *p2 = (VALUE *)RARRAY_CONST_PTR_TRANSIENT(dup) + len - 1; + do *p2-- = *p1++; while (--len > 0); } ARY_SET_LEN(dup, RARRAY_LEN(ary)); return dup; @@ -2296,73 +3300,143 @@ rotate_count(long cnt, long len) return (cnt < 0) ? (len - (~cnt % len) - 1) : (cnt % len); } +static void +ary_rotate_ptr(VALUE *ptr, long len, long cnt) +{ + if (cnt == 1) { + VALUE tmp = *ptr; + memmove(ptr, ptr + 1, sizeof(VALUE)*(len - 1)); + *(ptr + len - 1) = tmp; + } + else if (cnt == len - 1) { + VALUE tmp = *(ptr + len - 1); + memmove(ptr + 1, ptr, sizeof(VALUE)*(len - 1)); + *ptr = tmp; + } + else { + --len; + if (cnt < len) ary_reverse(ptr + cnt, ptr + len); + if (--cnt > 0) ary_reverse(ptr, ptr + cnt); + if (len > 0) ary_reverse(ptr, ptr + len); + } +} + VALUE rb_ary_rotate(VALUE ary, long cnt) { rb_ary_modify(ary); if (cnt != 0) { - VALUE *ptr = RARRAY_PTR(ary); - long len = RARRAY_LEN(ary); - - if (len > 0 && (cnt = rotate_count(cnt, len)) > 0) { - --len; - if (cnt < len) ary_reverse(ptr + cnt, ptr + len); - if (--cnt > 0) ary_reverse(ptr, ptr + cnt); - if (len > 0) ary_reverse(ptr, ptr + len); - return ary; - } + long len = RARRAY_LEN(ary); + if (len > 1 && (cnt = rotate_count(cnt, len)) > 0) { + RARRAY_PTR_USE_TRANSIENT(ary, ptr, ary_rotate_ptr(ptr, len, cnt)); + return ary; + } } - return Qnil; } /* * call-seq: - * ary.rotate!(count=1) -> ary + * array.rotate! -> self + * array.rotate!(count) -> self * - * Rotates +self+ in place so that the element at +count+ comes first, and - * returns +self+. + * Rotates +self+ in place by moving elements from one end to the other; returns +self+. + * + * When no argument given, rotates the first element to the last position: + * + * a = [:foo, 'bar', 2, 'bar'] + * a.rotate! # => ["bar", 2, "bar", :foo] + * + * When given a non-negative \Integer +count+, + * rotates +count+ elements from the beginning to the end: * - * If +count+ is negative then it rotates in the opposite direction, starting - * from the end of the array where +-1+ is the last element. + * a = [:foo, 'bar', 2] + * a.rotate!(2) + * a # => [2, :foo, "bar"] + * + * If +count+ is large, uses <tt>count % array.size</tt> as the count: + * + * a = [:foo, 'bar', 2] + * a.rotate!(20) + * a # => [2, :foo, "bar"] + * + * If +count+ is zero, returns +self+ unmodified: + * + * a = [:foo, 'bar', 2] + * a.rotate!(0) + * a # => [:foo, "bar", 2] + * + * When given a negative Integer +count+, rotates in the opposite direction, + * from end to beginning: + * + * a = [:foo, 'bar', 2] + * a.rotate!(-2) + * a # => ["bar", 2, :foo] + * + * If +count+ is small (far from zero), uses <tt>count % array.size</tt> as the count: + * + * a = [:foo, 'bar', 2] + * a.rotate!(-5) + * a # => ["bar", 2, :foo] * - * a = [ "a", "b", "c", "d" ] - * a.rotate! #=> ["b", "c", "d", "a"] - * a #=> ["b", "c", "d", "a"] - * a.rotate!(2) #=> ["d", "a", "b", "c"] - * a.rotate!(-3) #=> ["a", "b", "c", "d"] */ static VALUE rb_ary_rotate_bang(int argc, VALUE *argv, VALUE ary) { - long n = 1; - - switch (argc) { - case 1: n = NUM2LONG(argv[0]); - case 0: break; - default: rb_scan_args(argc, argv, "01", NULL); - } + long n = (rb_check_arity(argc, 0, 1) ? NUM2LONG(argv[0]) : 1); rb_ary_rotate(ary, n); return ary; } /* * call-seq: - * ary.rotate(count=1) -> new_ary + * array.rotate -> new_array + * array.rotate(count) -> new_array + * + * Returns a new \Array formed from +self+ with elements + * rotated from one end to the other. * - * Returns a new array by rotating +self+ so that the element at +count+ is - * the first element of the new array. + * When no argument given, returns a new \Array that is like +self+, + * except that the first element has been rotated to the last position: * - * If +count+ is negative then it rotates in the opposite direction, starting - * from the end of +self+ where +-1+ is the last element. + * a = [:foo, 'bar', 2, 'bar'] + * a1 = a.rotate + * a1 # => ["bar", 2, "bar", :foo] + * + * When given a non-negative \Integer +count+, + * returns a new \Array with +count+ elements rotated from the beginning to the end: + * + * a = [:foo, 'bar', 2] + * a1 = a.rotate(2) + * a1 # => [2, :foo, "bar"] + * + * If +count+ is large, uses <tt>count % array.size</tt> as the count: + * + * a = [:foo, 'bar', 2] + * a1 = a.rotate(20) + * a1 # => [2, :foo, "bar"] + * + * If +count+ is zero, returns a copy of +self+, unmodified: + * + * a = [:foo, 'bar', 2] + * a1 = a.rotate(0) + * a1 # => [:foo, "bar", 2] + * + * When given a negative \Integer +count+, rotates in the opposite direction, + * from end to beginning: + * + * a = [:foo, 'bar', 2] + * a1 = a.rotate(-2) + * a1 # => ["bar", 2, :foo] + * + * If +count+ is small (far from zero), uses <tt>count % array.size</tt> as the count: + * + * a = [:foo, 'bar', 2] + * a1 = a.rotate(-5) + * a1 # => ["bar", 2, :foo] * - * a = [ "a", "b", "c", "d" ] - * a.rotate #=> ["b", "c", "d", "a"] - * a #=> ["a", "b", "c", "d"] - * a.rotate(2) #=> ["c", "d", "a", "b"] - * a.rotate(-3) #=> ["b", "c", "d", "a"] */ static VALUE @@ -2370,22 +3444,17 @@ rb_ary_rotate_m(int argc, VALUE *argv, VALUE ary) { VALUE rotated; const VALUE *ptr; - long len, cnt = 1; - - switch (argc) { - case 1: cnt = NUM2LONG(argv[0]); - case 0: break; - default: rb_scan_args(argc, argv, "01", NULL); - } + long len; + long cnt = (rb_check_arity(argc, 0, 1) ? NUM2LONG(argv[0]) : 1); len = RARRAY_LEN(ary); rotated = rb_ary_new2(len); if (len > 0) { - cnt = rotate_count(cnt, len); - ptr = RARRAY_CONST_PTR(ary); - len -= cnt; - ary_memcpy(rotated, 0, len, ptr + cnt); - ary_memcpy(rotated, len, cnt, ptr); + cnt = rotate_count(cnt, len); + ptr = RARRAY_CONST_PTR_TRANSIENT(ary); + len -= cnt; + ary_memcpy(rotated, 0, len, ptr + cnt); + ary_memcpy(rotated, len, cnt, ptr); } ARY_SET_LEN(rotated, RARRAY_LEN(ary)); return rotated; @@ -2393,18 +3462,27 @@ rb_ary_rotate_m(int argc, VALUE *argv, VALUE ary) struct ary_sort_data { VALUE ary; - struct cmp_opt_data cmp_opt; + VALUE receiver; }; static VALUE sort_reentered(VALUE ary) { if (RBASIC(ary)->klass) { - rb_raise(rb_eRuntimeError, "sort reentered"); + rb_raise(rb_eRuntimeError, "sort reentered"); } return Qnil; } +static void +sort_returned(struct ary_sort_data *data) +{ + if (rb_obj_frozen_p(data->receiver)) { + rb_raise(rb_eFrozenError, "array frozen during sort"); + } + sort_reentered(data->ary); +} + static int sort_1(const void *ap, const void *bp, void *dummy) { @@ -2418,7 +3496,7 @@ sort_1(const void *ap, const void *bp, void *dummy) args[1] = b; retval = rb_yield_values2(2, args); n = rb_cmpint(retval, a, b); - sort_reentered(data->ary); + sort_returned(data); return n; } @@ -2430,47 +3508,64 @@ sort_2(const void *ap, const void *bp, void *dummy) VALUE a = *(const VALUE *)ap, b = *(const VALUE *)bp; int n; - if (FIXNUM_P(a) && FIXNUM_P(b) && CMP_OPTIMIZABLE(data->cmp_opt, Fixnum)) { - if ((long)a > (long)b) return 1; - if ((long)a < (long)b) return -1; - return 0; + if (FIXNUM_P(a) && FIXNUM_P(b) && CMP_OPTIMIZABLE(INTEGER)) { + if ((long)a > (long)b) return 1; + if ((long)a < (long)b) return -1; + return 0; } - if (STRING_P(a) && STRING_P(b) && CMP_OPTIMIZABLE(data->cmp_opt, String)) { - return rb_str_cmp(a, b); + if (STRING_P(a) && STRING_P(b) && CMP_OPTIMIZABLE(STRING)) { + return rb_str_cmp(a, b); } - if (RB_FLOAT_TYPE_P(a) && CMP_OPTIMIZABLE(data->cmp_opt, Float)) { - return rb_float_cmp(a, b); + if (RB_FLOAT_TYPE_P(a) && CMP_OPTIMIZABLE(FLOAT)) { + return rb_float_cmp(a, b); } retval = rb_funcallv(a, id_cmp, 1, &b); n = rb_cmpint(retval, a, b); - sort_reentered(data->ary); + sort_returned(data); return n; } /* * call-seq: - * ary.sort! -> ary - * ary.sort! { |a, b| block } -> ary + * array.sort! -> self + * array.sort! {|a, b| ... } -> self + * + * Returns +self+ with its elements sorted in place. + * + * With no block, compares elements using operator <tt><=></tt> + * (see Comparable): + * + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a.sort! + * a # => ["a", "b", "c", "d", "e"] + * + * With a block, calls the block with each element pair; + * for each element pair +a+ and +b+, the block should return an integer: * - * Sorts +self+ in place. + * - Negative when +b+ is to follow +a+. + * - Zero when +a+ and +b+ are equivalent. + * - Positive when +a+ is to follow +b+. * - * Comparisons for the sort will be done using the <code><=></code> operator - * or using an optional code block. + * Example: * - * The block must implement a comparison between +a+ and +b+ and return - * an integer less than 0 when +b+ follows +a+, +0+ when +a+ and +b+ - * are equivalent, or an integer greater than 0 when +a+ follows +b+. + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a.sort! {|a, b| a <=> b } + * a # => ["a", "b", "c", "d", "e"] + * a.sort! {|a, b| b <=> a } + * a # => ["e", "d", "c", "b", "a"] * - * The result is not guaranteed to be stable. When the comparison of two - * elements returns +0+, the order of the elements is unpredictable. + * When the block returns zero, the order for +a+ and +b+ is indeterminate, + * and may be unstable: * - * ary = [ "d", "a", "e", "c", "b" ] - * ary.sort! #=> ["a", "b", "c", "d", "e"] - * ary.sort! { |a, b| b <=> a } #=> ["e", "d", "c", "b", "a"] + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a.sort! {|a, b| 0 } + * a # => ["d", "e", "c", "a", "b"] * - * See also Enumerable#sort_by. */ VALUE @@ -2479,25 +3574,23 @@ rb_ary_sort_bang(VALUE ary) rb_ary_modify(ary); assert(!ARY_SHARED_P(ary)); if (RARRAY_LEN(ary) > 1) { - VALUE tmp = ary_make_substitution(ary); /* only ary refers tmp */ - struct ary_sort_data data; - long len = RARRAY_LEN(ary); - - RBASIC_CLEAR_CLASS(tmp); - data.ary = tmp; - data.cmp_opt.opt_methods = 0; - data.cmp_opt.opt_inited = 0; - RARRAY_PTR_USE(tmp, ptr, { - ruby_qsort(ptr, len, sizeof(VALUE), - rb_block_given_p()?sort_1:sort_2, &data); - }); /* WB: no new reference */ - rb_ary_modify(ary); + VALUE tmp = ary_make_substitution(ary); /* only ary refers tmp */ + struct ary_sort_data data; + long len = RARRAY_LEN(ary); + RBASIC_CLEAR_CLASS(tmp); + data.ary = tmp; + data.receiver = ary; + RARRAY_PTR_USE(tmp, ptr, { + ruby_qsort(ptr, len, sizeof(VALUE), + rb_block_given_p()?sort_1:sort_2, &data); + }); /* WB: no new reference */ + rb_ary_modify(ary); if (ARY_EMBED_P(tmp)) { if (ARY_SHARED_P(ary)) { /* ary might be destructively operated in the given block */ rb_ary_unshare(ary); - FL_SET_EMBED(ary); + FL_SET_EMBED(ary); } - ary_memcpy(ary, 0, ARY_EMBED_LEN(tmp), ARY_EMBED_PTR(tmp)); + ary_memcpy(ary, 0, ARY_EMBED_LEN(tmp), ARY_EMBED_PTR(tmp)); ARY_SET_LEN(ary, ARY_EMBED_LEN(tmp)); } else { @@ -2515,46 +3608,65 @@ rb_ary_sort_bang(VALUE ary) rb_ary_unshare(ary); } else { - ruby_sized_xfree((void *)ARY_HEAP_PTR(ary), ARY_HEAP_SIZE(ary)); + ary_heap_free(ary); } - ARY_SET_PTR(ary, RARRAY_CONST_PTR(tmp)); + ARY_SET_PTR(ary, ARY_HEAP_PTR(tmp)); ARY_SET_HEAP_LEN(ary, len); - ARY_SET_CAPA(ary, RARRAY_LEN(tmp)); + ARY_SET_CAPA(ary, ARY_HEAP_LEN(tmp)); } /* tmp was lost ownership for the ptr */ FL_UNSET(tmp, FL_FREEZE); FL_SET_EMBED(tmp); ARY_SET_EMBED_LEN(tmp, 0); FL_SET(tmp, FL_FREEZE); - } + } /* tmp will be GC'ed. */ RBASIC_SET_CLASS_RAW(tmp, rb_cArray); /* rb_cArray must be marked */ } + ary_verify(ary); return ary; } /* * call-seq: - * ary.sort -> new_ary - * ary.sort { |a, b| block } -> new_ary + * array.sort -> new_array + * array.sort {|a, b| ... } -> new_array + * + * Returns a new \Array whose elements are those from +self+, sorted. + * + * With no block, compares elements using operator <tt><=></tt> + * (see Comparable): * - * Returns a new array created by sorting +self+. + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a1 = a.sort + * a1 # => ["a", "b", "c", "d", "e"] * - * Comparisons for the sort will be done using the <code><=></code> operator - * or using an optional code block. + * With a block, calls the block with each element pair; + * for each element pair +a+ and +b+, the block should return an integer: * - * The block must implement a comparison between +a+ and +b+ and return - * an integer less than 0 when +b+ follows +a+, +0+ when +a+ and +b+ - * are equivalent, or an integer greater than 0 when +a+ follows +b+. + * - Negative when +b+ is to follow +a+. + * - Zero when +a+ and +b+ are equivalent. + * - Positive when +a+ is to follow +b+. * - * The result is not guaranteed to be stable. When the comparison of two - * elements returns +0+, the order of the elements is unpredictable. + * Example: * - * ary = [ "d", "a", "e", "c", "b" ] - * ary.sort #=> ["a", "b", "c", "d", "e"] - * ary.sort { |a, b| b <=> a } #=> ["e", "d", "c", "b", "a"] + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a1 = a.sort {|a, b| a <=> b } + * a1 # => ["a", "b", "c", "d", "e"] + * a2 = a.sort {|a, b| b <=> a } + * a2 # => ["e", "d", "c", "b", "a"] * - * See also Enumerable#sort_by. + * When the block returns zero, the order for +a+ and +b+ is indeterminate, + * and may be unstable: + * + * a = 'abcde'.split('').shuffle + * a # => ["e", "b", "d", "a", "c"] + * a1 = a.sort {|a, b| 0 } + * a1 # => ["c", "e", "b", "d", "a"] + * + * Related: Enumerable#sort_by. */ VALUE @@ -2569,55 +3681,12 @@ static VALUE rb_ary_bsearch_index(VALUE ary); /* * call-seq: - * ary.bsearch {|x| block } -> elem - * - * By using binary search, finds a value from this array which meets - * the given condition in O(log n) where n is the size of the array. - * - * You can use this method in two modes: a find-minimum mode and - * a find-any mode. In either case, the elements of the array must be - * monotone (or sorted) with respect to the block. - * - * In find-minimum mode (this is a good choice for typical use cases), - * the block must always return true or false, and there must be an index i - * (0 <= i <= ary.size) so that: - * - * - the block returns false for any element whose index is less than - * i, and - * - the block returns true for any element whose index is greater - * than or equal to i. - * - * This method returns the i-th element. If i is equal to ary.size, - * it returns nil. - * - * ary = [0, 4, 7, 10, 12] - * ary.bsearch {|x| x >= 4 } #=> 4 - * ary.bsearch {|x| x >= 6 } #=> 7 - * ary.bsearch {|x| x >= -1 } #=> 0 - * ary.bsearch {|x| x >= 100 } #=> nil - * - * In find-any mode (this behaves like libc's bsearch(3)), the block - * must always return a number, and there must be two indices i and j - * (0 <= i <= j <= ary.size) so that: - * - * - the block returns a positive number for ary[k] if 0 <= k < i, - * - the block returns zero for ary[k] if i <= k < j, and - * - the block returns a negative number for ary[k] if - * j <= k < ary.size. - * - * Under this condition, this method returns any element whose index - * is within i...j. If i is equal to j (i.e., there is no element - * that satisfies the block), this method returns nil. - * - * ary = [0, 4, 7, 10, 12] - * # try to find v such that 4 <= v < 8 - * ary.bsearch {|x| 1 - x / 4 } #=> 4 or 7 - * # try to find v such that 8 <= v < 10 - * ary.bsearch {|x| 4 - x / 2 } #=> nil - * - * You must not mix the two modes at a time; the block must always - * return either true/false, or always return a number. It is - * undefined which value is actually picked up at each iteration. + * array.bsearch {|element| ... } -> object + * array.bsearch -> new_enumerator + * + * Returns an element from +self+ selected by a binary search. + * + * See {Binary Searching}[rdoc-ref:bsearch.rdoc]. */ static VALUE @@ -2626,22 +3695,18 @@ rb_ary_bsearch(VALUE ary) VALUE index_result = rb_ary_bsearch_index(ary); if (FIXNUM_P(index_result)) { - return rb_ary_entry(ary, FIX2LONG(index_result)); + return rb_ary_entry(ary, FIX2LONG(index_result)); } return index_result; } /* * call-seq: - * ary.bsearch_index {|x| block } -> int or nil - * - * By using binary search, finds an index of a value from this array which - * meets the given condition in O(log n) where n is the size of the array. + * array.bsearch_index {|element| ... } -> integer or nil + * array.bsearch_index -> new_enumerator * - * It supports two modes, depending on the nature of the block. They are - * exactly the same as in the case of the #bsearch method, with the only difference - * being that this method returns the index of the element instead of the - * element itself. For more details consult the documentation for #bsearch. + * Searches +self+ as described at method #bsearch, + * but returns the _index_ of the found element instead of the element itself. */ static VALUE @@ -2653,39 +3718,39 @@ rb_ary_bsearch_index(VALUE ary) RETURN_ENUMERATOR(ary, 0, 0); while (low < high) { - mid = low + ((high - low) / 2); - val = rb_ary_entry(ary, mid); - v = rb_yield(val); - if (FIXNUM_P(v)) { - if (v == INT2FIX(0)) return INT2FIX(mid); - smaller = (SIGNED_VALUE)v < 0; /* Fixnum preserves its sign-bit */ - } - else if (v == Qtrue) { - satisfied = 1; - smaller = 1; - } - else if (v == Qfalse || v == Qnil) { - smaller = 0; - } - else if (rb_obj_is_kind_of(v, rb_cNumeric)) { - const VALUE zero = INT2FIX(0); - switch (rb_cmpint(rb_funcallv(v, id_cmp, 1, &zero), v, zero)) { - case 0: return INT2FIX(mid); - case 1: smaller = 1; break; - case -1: smaller = 0; - } - } - else { - rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE - " (must be numeric, true, false or nil)", - rb_obj_class(v)); - } - if (smaller) { - high = mid; - } - else { - low = mid + 1; - } + mid = low + ((high - low) / 2); + val = rb_ary_entry(ary, mid); + v = rb_yield(val); + if (FIXNUM_P(v)) { + if (v == INT2FIX(0)) return INT2FIX(mid); + smaller = (SIGNED_VALUE)v < 0; /* Fixnum preserves its sign-bit */ + } + else if (v == Qtrue) { + satisfied = 1; + smaller = 1; + } + else if (!RTEST(v)) { + smaller = 0; + } + else if (rb_obj_is_kind_of(v, rb_cNumeric)) { + const VALUE zero = INT2FIX(0); + switch (rb_cmpint(rb_funcallv(v, id_cmp, 1, &zero), v, zero)) { + case 0: return INT2FIX(mid); + case 1: smaller = 0; break; + case -1: smaller = 1; + } + } + else { + rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE + " (must be numeric, true, false or nil)", + rb_obj_class(v)); + } + if (smaller) { + high = mid; + } + else { + low = mid + 1; + } } if (!satisfied) return Qnil; return INT2FIX(low); @@ -2700,18 +3765,28 @@ sort_by_i(RB_BLOCK_CALL_FUNC_ARGLIST(i, dummy)) /* * call-seq: - * ary.sort_by! { |obj| block } -> ary - * ary.sort_by! -> Enumerator + * array.sort_by! {|element| ... } -> self + * array.sort_by! -> new_enumerator * - * Sorts +self+ in place using a set of keys generated by mapping the - * values in +self+ through the given block. + * Sorts the elements of +self+ in place, + * using an ordering determined by the block; returns self. * - * The result is not guaranteed to be stable. When two keys are equal, - * the order of the corresponding elements is unpredictable. + * Calls the block with each successive element; + * sorts elements based on the values returned from the block. * - * If no block is given, an Enumerator is returned instead. + * For duplicates returned by the block, the ordering is indeterminate, and may be unstable. + * + * This example sorts strings based on their sizes: + * + * a = ['aaaa', 'bbb', 'cc', 'd'] + * a.sort_by! {|element| element.size } + * a # => ["d", "cc", "bbb", "aaaa"] + * + * Returns a new \Enumerator if no block given: + * + * a = ['aaaa', 'bbb', 'cc', 'd'] + * a.sort_by! # => #<Enumerator: ["aaaa", "bbb", "cc", "d"]:sort_by!> * - * See also Enumerable#sort_by. */ static VALUE @@ -2729,23 +3804,22 @@ rb_ary_sort_by_bang(VALUE ary) /* * call-seq: - * ary.collect { |item| block } -> new_ary - * ary.map { |item| block } -> new_ary - * ary.collect -> Enumerator - * ary.map -> Enumerator - * - * Invokes the given block once for each element of +self+. + * array.map {|element| ... } -> new_array + * array.map -> new_enumerator * - * Creates a new array containing the values returned by the block. + * Calls the block, if given, with each element of +self+; + * returns a new \Array whose elements are the return values from the block: * - * See also Enumerable#collect. + * a = [:foo, 'bar', 2] + * a1 = a.map {|element| element.class } + * a1 # => [Symbol, String, Integer] * - * If no block is given, an Enumerator is returned instead. + * Returns a new \Enumerator if no block given: + * a = [:foo, 'bar', 2] + * a1 = a.map + * a1 # => #<Enumerator: [:foo, "bar", 2]:map> * - * a = [ "a", "b", "c", "d" ] - * a.collect { |x| x + "!" } #=> ["a!", "b!", "c!", "d!"] - * a.map.with_index { |x, i| x * i } #=> ["", "b", "cc", "ddd"] - * a #=> ["a", "b", "c", "d"] + * Array#collect is an alias for Array#map. */ static VALUE @@ -2757,7 +3831,7 @@ rb_ary_collect(VALUE ary) RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); collect = rb_ary_new2(RARRAY_LEN(ary)); for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_ary_push(collect, rb_yield(RARRAY_AREF(ary, i))); + rb_ary_push(collect, rb_yield(RARRAY_AREF(ary, i))); } return collect; } @@ -2765,23 +3839,22 @@ rb_ary_collect(VALUE ary) /* * call-seq: - * ary.collect! {|item| block } -> ary - * ary.map! {|item| block } -> ary - * ary.collect! -> Enumerator - * ary.map! -> Enumerator + * array.map! {|element| ... } -> self + * array.map! -> new_enumerator + * + * Calls the block, if given, with each element; + * replaces the element with the block's return value: * - * Invokes the given block once for each element of +self+, replacing the - * element with the value returned by the block. + * a = [:foo, 'bar', 2] + * a.map! { |element| element.class } # => [Symbol, String, Integer] * - * See also Enumerable#collect. + * Returns a new \Enumerator if no block given: * - * If no block is given, an Enumerator is returned instead. + * a = [:foo, 'bar', 2] + * a1 = a.map! + * a1 # => #<Enumerator: [:foo, "bar", 2]:map!> * - * a = [ "a", "b", "c", "d" ] - * a.map! {|x| x + "!" } - * a #=> [ "a!", "b!", "c!", "d!" ] - * a.collect!.with_index {|x, i| x[0...i] } - * a #=> ["", "b", "c!", "d!"] + * Array#collect! is an alias for Array#map!. */ static VALUE @@ -2792,7 +3865,7 @@ rb_ary_collect_bang(VALUE ary) RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); rb_ary_modify(ary); for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_ary_store(ary, i, rb_yield(RARRAY_AREF(ary, i))); + rb_ary_store(ary, i, rb_yield(RARRAY_AREF(ary, i))); } return ary; } @@ -2804,66 +3877,128 @@ rb_get_values_at(VALUE obj, long olen, int argc, const VALUE *argv, VALUE (*func long beg, len, i, j; for (i=0; i<argc; i++) { - if (FIXNUM_P(argv[i])) { - rb_ary_push(result, (*func)(obj, FIX2LONG(argv[i]))); - continue; - } - /* check if idx is Range */ - if (rb_range_beg_len(argv[i], &beg, &len, olen, 1)) { - long end = olen < beg+len ? olen : beg+len; - for (j = beg; j < end; j++) { - rb_ary_push(result, (*func)(obj, j)); - } - if (beg + len > j) - rb_ary_resize(result, RARRAY_LEN(result) + (beg + len) - j); - continue; - } - rb_ary_push(result, (*func)(obj, NUM2LONG(argv[i]))); + if (FIXNUM_P(argv[i])) { + rb_ary_push(result, (*func)(obj, FIX2LONG(argv[i]))); + continue; + } + /* check if idx is Range */ + if (rb_range_beg_len(argv[i], &beg, &len, olen, 1)) { + long end = olen < beg+len ? olen : beg+len; + for (j = beg; j < end; j++) { + rb_ary_push(result, (*func)(obj, j)); + } + if (beg + len > j) + rb_ary_resize(result, RARRAY_LEN(result) + (beg + len) - j); + continue; + } + rb_ary_push(result, (*func)(obj, NUM2LONG(argv[i]))); } return result; } +static VALUE +append_values_at_single(VALUE result, VALUE ary, long olen, VALUE idx) +{ + long beg, len; + if (FIXNUM_P(idx)) { + beg = FIX2LONG(idx); + } + /* check if idx is Range */ + else if (rb_range_beg_len(idx, &beg, &len, olen, 1)) { + if (len > 0) { + const VALUE *const src = RARRAY_CONST_PTR_TRANSIENT(ary); + const long end = beg + len; + const long prevlen = RARRAY_LEN(result); + if (beg < olen) { + rb_ary_cat(result, src + beg, end > olen ? olen-beg : len); + } + if (end > olen) { + rb_ary_store(result, prevlen + len - 1, Qnil); + } + } + return result; + } + else { + beg = NUM2LONG(idx); + } + return rb_ary_push(result, rb_ary_entry(ary, beg)); +} + /* * call-seq: - * ary.values_at(selector, ...) -> new_ary + * array.values_at(*indexes) -> new_array + * + * Returns a new \Array whose elements are the elements + * of +self+ at the given \Integer or \Range +indexes+. + * + * For each positive +index+, returns the element at offset +index+: + * + * a = [:foo, 'bar', 2] + * a.values_at(0, 2) # => [:foo, 2] + * a.values_at(0..1) # => [:foo, "bar"] + * + * The given +indexes+ may be in any order, and may repeat: + * + * a = [:foo, 'bar', 2] + * a.values_at(2, 0, 1, 0, 2) # => [2, :foo, "bar", :foo, 2] + * a.values_at(1, 0..2) # => ["bar", :foo, "bar", 2] + * + * Assigns +nil+ for an +index+ that is too large: + * + * a = [:foo, 'bar', 2] + * a.values_at(0, 3, 1, 3) # => [:foo, nil, "bar", nil] + * + * Returns a new empty \Array if no arguments given. + * + * For each negative +index+, counts backward from the end of the array: * - * Returns an array containing the elements in +self+ corresponding to the - * given +selector+(s). + * a = [:foo, 'bar', 2] + * a.values_at(-1, -3) # => [2, :foo] * - * The selectors may be either integer indices or ranges. + * Assigns +nil+ for an +index+ that is too small: * - * See also Array#select. + * a = [:foo, 'bar', 2] + * a.values_at(0, -5, 1, -6, 2) # => [:foo, nil, "bar", nil, 2] + * + * The given +indexes+ may have a mixture of signs: + * + * a = [:foo, 'bar', 2] + * a.values_at(0, -2, 1, -1) # => [:foo, "bar", "bar", 2] * - * a = %w{ a b c d e f } - * a.values_at(1, 3, 5) # => ["b", "d", "f"] - * a.values_at(1, 3, 5, 7) # => ["b", "d", "f", nil] - * a.values_at(-1, -2, -2, -7) # => ["f", "e", "e", nil] - * a.values_at(4..6, 3...6) # => ["e", "f", nil, "d", "e", "f"] */ static VALUE rb_ary_values_at(int argc, VALUE *argv, VALUE ary) { - return rb_get_values_at(ary, RARRAY_LEN(ary), argc, argv, rb_ary_entry); + long i, olen = RARRAY_LEN(ary); + VALUE result = rb_ary_new_capa(argc); + for (i = 0; i < argc; ++i) { + append_values_at_single(result, ary, olen, argv[i]); + } + RB_GC_GUARD(ary); + return result; } /* * call-seq: - * ary.select { |item| block } -> new_ary - * ary.select -> Enumerator + * array.select {|element| ... } -> new_array + * array.select -> new_enumerator * - * Returns a new array containing all elements of +ary+ - * for which the given +block+ returns a true value. + * Calls the block, if given, with each element of +self+; + * returns a new \Array containing those elements of +self+ + * for which the block returns a truthy value: * - * If no block is given, an Enumerator is returned instead. + * a = [:foo, 'bar', 2, :bam] + * a1 = a.select {|element| element.to_s.start_with?('b') } + * a1 # => ["bar", :bam] * - * [1,2,3,4,5].select { |num| num.even? } #=> [2, 4] + * Returns a new \Enumerator if no block given: * - * a = %w{ a b c d e f } - * a.select { |v| v =~ /[aeiou]/ } #=> ["a", "e"] + * a = [:foo, 'bar', 2, :bam] + * a.select # => #<Enumerator: [:foo, "bar", 2, :bam]:select> * - * See also Enumerable#select. + * Array#filter is an alias for Array#select. */ static VALUE @@ -2875,9 +4010,9 @@ rb_ary_select(VALUE ary) RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); result = rb_ary_new2(RARRAY_LEN(ary)); for (i = 0; i < RARRAY_LEN(ary); i++) { - if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) { - rb_ary_push(result, rb_ary_elt(ary, i)); - } + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) { + rb_ary_push(result, rb_ary_elt(ary, i)); + } } return result; } @@ -2895,12 +4030,12 @@ select_bang_i(VALUE a) long i1, i2; for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); arg->len[0] = ++i1) { - VALUE v = RARRAY_AREF(ary, i1); - if (!RTEST(rb_yield(v))) continue; - if (i1 != i2) { - rb_ary_store(ary, i2, v); - } - arg->len[1] = ++i2; + VALUE v = RARRAY_AREF(ary, i1); + if (!RTEST(rb_yield(v))) continue; + if (i1 != i2) { + rb_ary_store(ary, i2, v); + } + arg->len[1] = ++i2; } return (i1 == i2) ? Qnil : ary; } @@ -2914,34 +4049,40 @@ select_bang_ensure(VALUE a) long i1 = arg->len[0], i2 = arg->len[1]; if (i2 < len && i2 < i1) { - long tail = 0; - if (i1 < len) { - tail = len - i1; - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr + i2, ptr + i1, VALUE, tail); - }); - } - ARY_SET_LEN(ary, i2 + tail); + long tail = 0; + rb_ary_modify(ary); + if (i1 < len) { + tail = len - i1; + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMMOVE(ptr + i2, ptr + i1, VALUE, tail); + }); + } + ARY_SET_LEN(ary, i2 + tail); } return ary; } /* * call-seq: - * ary.select! {|item| block } -> ary or nil - * ary.select! -> Enumerator + * array.select! {|element| ... } -> self or nil + * array.select! -> new_enumerator * - * Invokes the given block passing in successive elements from +self+, - * deleting elements for which the block returns a +false+ value. + * Calls the block, if given with each element of +self+; + * removes from +self+ those elements for which the block returns +false+ or +nil+. * - * The array may not be changed instantly every time the block is called. + * Returns +self+ if any elements were removed: * - * If changes were made, it will return +self+, otherwise it returns +nil+. + * a = [:foo, 'bar', 2, :bam] + * a.select! {|element| element.to_s.start_with?('b') } # => ["bar", :bam] * - * See also Array#keep_if + * Returns +nil+ if no elements were removed. * - * If no block is given, an Enumerator is returned instead. + * Returns a new \Enumerator if no block given: * + * a = [:foo, 'bar', 2, :bam] + * a.select! # => #<Enumerator: [:foo, "bar", 2, :bam]:select!> + * + * Array#filter! is an alias for Array#select!. */ static VALUE @@ -2959,18 +4100,20 @@ rb_ary_select_bang(VALUE ary) /* * call-seq: - * ary.keep_if { |item| block } -> ary - * ary.keep_if -> Enumerator + * array.keep_if {|element| ... } -> self + * array.keep_if -> new_enumeration + * + * Retains those elements for which the block returns a truthy value; + * deletes all other elements; returns +self+: * - * Deletes every element of +self+ for which the given block evaluates to - * +false+. + * a = [:foo, 'bar', 2, :bam] + * a.keep_if {|element| element.to_s.start_with?('b') } # => ["bar", :bam] * - * See also Array#select! + * Returns a new \Enumerator if no block given: * - * If no block is given, an Enumerator is returned instead. + * a = [:foo, 'bar', 2, :bam] + * a.keep_if # => #<Enumerator: [:foo, "bar", 2, :bam]:keep_if> * - * a = %w{ a b c d e f } - * a.keep_if { |v| v =~ /[aeiou]/ } #=> ["a", "e"] */ static VALUE @@ -2986,32 +4129,48 @@ ary_resize_smaller(VALUE ary, long len) { rb_ary_modify(ary); if (RARRAY_LEN(ary) > len) { - ARY_SET_LEN(ary, len); - if (len * 2 < ARY_CAPA(ary) && - ARY_CAPA(ary) > ARY_DEFAULT_SIZE) { - ary_resize_capa(ary, len * 2); - } + ARY_SET_LEN(ary, len); + if (len * 2 < ARY_CAPA(ary) && + ARY_CAPA(ary) > ARY_DEFAULT_SIZE) { + ary_resize_capa(ary, len * 2); + } } } /* * call-seq: - * ary.delete(obj) -> item or nil - * ary.delete(obj) { block } -> item or result of block + * array.delete(obj) -> deleted_object + * array.delete(obj) {|nosuch| ... } -> deleted_object or block_return + * + * Removes zero or more elements from +self+. + * + * When no block is given, + * removes from +self+ each element +ele+ such that <tt>ele == obj</tt>; + * returns the last deleted element: + * + * s1 = 'bar'; s2 = 'bar' + * a = [:foo, s1, 2, s2] + * a.delete('bar') # => "bar" + * a # => [:foo, 2] * - * Deletes all items from +self+ that are equal to +obj+. + * Returns +nil+ if no elements removed. * - * Returns the last deleted item, or +nil+ if no matching item is found. + * When a block is given, + * removes from +self+ each element +ele+ such that <tt>ele == obj</tt>. * - * If the optional code block is given, the result of the block is returned if - * the item is not found. (To remove +nil+ elements and get an informative - * return value, use Array#compact!) + * If any such elements are found, ignores the block + * and returns the last deleted element: + * + * s1 = 'bar'; s2 = 'bar' + * a = [:foo, s1, 2, s2] + * deleted_obj = a.delete('bar') {|obj| fail 'Cannot happen' } + * a # => [:foo, 2] + * + * If no such elements are found, returns the block's return value: + * + * a = [:foo, 'bar', 2] + * a.delete(:nosuch) {|obj| "#{obj} not found" } # => "nosuch not found" * - * a = [ "a", "b", "b", "b", "c" ] - * a.delete("b") #=> "b" - * a #=> ["a", "c"] - * a.delete("z") #=> nil - * a.delete("z") { "not found" } #=> "not found" */ VALUE @@ -3021,26 +4180,27 @@ rb_ary_delete(VALUE ary, VALUE item) long i1, i2; for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); i1++) { - VALUE e = RARRAY_AREF(ary, i1); + VALUE e = RARRAY_AREF(ary, i1); - if (rb_equal(e, item)) { - v = e; - continue; - } - if (i1 != i2) { - rb_ary_store(ary, i2, e); - } - i2++; + if (rb_equal(e, item)) { + v = e; + continue; + } + if (i1 != i2) { + rb_ary_store(ary, i2, e); + } + i2++; } if (RARRAY_LEN(ary) == i2) { - if (rb_block_given_p()) { - return rb_yield(item); - } - return Qnil; + if (rb_block_given_p()) { + return rb_yield(item); + } + return Qnil; } ary_resize_smaller(ary, i2); + ary_verify(ary); return v; } @@ -3050,18 +4210,18 @@ rb_ary_delete_same(VALUE ary, VALUE item) long i1, i2; for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); i1++) { - VALUE e = RARRAY_AREF(ary, i1); + VALUE e = RARRAY_AREF(ary, i1); - if (e == item) { - continue; - } - if (i1 != i2) { - rb_ary_store(ary, i2, e); - } - i2++; + if (e == item) { + continue; + } + if (i1 != i2) { + rb_ary_store(ary, i2, e); + } + i2++; } if (RARRAY_LEN(ary) == i2) { - return; + return; } ary_resize_smaller(ary, i2); @@ -3075,33 +4235,41 @@ rb_ary_delete_at(VALUE ary, long pos) if (pos >= len) return Qnil; if (pos < 0) { - pos += len; - if (pos < 0) return Qnil; + pos += len; + if (pos < 0) return Qnil; } rb_ary_modify(ary); del = RARRAY_AREF(ary, pos); - RARRAY_PTR_USE(ary, ptr, { - MEMMOVE(ptr+pos, ptr+pos+1, VALUE, len-pos-1); + RARRAY_PTR_USE_TRANSIENT(ary, ptr, { + MEMMOVE(ptr+pos, ptr+pos+1, VALUE, len-pos-1); }); ARY_INCREASE_LEN(ary, -1); - + ary_verify(ary); return del; } /* * call-seq: - * ary.delete_at(index) -> obj or nil + * array.delete_at(index) -> deleted_object or nil + * + * Deletes an element from +self+, per the given \Integer +index+. * - * Deletes the element at the specified +index+, returning that element, or - * +nil+ if the +index+ is out of range. + * When +index+ is non-negative, deletes the element at offset +index+: * - * See also Array#slice! + * a = [:foo, 'bar', 2] + * a.delete_at(1) # => "bar" + * a # => [:foo, 2] * - * a = ["ant", "bat", "cat", "dog"] - * a.delete_at(2) #=> "cat" - * a #=> ["ant", "bat", "dog"] - * a.delete_at(99) #=> nil + * If index is too large, returns +nil+. + * + * When +index+ is negative, counts backward from the end of the array: + * + * a = [:foo, 'bar', 2] + * a.delete_at(-2) # => "bar" + * a # => [:foo, 2] + * + * If +index+ is too small (far from zero), returns nil. */ static VALUE @@ -3110,73 +4278,133 @@ rb_ary_delete_at_m(VALUE ary, VALUE pos) return rb_ary_delete_at(ary, NUM2LONG(pos)); } +static VALUE +ary_slice_bang_by_rb_ary_splice(VALUE ary, long pos, long len) +{ + const long orig_len = RARRAY_LEN(ary); + + if (len < 0) { + return Qnil; + } + else if (pos < -orig_len) { + return Qnil; + } + else if (pos < 0) { + pos += orig_len; + } + else if (orig_len < pos) { + return Qnil; + } + if (orig_len < pos + len) { + len = orig_len - pos; + } + if (len == 0) { + return rb_ary_new2(0); + } + else { + VALUE arg2 = rb_ary_new4(len, RARRAY_CONST_PTR_TRANSIENT(ary)+pos); + rb_ary_splice(ary, pos, len, 0, 0); + return arg2; + } +} + /* * call-seq: - * ary.slice!(index) -> obj or nil - * ary.slice!(start, length) -> new_ary or nil - * ary.slice!(range) -> new_ary or nil - * - * Deletes the element(s) given by an +index+ (optionally up to +length+ - * elements) or by a +range+. - * - * Returns the deleted object (or objects), or +nil+ if the +index+ is out of - * range. - * - * a = [ "a", "b", "c" ] - * a.slice!(1) #=> "b" - * a #=> ["a", "c"] - * a.slice!(-1) #=> "c" - * a #=> ["a"] - * a.slice!(100) #=> nil - * a #=> ["a"] + * array.slice!(n) -> object or nil + * array.slice!(start, length) -> new_array or nil + * array.slice!(range) -> new_array or nil + * + * Removes and returns elements from +self+. + * + * When the only argument is an \Integer +n+, + * removes and returns the _nth_ element in +self+: + * + * a = [:foo, 'bar', 2] + * a.slice!(1) # => "bar" + * a # => [:foo, 2] + * + * If +n+ is negative, counts backwards from the end of +self+: + * + * a = [:foo, 'bar', 2] + * a.slice!(-1) # => 2 + * a # => [:foo, "bar"] + * + * If +n+ is out of range, returns +nil+. + * + * When the only arguments are Integers +start+ and +length+, + * removes +length+ elements from +self+ beginning at offset +start+; + * returns the deleted objects in a new \Array: + * + * a = [:foo, 'bar', 2] + * a.slice!(0, 2) # => [:foo, "bar"] + * a # => [2] + * + * If <tt>start + length</tt> exceeds the array size, + * removes and returns all elements from offset +start+ to the end: + * + * a = [:foo, 'bar', 2] + * a.slice!(1, 50) # => ["bar", 2] + * a # => [:foo] + * + * If <tt>start == a.size</tt> and +length+ is non-negative, + * returns a new empty \Array. + * + * If +length+ is negative, returns +nil+. + * + * When the only argument is a \Range object +range+, + * treats <tt>range.min</tt> as +start+ above and <tt>range.size</tt> as +length+ above: + * + * a = [:foo, 'bar', 2] + * a.slice!(1..2) # => ["bar", 2] + * a # => [:foo] + * + * If <tt>range.start == a.size</tt>, returns a new empty \Array. + * + * If <tt>range.start</tt> is larger than the array size, returns +nil+. + * + * If <tt>range.end</tt> is negative, counts backwards from the end of the array: + * + * a = [:foo, 'bar', 2] + * a.slice!(0..-2) # => [:foo, "bar"] + * a # => [2] + * + * If <tt>range.start</tt> is negative, + * calculates the start index backwards from the end of the array: + * + * a = [:foo, 'bar', 2] + * a.slice!(-2..2) # => ["bar", 2] + * a # => [:foo] + * */ static VALUE rb_ary_slice_bang(int argc, VALUE *argv, VALUE ary) { - VALUE arg1, arg2; - long pos, len, orig_len; + VALUE arg1; + long pos, len; rb_ary_modify_check(ary); + rb_check_arity(argc, 1, 2); + arg1 = argv[0]; + if (argc == 2) { - pos = NUM2LONG(argv[0]); - len = NUM2LONG(argv[1]); - delete_pos_len: - if (len < 0) return Qnil; - orig_len = RARRAY_LEN(ary); - if (pos < 0) { - pos += orig_len; - if (pos < 0) return Qnil; - } - else if (orig_len < pos) return Qnil; - if (orig_len < pos + len) { - len = orig_len - pos; - } - if (len == 0) return rb_ary_new2(0); - arg2 = rb_ary_new4(len, RARRAY_CONST_PTR(ary)+pos); - RBASIC_SET_CLASS(arg2, rb_obj_class(ary)); - rb_ary_splice(ary, pos, len, 0, 0); - return arg2; - } - - if (argc != 1) { - /* error report */ - rb_scan_args(argc, argv, "11", NULL, NULL); + pos = NUM2LONG(argv[0]); + len = NUM2LONG(argv[1]); + return ary_slice_bang_by_rb_ary_splice(ary, pos, len); } - arg1 = argv[0]; if (!FIXNUM_P(arg1)) { - switch (rb_range_beg_len(arg1, &pos, &len, RARRAY_LEN(ary), 0)) { - case Qtrue: - /* valid range */ - goto delete_pos_len; - case Qnil: - /* invalid range */ - return Qnil; - default: - /* not a range */ - break; - } + switch (rb_range_beg_len(arg1, &pos, &len, RARRAY_LEN(ary), 0)) { + case Qtrue: + /* valid range */ + return ary_slice_bang_by_rb_ary_splice(ary, pos, len); + case Qnil: + /* invalid range */ + return Qnil; + default: + /* not a range */ + break; + } } return rb_ary_delete_at(ary, NUM2LONG(arg1)); @@ -3188,10 +4416,11 @@ ary_reject(VALUE orig, VALUE result) long i; for (i = 0; i < RARRAY_LEN(orig); i++) { - VALUE v = RARRAY_AREF(orig, i); - if (!RTEST(rb_yield(v))) { - rb_ary_push(result, v); - } + VALUE v = RARRAY_AREF(orig, i); + + if (!RTEST(rb_yield(v))) { + rb_ary_push(result, v); + } } return result; } @@ -3204,12 +4433,12 @@ reject_bang_i(VALUE a) long i1, i2; for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); arg->len[0] = ++i1) { - VALUE v = RARRAY_AREF(ary, i1); - if (RTEST(rb_yield(v))) continue; - if (i1 != i2) { - rb_ary_store(ary, i2, v); - } - arg->len[1] = ++i2; + VALUE v = RARRAY_AREF(ary, i1); + if (RTEST(rb_yield(v))) continue; + if (i1 != i2) { + rb_ary_store(ary, i2, v); + } + arg->len[1] = ++i2; } return (i1 == i2) ? Qnil : ary; } @@ -3218,7 +4447,6 @@ static VALUE ary_reject_bang(VALUE ary) { struct select_bang_arg args; - rb_ary_modify_check(ary); args.ary = ary; args.len[0] = args.len[1] = 0; @@ -3227,37 +4455,50 @@ ary_reject_bang(VALUE ary) /* * call-seq: - * ary.reject! { |item| block } -> ary or nil - * ary.reject! -> Enumerator + * array.reject! {|element| ... } -> self or nil + * array.reject! -> new_enumerator * - * Deletes every element of +self+ for which the block evaluates to +true+, - * if no changes were made returns +nil+. + * Removes each element for which the block returns a truthy value. * - * The array may not be changed instantly every time the block is called. + * Returns +self+ if any elements removed: * - * See also Enumerable#reject and Array#delete_if. + * a = [:foo, 'bar', 2, 'bat'] + * a.reject! {|element| element.to_s.start_with?('b') } # => [:foo, 2] + * + * Returns +nil+ if no elements removed. + * + * Returns a new \Enumerator if no block given: + * + * a = [:foo, 'bar', 2] + * a.reject! # => #<Enumerator: [:foo, "bar", 2]:reject!> * - * If no block is given, an Enumerator is returned instead. */ static VALUE rb_ary_reject_bang(VALUE ary) { RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); + rb_ary_modify(ary); return ary_reject_bang(ary); } /* * call-seq: - * ary.reject {|item| block } -> new_ary - * ary.reject -> Enumerator + * array.reject {|element| ... } -> new_array + * array.reject -> new_enumerator * - * Returns a new array containing the items in +self+ for which the given - * block is not +true+. The ordering of non-rejected elements is maintained. + * Returns a new \Array whose elements are all those from +self+ + * for which the block returns +false+ or +nil+: * - * See also Array#delete_if + * a = [:foo, 'bar', 2, 'bat'] + * a1 = a.reject {|element| element.to_s.start_with?('b') } + * a1 # => [:foo, 2] + * + * Returns a new \Enumerator if no block given: + * + * a = [:foo, 'bar', 2] + * a.reject # => #<Enumerator: [:foo, "bar", 2]:reject> * - * If no block is given, an Enumerator is returned instead. */ static VALUE @@ -3273,25 +4514,26 @@ rb_ary_reject(VALUE ary) /* * call-seq: - * ary.delete_if { |item| block } -> ary - * ary.delete_if -> Enumerator + * array.delete_if {|element| ... } -> self + * array.delete_if -> Enumerator * - * Deletes every element of +self+ for which block evaluates to +true+. + * Removes each element in +self+ for which the block returns a truthy value; + * returns +self+: * - * The array is changed instantly every time the block is called, not after - * the iteration is over. + * a = [:foo, 'bar', 2, 'bat'] + * a.delete_if {|element| element.to_s.start_with?('b') } # => [:foo, 2] * - * See also Array#reject! + * Returns a new \Enumerator if no block given: * - * If no block is given, an Enumerator is returned instead. + * a = [:foo, 'bar', 2] + * a.delete_if # => #<Enumerator: [:foo, "bar", 2]:delete_if> * - * scores = [ 97, 42, 75 ] - * scores.delete_if {|score| score < 80 } #=> [97] - */ +3 */ static VALUE rb_ary_delete_if(VALUE ary) { + ary_verify(ary); RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length); ary_reject_bang(ary); return ary; @@ -3301,9 +4543,9 @@ static VALUE take_i(RB_BLOCK_CALL_FUNC_ARGLIST(val, cbarg)) { VALUE *args = (VALUE *)cbarg; - if (args[1]-- == 0) rb_iter_break(); if (argc > 1) val = rb_ary_new4(argc, argv); rb_ary_push(args[0], val); + if (--args[1] == 0) rb_iter_break(); return Qnil; } @@ -3313,38 +4555,71 @@ take_items(VALUE obj, long n) VALUE result = rb_check_array_type(obj); VALUE args[2]; + if (n == 0) return result; if (!NIL_P(result)) return rb_ary_subseq(result, 0, n); result = rb_ary_new2(n); args[0] = result; args[1] = (VALUE)n; - if (rb_check_block_call(obj, idEach, 0, 0, take_i, (VALUE)args) == Qundef) - rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (must respond to :each)", - rb_obj_class(obj)); + if (UNDEF_P(rb_check_block_call(obj, idEach, 0, 0, take_i, (VALUE)args))) + rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (must respond to :each)", + rb_obj_class(obj)); return result; } /* * call-seq: - * ary.zip(arg, ...) -> new_ary - * ary.zip(arg, ...) { |arr| block } -> nil + * array.zip(*other_arrays) -> new_array + * array.zip(*other_arrays) {|other_array| ... } -> nil + * + * When no block given, returns a new \Array +new_array+ of size <tt>self.size</tt> + * whose elements are Arrays. + * + * Each nested array <tt>new_array[n]</tt> is of size <tt>other_arrays.size+1</tt>, + * and contains: + * + * - The _nth_ element of +self+. + * - The _nth_ element of each of the +other_arrays+. + * + * If all +other_arrays+ and +self+ are the same size: + * + * a = [:a0, :a1, :a2, :a3] + * b = [:b0, :b1, :b2, :b3] + * c = [:c0, :c1, :c2, :c3] + * d = a.zip(b, c) + * d # => [[:a0, :b0, :c0], [:a1, :b1, :c1], [:a2, :b2, :c2], [:a3, :b3, :c3]] + * + * If any array in +other_arrays+ is smaller than +self+, + * fills to <tt>self.size</tt> with +nil+: * - * Converts any arguments to arrays, then merges elements of +self+ with - * corresponding elements from each argument. + * a = [:a0, :a1, :a2, :a3] + * b = [:b0, :b1, :b2] + * c = [:c0, :c1] + * d = a.zip(b, c) + * d # => [[:a0, :b0, :c0], [:a1, :b1, :c1], [:a2, :b2, nil], [:a3, nil, nil]] * - * This generates a sequence of <code>ary.size</code> _n_-element arrays, - * where _n_ is one more than the count of arguments. + * If any array in +other_arrays+ is larger than +self+, + * its trailing elements are ignored: * - * If the size of any argument is less than the size of the initial array, - * +nil+ values are supplied. + * a = [:a0, :a1, :a2, :a3] + * b = [:b0, :b1, :b2, :b3, :b4] + * c = [:c0, :c1, :c2, :c3, :c4, :c5] + * d = a.zip(b, c) + * d # => [[:a0, :b0, :c0], [:a1, :b1, :c1], [:a2, :b2, :c2], [:a3, :b3, :c3]] * - * If a block is given, it is invoked for each output +array+, otherwise an - * array of arrays is returned. + * When a block is given, calls the block with each of the sub-arrays (formed as above); returns +nil+: + * + * a = [:a0, :a1, :a2, :a3] + * b = [:b0, :b1, :b2, :b3] + * c = [:c0, :c1, :c2, :c3] + * a.zip(b, c) {|sub_array| p sub_array} # => nil + * + * Output: + * + * [:a0, :b0, :c0] + * [:a1, :b1, :c1] + * [:a2, :b2, :c2] + * [:a3, :b3, :c3] * - * a = [ 4, 5, 6 ] - * b = [ 7, 8, 9 ] - * [1, 2, 3].zip(a, b) #=> [[1, 4, 7], [2, 5, 8], [3, 6, 9]] - * [1, 2].zip(a, b) #=> [[1, 4, 7], [2, 5, 8]] - * a.zip([1, 2], [8]) #=> [[4, 1, 8], [5, 2, nil], [6, nil, nil]] */ static VALUE @@ -3355,51 +4630,51 @@ rb_ary_zip(int argc, VALUE *argv, VALUE ary) VALUE result = Qnil; for (i=0; i<argc; i++) { - argv[i] = take_items(argv[i], len); + argv[i] = take_items(argv[i], len); } if (rb_block_given_p()) { - int arity = rb_block_arity(); - - if (arity > 1) { - VALUE work, *tmp; - - tmp = ALLOCV_N(VALUE, work, argc+1); - - for (i=0; i<RARRAY_LEN(ary); i++) { - tmp[0] = RARRAY_AREF(ary, i); - for (j=0; j<argc; j++) { - tmp[j+1] = rb_ary_elt(argv[j], i); - } - rb_yield_values2(argc+1, tmp); - } - - if (work) ALLOCV_END(work); - } - else { - for (i=0; i<RARRAY_LEN(ary); i++) { - VALUE tmp = rb_ary_new2(argc+1); - - rb_ary_push(tmp, RARRAY_AREF(ary, i)); - for (j=0; j<argc; j++) { - rb_ary_push(tmp, rb_ary_elt(argv[j], i)); - } - rb_yield(tmp); - } - } + int arity = rb_block_arity(); + + if (arity > 1) { + VALUE work, *tmp; + + tmp = ALLOCV_N(VALUE, work, argc+1); + + for (i=0; i<RARRAY_LEN(ary); i++) { + tmp[0] = RARRAY_AREF(ary, i); + for (j=0; j<argc; j++) { + tmp[j+1] = rb_ary_elt(argv[j], i); + } + rb_yield_values2(argc+1, tmp); + } + + if (work) ALLOCV_END(work); + } + else { + for (i=0; i<RARRAY_LEN(ary); i++) { + VALUE tmp = rb_ary_new2(argc+1); + + rb_ary_push(tmp, RARRAY_AREF(ary, i)); + for (j=0; j<argc; j++) { + rb_ary_push(tmp, rb_ary_elt(argv[j], i)); + } + rb_yield(tmp); + } + } } else { - result = rb_ary_new_capa(len); + result = rb_ary_new_capa(len); - for (i=0; i<len; i++) { - VALUE tmp = rb_ary_new_capa(argc+1); + for (i=0; i<len; i++) { + VALUE tmp = rb_ary_new_capa(argc+1); - rb_ary_push(tmp, RARRAY_AREF(ary, i)); - for (j=0; j<argc; j++) { - rb_ary_push(tmp, rb_ary_elt(argv[j], i)); - } - rb_ary_push(result, tmp); - } + rb_ary_push(tmp, RARRAY_AREF(ary, i)); + for (j=0; j<argc; j++) { + rb_ary_push(tmp, rb_ary_elt(argv[j], i)); + } + rb_ary_push(result, tmp); + } } return result; @@ -3407,15 +4682,14 @@ rb_ary_zip(int argc, VALUE *argv, VALUE ary) /* * call-seq: - * ary.transpose -> new_ary + * array.transpose -> new_array * - * Assumes that +self+ is an array of arrays and transposes the rows and - * columns. + * Transposes the rows and columns in an \Array of Arrays; + * the nested Arrays must all be the same size: * - * a = [[1,2], [3,4], [5,6]] - * a.transpose #=> [[1, 3, 5], [2, 4, 6]] + * a = [[:a0, :a1], [:b0, :b1], [:c0, :c1]] + * a.transpose # => [[:a0, :b0, :c0], [:a1, :b1, :c1]] * - * If the length of the subarrays don't match, an IndexError is raised. */ static VALUE @@ -3427,36 +4701,34 @@ rb_ary_transpose(VALUE ary) alen = RARRAY_LEN(ary); if (alen == 0) return rb_ary_dup(ary); for (i=0; i<alen; i++) { - tmp = to_ary(rb_ary_elt(ary, i)); - if (elen < 0) { /* first element */ - elen = RARRAY_LEN(tmp); - result = rb_ary_new2(elen); - for (j=0; j<elen; j++) { - rb_ary_store(result, j, rb_ary_new2(alen)); - } - } - else if (elen != RARRAY_LEN(tmp)) { - rb_raise(rb_eIndexError, "element size differs (%ld should be %ld)", - RARRAY_LEN(tmp), elen); - } - for (j=0; j<elen; j++) { - rb_ary_store(rb_ary_elt(result, j), i, rb_ary_elt(tmp, j)); - } + tmp = to_ary(rb_ary_elt(ary, i)); + if (elen < 0) { /* first element */ + elen = RARRAY_LEN(tmp); + result = rb_ary_new2(elen); + for (j=0; j<elen; j++) { + rb_ary_store(result, j, rb_ary_new2(alen)); + } + } + else if (elen != RARRAY_LEN(tmp)) { + rb_raise(rb_eIndexError, "element size differs (%ld should be %ld)", + RARRAY_LEN(tmp), elen); + } + for (j=0; j<elen; j++) { + rb_ary_store(rb_ary_elt(result, j), i, rb_ary_elt(tmp, j)); + } } return result; } /* * call-seq: - * ary.replace(other_ary) -> ary - * ary.initialize_copy(other_ary) -> ary + * array.replace(other_array) -> self + * + * Replaces the content of +self+ with the content of +other_array+; returns +self+: * - * Replaces the contents of +self+ with the contents of +other_ary+, - * truncating or expanding if necessary. + * a = [:foo, 'bar', 2] + * a.replace(['foo', :bar, 3]) # => ["foo", :bar, 3] * - * a = [ "a", "b", "c", "d", "e" ] - * a.replace([ "x", "y", "z" ]) #=> ["x", "y", "z"] - * a #=> ["x", "y", "z"] */ VALUE @@ -3466,94 +4738,270 @@ rb_ary_replace(VALUE copy, VALUE orig) orig = to_ary(orig); if (copy == orig) return copy; - if (RARRAY_LEN(orig) <= RARRAY_EMBED_LEN_MAX) { - VALUE shared = 0; + rb_ary_reset(copy); - if (ARY_OWNS_HEAP_P(copy)) { - RARRAY_PTR_USE(copy, ptr, ruby_sized_xfree(ptr, ARY_HEAP_SIZE(copy))); - } - else if (ARY_SHARED_P(copy)) { - shared = ARY_SHARED(copy); - FL_UNSET_SHARED(copy); - } - FL_SET_EMBED(copy); - ary_memcpy(copy, 0, RARRAY_LEN(orig), RARRAY_CONST_PTR(orig)); - if (shared) { - rb_ary_decrement_share(shared); - } - ARY_SET_LEN(copy, RARRAY_LEN(orig)); + /* orig has enough space to embed the contents of orig. */ + if (RARRAY_LEN(orig) <= ary_embed_capa(copy)) { + assert(ARY_EMBED_P(copy)); + ary_memcpy(copy, 0, RARRAY_LEN(orig), RARRAY_CONST_PTR_TRANSIENT(orig)); + ARY_SET_EMBED_LEN(copy, RARRAY_LEN(orig)); } +#if USE_RVARGC + /* orig is embedded but copy does not have enough space to embed the + * contents of orig. */ + else if (ARY_EMBED_P(orig)) { + long len = ARY_EMBED_LEN(orig); + VALUE *ptr = ary_heap_alloc(copy, len); + + FL_UNSET_EMBED(copy); + ARY_SET_PTR(copy, ptr); + ARY_SET_LEN(copy, len); + ARY_SET_CAPA(copy, len); + + // No allocation and exception expected that could leave `copy` in a + // bad state from the edits above. + ary_memcpy(copy, 0, len, RARRAY_CONST_PTR_TRANSIENT(orig)); + } +#endif + /* Otherwise, orig is on heap and copy does not have enough space to embed + * the contents of orig. */ else { - VALUE shared = ary_make_shared(orig); - if (ARY_OWNS_HEAP_P(copy)) { - RARRAY_PTR_USE(copy, ptr, ruby_sized_xfree(ptr, ARY_HEAP_SIZE(copy))); - } - else { - rb_ary_unshare_safe(copy); - } + VALUE shared_root = ary_make_shared(orig); FL_UNSET_EMBED(copy); - ARY_SET_PTR(copy, RARRAY_CONST_PTR(orig)); - ARY_SET_LEN(copy, RARRAY_LEN(orig)); - rb_ary_set_shared(copy, shared); + ARY_SET_PTR(copy, ARY_HEAP_PTR(orig)); + ARY_SET_LEN(copy, ARY_HEAP_LEN(orig)); + rb_ary_set_shared(copy, shared_root); } + ary_verify(copy); return copy; } /* * call-seq: - * ary.clear -> ary + * array.clear -> self * - * Removes all elements from +self+. + * Removes all elements from +self+: + * + * a = [:foo, 'bar', 2] + * a.clear # => [] * - * a = [ "a", "b", "c", "d", "e" ] - * a.clear #=> [ ] */ VALUE rb_ary_clear(VALUE ary) { rb_ary_modify_check(ary); - ARY_SET_LEN(ary, 0); if (ARY_SHARED_P(ary)) { - if (!ARY_EMBED_P(ary)) { - rb_ary_unshare(ary); - FL_SET_EMBED(ary); - } + if (!ARY_EMBED_P(ary)) { + rb_ary_unshare(ary); + FL_SET_EMBED(ary); + ARY_SET_EMBED_LEN(ary, 0); + } } - else if (ARY_DEFAULT_SIZE * 2 < ARY_CAPA(ary)) { - ary_resize_capa(ary, ARY_DEFAULT_SIZE * 2); + else { + ARY_SET_LEN(ary, 0); + if (ARY_DEFAULT_SIZE * 2 < ARY_CAPA(ary)) { + ary_resize_capa(ary, ARY_DEFAULT_SIZE * 2); + } } + ary_verify(ary); return ary; } /* * call-seq: - * ary.fill(obj) -> ary - * ary.fill(obj, start [, length]) -> ary - * ary.fill(obj, range ) -> ary - * ary.fill { |index| block } -> ary - * ary.fill(start [, length] ) { |index| block } -> ary - * ary.fill(range) { |index| block } -> ary + * array.fill(obj) -> self + * array.fill(obj, start) -> self + * array.fill(obj, start, length) -> self + * array.fill(obj, range) -> self + * array.fill {|index| ... } -> self + * array.fill(start) {|index| ... } -> self + * array.fill(start, length) {|index| ... } -> self + * array.fill(range) {|index| ... } -> self + * + * Replaces specified elements in +self+ with specified objects; returns +self+. + * + * With argument +obj+ and no block given, replaces all elements with that one object: + * + * a = ['a', 'b', 'c', 'd'] + * a # => ["a", "b", "c", "d"] + * a.fill(:X) # => [:X, :X, :X, :X] + * + * With arguments +obj+ and \Integer +start+, and no block given, + * replaces elements based on the given start. + * + * If +start+ is in range (<tt>0 <= start < array.size</tt>), + * replaces all elements from offset +start+ through the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 2) # => ["a", "b", :X, :X] + * + * If +start+ is too large (<tt>start >= array.size</tt>), does nothing: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 4) # => ["a", "b", "c", "d"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 5) # => ["a", "b", "c", "d"] + * + * If +start+ is negative, counts from the end (starting index is <tt>start + array.size</tt>): + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, -2) # => ["a", "b", :X, :X] + * + * If +start+ is too small (less than and far from zero), replaces all elements: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, -6) # => [:X, :X, :X, :X] + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, -50) # => [:X, :X, :X, :X] + * + * With arguments +obj+, \Integer +start+, and \Integer +length+, and no block given, + * replaces elements based on the given +start+ and +length+. + * + * If +start+ is in range, replaces +length+ elements beginning at offset +start+: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 1, 1) # => ["a", :X, "c", "d"] + * + * If +start+ is negative, counts from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, -2, 1) # => ["a", "b", :X, "d"] + * + * If +start+ is large (<tt>start >= array.size</tt>), extends +self+ with +nil+: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 5, 0) # => ["a", "b", "c", "d", nil] + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 5, 2) # => ["a", "b", "c", "d", nil, :X, :X] + * + * If +length+ is zero or negative, replaces no elements: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, 1, 0) # => ["a", "b", "c", "d"] + * a.fill(:X, 1, -1) # => ["a", "b", "c", "d"] + * + * With arguments +obj+ and \Range +range+, and no block given, + * replaces elements based on the given range. + * + * If the range is positive and ascending (<tt>0 < range.begin <= range.end</tt>), + * replaces elements from <tt>range.begin</tt> to <tt>range.end</tt>: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (1..1)) # => ["a", :X, "c", "d"] + * + * If <tt>range.first</tt> is negative, replaces no elements: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (-1..1)) # => ["a", "b", "c", "d"] + * + * If <tt>range.last</tt> is negative, counts from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (0..-2)) # => [:X, :X, :X, "d"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (1..-2)) # => ["a", :X, :X, "d"] + * + * If <tt>range.last</tt> and <tt>range.last</tt> are both negative, + * both count from the end of the array: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (-1..-1)) # => ["a", "b", "c", :X] + * a = ['a', 'b', 'c', 'd'] + * a.fill(:X, (-2..-2)) # => ["a", "b", :X, "d"] + * + * With no arguments and a block given, calls the block with each index; + * replaces the corresponding element with the block's return value: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill { |index| "new_#{index}" } # => ["new_0", "new_1", "new_2", "new_3"] + * + * With argument +start+ and a block given, calls the block with each index + * from offset +start+ to the end; replaces the corresponding element + * with the block's return value. + * + * If start is in range (<tt>0 <= start < array.size</tt>), + * replaces from offset +start+ to the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(1) { |index| "new_#{index}" } # => ["a", "new_1", "new_2", "new_3"] + * + * If +start+ is too large(<tt>start >= array.size</tt>), does nothing: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(4) { |index| fail 'Cannot happen' } # => ["a", "b", "c", "d"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(4) { |index| fail 'Cannot happen' } # => ["a", "b", "c", "d"] + * + * If +start+ is negative, counts from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(-2) { |index| "new_#{index}" } # => ["a", "b", "new_2", "new_3"] + * + * If start is too small (<tt>start <= -array.size</tt>, replaces all elements: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(-6) { |index| "new_#{index}" } # => ["new_0", "new_1", "new_2", "new_3"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(-50) { |index| "new_#{index}" } # => ["new_0", "new_1", "new_2", "new_3"] + * + * With arguments +start+ and +length+, and a block given, + * calls the block for each index specified by start length; + * replaces the corresponding element with the block's return value. + * + * If +start+ is in range, replaces +length+ elements beginning at offset +start+: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(1, 1) { |index| "new_#{index}" } # => ["a", "new_1", "c", "d"] + * + * If start is negative, counts from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(-2, 1) { |index| "new_#{index}" } # => ["a", "b", "new_2", "d"] + * + * If +start+ is large (<tt>start >= array.size</tt>), extends +self+ with +nil+: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(5, 0) { |index| "new_#{index}" } # => ["a", "b", "c", "d", nil] + * a = ['a', 'b', 'c', 'd'] + * a.fill(5, 2) { |index| "new_#{index}" } # => ["a", "b", "c", "d", nil, "new_5", "new_6"] + * + * If +length+ is zero or less, replaces no elements: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(1, 0) { |index| "new_#{index}" } # => ["a", "b", "c", "d"] + * a.fill(1, -1) { |index| "new_#{index}" } # => ["a", "b", "c", "d"] * - * The first three forms set the selected elements of +self+ (which - * may be the entire array) to +obj+. + * With arguments +obj+ and +range+, and a block given, + * calls the block with each index in the given range; + * replaces the corresponding element with the block's return value. * - * A +start+ of +nil+ is equivalent to zero. + * If the range is positive and ascending (<tt>range 0 < range.begin <= range.end</tt>, + * replaces elements from <tt>range.begin</tt> to <tt>range.end</tt>: * - * A +length+ of +nil+ is equivalent to the length of the array. + * a = ['a', 'b', 'c', 'd'] + * a.fill(1..1) { |index| "new_#{index}" } # => ["a", "new_1", "c", "d"] * - * The last three forms fill the array with the value of the given block, - * which is passed the absolute index of each element to be filled. + * If +range.first+ is negative, does nothing: * - * Negative values of +start+ count from the end of the array, where +-1+ is - * the last element. + * a = ['a', 'b', 'c', 'd'] + * a.fill(-1..1) { |index| fail 'Cannot happen' } # => ["a", "b", "c", "d"] + * + * If <tt>range.last</tt> is negative, counts from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(0..-2) { |index| "new_#{index}" } # => ["new_0", "new_1", "new_2", "d"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(1..-2) { |index| "new_#{index}" } # => ["a", "new_1", "new_2", "d"] + * + * If <tt>range.first</tt> and <tt>range.last</tt> are both negative, + * both count from the end: + * + * a = ['a', 'b', 'c', 'd'] + * a.fill(-1..-1) { |index| "new_#{index}" } # => ["a", "b", "c", "new_3"] + * a = ['a', 'b', 'c', 'd'] + * a.fill(-2..-2) { |index| "new_#{index}" } # => ["a", "b", "new_2", "d"] * - * a = [ "a", "b", "c", "d" ] - * a.fill("x") #=> ["x", "x", "x", "x"] - * a.fill("z", 2, 2) #=> ["x", "x", "z", "z"] - * a.fill("y", 0..1) #=> ["y", "y", "z", "z"] - * a.fill { |i| i*i } #=> [0, 1, 4, 9] - * a.fill(-2) { |i| i*i*i } #=> [0, 1, 8, 27] */ static VALUE @@ -3563,84 +5011,74 @@ rb_ary_fill(int argc, VALUE *argv, VALUE ary) long beg = 0, end = 0, len = 0; if (rb_block_given_p()) { - rb_scan_args(argc, argv, "02", &arg1, &arg2); - argc += 1; /* hackish */ + rb_scan_args(argc, argv, "02", &arg1, &arg2); + argc += 1; /* hackish */ } else { - rb_scan_args(argc, argv, "12", &item, &arg1, &arg2); + rb_scan_args(argc, argv, "12", &item, &arg1, &arg2); } switch (argc) { case 1: - beg = 0; - len = RARRAY_LEN(ary); - break; + beg = 0; + len = RARRAY_LEN(ary); + break; case 2: - if (rb_range_beg_len(arg1, &beg, &len, RARRAY_LEN(ary), 1)) { - break; - } - /* fall through */ + if (rb_range_beg_len(arg1, &beg, &len, RARRAY_LEN(ary), 1)) { + break; + } + /* fall through */ case 3: - beg = NIL_P(arg1) ? 0 : NUM2LONG(arg1); - if (beg < 0) { - beg = RARRAY_LEN(ary) + beg; - if (beg < 0) beg = 0; - } - len = NIL_P(arg2) ? RARRAY_LEN(ary) - beg : NUM2LONG(arg2); - break; + beg = NIL_P(arg1) ? 0 : NUM2LONG(arg1); + if (beg < 0) { + beg = RARRAY_LEN(ary) + beg; + if (beg < 0) beg = 0; + } + len = NIL_P(arg2) ? RARRAY_LEN(ary) - beg : NUM2LONG(arg2); + break; } rb_ary_modify(ary); if (len < 0) { return ary; } if (beg >= ARY_MAX_SIZE || len > ARY_MAX_SIZE - beg) { - rb_raise(rb_eArgError, "argument too big"); + rb_raise(rb_eArgError, "argument too big"); } end = beg + len; if (RARRAY_LEN(ary) < end) { - if (end >= ARY_CAPA(ary)) { - ary_resize_capa(ary, end); - } - ary_mem_clear(ary, RARRAY_LEN(ary), end - RARRAY_LEN(ary)); - ARY_SET_LEN(ary, end); + if (end >= ARY_CAPA(ary)) { + ary_resize_capa(ary, end); + } + ary_mem_clear(ary, RARRAY_LEN(ary), end - RARRAY_LEN(ary)); + ARY_SET_LEN(ary, end); } - if (item == Qundef) { - VALUE v; - long i; + if (UNDEF_P(item)) { + VALUE v; + long i; - for (i=beg; i<end; i++) { - v = rb_yield(LONG2NUM(i)); - if (i>=RARRAY_LEN(ary)) break; - ARY_SET(ary, i, v); - } + for (i=beg; i<end; i++) { + v = rb_yield(LONG2NUM(i)); + if (i>=RARRAY_LEN(ary)) break; + ARY_SET(ary, i, v); + } } else { - ary_memfill(ary, beg, len, item); + ary_memfill(ary, beg, len, item); } return ary; } /* * call-seq: - * ary + other_ary -> new_ary - * - * Concatenation --- Returns a new array built by concatenating the - * two arrays together to produce a third array. + * array + other_array -> new_array * - * [ 1, 2, 3 ] + [ 4, 5 ] #=> [ 1, 2, 3, 4, 5 ] - * a = [ "a", "b", "c" ] - * c = a + [ "d", "e", "f" ] - * c #=> [ "a", "b", "c", "d", "e", "f" ] - * a #=> [ "a", "b", "c" ] + * Returns a new \Array containing all elements of +array+ + * followed by all elements of +other_array+: * - * Note that - * x += y - * is the same as - * x = x + y - * This means that it produces a new array. As a consequence, - * repeated use of <code>+=</code> on arrays can be quite inefficient. + * a = [0, 1] + [2, 3] + * a # => [0, 1, 2, 3] * - * See also Array#concat. + * Related: #concat. */ VALUE @@ -3655,8 +5093,8 @@ rb_ary_plus(VALUE x, VALUE y) len = xlen + ylen; z = rb_ary_new2(len); - ary_memcpy(z, 0, xlen, RARRAY_CONST_PTR(x)); - ary_memcpy(z, xlen, ylen, RARRAY_CONST_PTR(y)); + ary_memcpy(z, 0, xlen, RARRAY_CONST_PTR_TRANSIENT(x)); + ary_memcpy(z, xlen, ylen, RARRAY_CONST_PTR_TRANSIENT(y)); ARY_SET_LEN(z, len); return z; } @@ -3666,29 +5104,20 @@ ary_append(VALUE x, VALUE y) { long n = RARRAY_LEN(y); if (n > 0) { - rb_ary_splice(x, RARRAY_LEN(x), 0, RARRAY_CONST_PTR(y), n); + rb_ary_splice(x, RARRAY_LEN(x), 0, RARRAY_CONST_PTR_TRANSIENT(y), n); } + RB_GC_GUARD(y); return x; } /* * call-seq: - * ary.concat(other_ary1, other_ary2,...) -> ary - * - * Appends the elements of +other_ary+s to +self+. - * - * [ "a", "b" ].concat( ["c", "d"] ) #=> [ "a", "b", "c", "d" ] - * [ "a" ].concat( ["b"], ["c", "d"] ) #=> [ "a", "b", "c", "d" ] - * [ "a" ].concat #=> [ "a" ] - * - * a = [ 1, 2, 3 ] - * a.concat( [ 4, 5 ] ) - * a #=> [ 1, 2, 3, 4, 5 ] + * array.concat(*other_arrays) -> self * - * a = [ 1, 2 ] - * a.concat(a, a) #=> [1, 2, 1, 2, 1, 2] + * Adds to +array+ all elements from each \Array in +other_arrays+; returns +self+: * - * See also Array#+. + * a = [0, 1] + * a.concat([2, 3], [4, 5]) # => [0, 1, 2, 3, 4, 5] */ static VALUE @@ -3697,17 +5126,18 @@ rb_ary_concat_multi(int argc, VALUE *argv, VALUE ary) rb_ary_modify_check(ary); if (argc == 1) { - rb_ary_concat(ary, argv[0]); + rb_ary_concat(ary, argv[0]); } else if (argc > 1) { - int i; - VALUE args = rb_ary_tmp_new(argc); - for (i = 0; i < argc; i++) { - rb_ary_concat(args, argv[i]); - } - ary_append(ary, args); + int i; + VALUE args = rb_ary_hidden_new(argc); + for (i = 0; i < argc; i++) { + rb_ary_concat(args, argv[i]); + } + ary_append(ary, args); } + ary_verify(ary); return ary; } @@ -3719,18 +5149,19 @@ rb_ary_concat(VALUE x, VALUE y) /* * call-seq: - * ary * int -> new_ary - * ary * str -> new_string + * array * n -> new_array + * array * string_separator -> new_string * - * Repetition --- With a String argument, equivalent to - * <code>ary.join(str)</code>. + * When non-negative argument \Integer +n+ is given, + * returns a new \Array built by concatenating the +n+ copies of +self+: * - * Otherwise, returns a new array built by concatenating the +int+ copies of - * +self+. + * a = ['x', 'y'] + * a * 3 # => ["x", "y", "x", "y", "x", "y"] * + * When \String argument +string_separator+ is given, + * equivalent to <tt>array.join(string_separator)</tt>: * - * [ 1, 2, 3 ] * 3 #=> [ 1, 2, 3, 1, 2, 3, 1, 2, 3 ] - * [ 1, 2, 3 ] * "," #=> "1,2,3" + * [0, [0, 1], {foo: 0}] * ', ' # => "0, 0, 1, {:foo=>0}" * */ @@ -3743,61 +5174,54 @@ rb_ary_times(VALUE ary, VALUE times) tmp = rb_check_string_type(times); if (!NIL_P(tmp)) { - return rb_ary_join(ary, tmp); + return rb_ary_join(ary, tmp); } len = NUM2LONG(times); if (len == 0) { - ary2 = ary_new(rb_obj_class(ary), 0); - goto out; + ary2 = ary_new(rb_cArray, 0); + goto out; } if (len < 0) { - rb_raise(rb_eArgError, "negative argument"); + rb_raise(rb_eArgError, "negative argument"); } if (ARY_MAX_SIZE/len < RARRAY_LEN(ary)) { - rb_raise(rb_eArgError, "argument too big"); + rb_raise(rb_eArgError, "argument too big"); } len *= RARRAY_LEN(ary); - ary2 = ary_new(rb_obj_class(ary), len); + ary2 = ary_new(rb_cArray, len); ARY_SET_LEN(ary2, len); - ptr = RARRAY_CONST_PTR(ary); + ptr = RARRAY_CONST_PTR_TRANSIENT(ary); t = RARRAY_LEN(ary); if (0 < t) { - ary_memcpy(ary2, 0, t, ptr); - while (t <= len/2) { - ary_memcpy(ary2, t, t, RARRAY_CONST_PTR(ary2)); + ary_memcpy(ary2, 0, t, ptr); + while (t <= len/2) { + ary_memcpy(ary2, t, t, RARRAY_CONST_PTR_TRANSIENT(ary2)); t *= 2; } if (t < len) { - ary_memcpy(ary2, t, len-t, RARRAY_CONST_PTR(ary2)); + ary_memcpy(ary2, t, len-t, RARRAY_CONST_PTR_TRANSIENT(ary2)); } } out: - OBJ_INFECT(ary2, ary); - return ary2; } /* * call-seq: - * ary.assoc(obj) -> element_ary or nil + * array.assoc(obj) -> found_array or nil * - * Searches through an array whose elements are also arrays comparing +obj+ - * with the first element of each contained array using <code>obj.==</code>. + * Returns the first element in +self+ that is an \Array + * whose first element <tt>==</tt> +obj+: * - * Returns the first contained array that matches (that is, the first - * associated array), or +nil+ if no match is found. + * a = [{foo: 0}, [2, 4], [4, 5, 6], [4, 5]] + * a.assoc(4) # => [4, 5, 6] * - * See also Array#rassoc + * Returns +nil+ if no such element is found. * - * s1 = [ "colors", "red", "blue", "green" ] - * s2 = [ "letters", "a", "b", "c" ] - * s3 = "foo" - * a = [ s1, s2, s3 ] - * a.assoc("letters") #=> [ "letters", "a", "b", "c" ] - * a.assoc("foo") #=> nil + * Related: #rassoc. */ VALUE @@ -3807,30 +5231,27 @@ rb_ary_assoc(VALUE ary, VALUE key) VALUE v; for (i = 0; i < RARRAY_LEN(ary); ++i) { - v = rb_check_array_type(RARRAY_AREF(ary, i)); - if (!NIL_P(v) && RARRAY_LEN(v) > 0 && - rb_equal(RARRAY_AREF(v, 0), key)) - return v; + v = rb_check_array_type(RARRAY_AREF(ary, i)); + if (!NIL_P(v) && RARRAY_LEN(v) > 0 && + rb_equal(RARRAY_AREF(v, 0), key)) + return v; } return Qnil; } /* * call-seq: - * ary.rassoc(obj) -> element_ary or nil + * array.rassoc(obj) -> found_array or nil * - * Searches through the array whose elements are also arrays. + * Returns the first element in +self+ that is an \Array + * whose second element <tt>==</tt> +obj+: * - * Compares +obj+ with the second element of each contained array using - * <code>obj.==</code>. + * a = [{foo: 0}, [2, 4], [4, 5, 6], [4, 5]] + * a.rassoc(4) # => [2, 4] * - * Returns the first contained array that matches +obj+. + * Returns +nil+ if no such element is found. * - * See also Array#assoc. - * - * a = [ [ 1, "one"], [2, "two"], [3, "three"], ["ii", "two"] ] - * a.rassoc("two") #=> [2, "two"] - * a.rassoc("four") #=> nil + * Related: #assoc. */ VALUE @@ -3840,11 +5261,11 @@ rb_ary_rassoc(VALUE ary, VALUE value) VALUE v; for (i = 0; i < RARRAY_LEN(ary); ++i) { - v = RARRAY_AREF(ary, i); - if (RB_TYPE_P(v, T_ARRAY) && - RARRAY_LEN(v) > 1 && - rb_equal(RARRAY_AREF(v, 1), value)) - return v; + v = RARRAY_AREF(ary, i); + if (RB_TYPE_P(v, T_ARRAY) && + RARRAY_LEN(v) > 1 && + rb_equal(RARRAY_AREF(v, 1), value)) + return v; } return Qnil; } @@ -3857,43 +5278,48 @@ recursive_equal(VALUE ary1, VALUE ary2, int recur) if (recur) return Qtrue; /* Subtle! */ + /* rb_equal() can evacuate ptrs */ p1 = RARRAY_CONST_PTR(ary1); p2 = RARRAY_CONST_PTR(ary2); len1 = RARRAY_LEN(ary1); for (i = 0; i < len1; i++) { - if (*p1 != *p2) { - if (rb_equal(*p1, *p2)) { - len1 = RARRAY_LEN(ary1); - if (len1 != RARRAY_LEN(ary2)) - return Qfalse; - if (len1 < i) - return Qtrue; - p1 = RARRAY_CONST_PTR(ary1) + i; - p2 = RARRAY_CONST_PTR(ary2) + i; - } - else { - return Qfalse; - } - } - p1++; - p2++; + if (*p1 != *p2) { + if (rb_equal(*p1, *p2)) { + len1 = RARRAY_LEN(ary1); + if (len1 != RARRAY_LEN(ary2)) + return Qfalse; + if (len1 < i) + return Qtrue; + p1 = RARRAY_CONST_PTR(ary1) + i; + p2 = RARRAY_CONST_PTR(ary2) + i; + } + else { + return Qfalse; + } + } + p1++; + p2++; } return Qtrue; } /* * call-seq: - * ary == other_ary -> bool + * array == other_array -> true or false * - * Equality --- Two arrays are equal if they contain the same number of - * elements and if each element is equal to (according to Object#==) the - * corresponding element in +other_ary+. + * Returns +true+ if both <tt>array.size == other_array.size</tt> + * and for each index +i+ in +array+, <tt>array[i] == other_array[i]</tt>: * - * [ "a", "c" ] == [ "a", "c", 7 ] #=> false - * [ "a", "c", 7 ] == [ "a", "c", 7 ] #=> true - * [ "a", "c", 7 ] == [ "a", "d", "f" ] #=> false + * a0 = [:foo, 'bar', 2] + * a1 = [:foo, 'bar', 2.0] + * a1 == a0 # => true + * [] == [] # => true * + * Otherwise, returns +false+. + * + * This method is different from method Array#eql?, + * which compares elements using <tt>Object#eql?</tt>. */ static VALUE @@ -3901,13 +5327,13 @@ rb_ary_equal(VALUE ary1, VALUE ary2) { if (ary1 == ary2) return Qtrue; if (!RB_TYPE_P(ary2, T_ARRAY)) { - if (!rb_respond_to(ary2, idTo_ary)) { - return Qfalse; - } - return rb_equal(ary2, ary1); + if (!rb_respond_to(ary2, idTo_ary)) { + return Qfalse; + } + return rb_equal(ary2, ary1); } if (RARRAY_LEN(ary1) != RARRAY_LEN(ary2)) return Qfalse; - if (RARRAY_CONST_PTR(ary1) == RARRAY_CONST_PTR(ary2)) return Qtrue; + if (RARRAY_CONST_PTR_TRANSIENT(ary1) == RARRAY_CONST_PTR_TRANSIENT(ary2)) return Qtrue; return rb_exec_recursive_paired(recursive_equal, ary1, ary2, ary2); } @@ -3918,18 +5344,27 @@ recursive_eql(VALUE ary1, VALUE ary2, int recur) if (recur) return Qtrue; /* Subtle! */ for (i=0; i<RARRAY_LEN(ary1); i++) { - if (!rb_eql(rb_ary_elt(ary1, i), rb_ary_elt(ary2, i))) - return Qfalse; + if (!rb_eql(rb_ary_elt(ary1, i), rb_ary_elt(ary2, i))) + return Qfalse; } return Qtrue; } /* * call-seq: - * ary.eql?(other) -> true or false + * array.eql? other_array -> true or false + * + * Returns +true+ if +self+ and +other_array+ are the same size, + * and if, for each index +i+ in +self+, <tt>self[i].eql? other_array[i]</tt>: * - * Returns +true+ if +self+ and +other+ are the same object, - * or are both arrays with the same content (according to Object#eql?). + * a0 = [:foo, 'bar', 2] + * a1 = [:foo, 'bar', 2] + * a1.eql?(a0) # => true + * + * Otherwise, returns +false+. + * + * This method is different from method Array#==, + * which compares using method <tt>Object#==</tt>. */ static VALUE @@ -3938,20 +5373,21 @@ rb_ary_eql(VALUE ary1, VALUE ary2) if (ary1 == ary2) return Qtrue; if (!RB_TYPE_P(ary2, T_ARRAY)) return Qfalse; if (RARRAY_LEN(ary1) != RARRAY_LEN(ary2)) return Qfalse; - if (RARRAY_CONST_PTR(ary1) == RARRAY_CONST_PTR(ary2)) return Qtrue; + if (RARRAY_CONST_PTR_TRANSIENT(ary1) == RARRAY_CONST_PTR_TRANSIENT(ary2)) return Qtrue; return rb_exec_recursive_paired(recursive_eql, ary1, ary2, ary2); } /* * call-seq: - * ary.hash -> integer + * array.hash -> integer + * + * Returns the integer hash value for +self+. * - * Compute a hash-code for this array. + * Two arrays with the same content will have the same hash code (and will compare using eql?): * - * Two arrays with the same content will have the same hash code (and will - * compare using #eql?). + * [0, 1, 2].hash == [0, 1, 2].hash # => true + * [0, 1, 2].hash == [0, 1, 3].hash # => false * - * See also Object#hash. */ static VALUE @@ -3964,8 +5400,8 @@ rb_ary_hash(VALUE ary) h = rb_hash_start(RARRAY_LEN(ary)); h = rb_hash_uint(h, (st_index_t)rb_ary_hash); for (i=0; i<RARRAY_LEN(ary); i++) { - n = rb_hash(RARRAY_AREF(ary, i)); - h = rb_hash_uint(h, NUM2LONG(n)); + n = rb_hash(RARRAY_AREF(ary, i)); + h = rb_hash_uint(h, NUM2LONG(n)); } h = rb_hash_end(h); return ST2FIX(h); @@ -3973,14 +5409,13 @@ rb_ary_hash(VALUE ary) /* * call-seq: - * ary.include?(object) -> true or false + * array.include?(obj) -> true or false * - * Returns +true+ if the given +object+ is present in +self+ (that is, if any - * element <code>==</code> +object+), otherwise returns +false+. + * Returns +true+ if for some index +i+ in +self+, <tt>obj == self[i]</tt>; + * otherwise +false+: * - * a = [ "a", "b", "c" ] - * a.include?("b") #=> true - * a.include?("z") #=> false + * [0, 1, 2].include?(2) # => true + * [0, 1, 2].include?(3) # => false */ VALUE @@ -3990,10 +5425,10 @@ rb_ary_includes(VALUE ary, VALUE item) VALUE e; for (i=0; i<RARRAY_LEN(ary); i++) { - e = RARRAY_AREF(ary, i); - if (rb_equal(e, item)) { - return Qtrue; - } + e = RARRAY_AREF(ary, i); + if (rb_equal(e, item)) { + return Qtrue; + } } return Qfalse; } @@ -4005,10 +5440,10 @@ rb_ary_includes_by_eql(VALUE ary, VALUE item) VALUE e; for (i=0; i<RARRAY_LEN(ary); i++) { - e = RARRAY_AREF(ary, i); - if (rb_eql(item, e)) { - return Qtrue; - } + e = RARRAY_AREF(ary, i); + if (rb_eql(item, e)) { + return Qtrue; + } } return Qfalse; } @@ -4021,45 +5456,46 @@ recursive_cmp(VALUE ary1, VALUE ary2, int recur) if (recur) return Qundef; /* Subtle! */ len = RARRAY_LEN(ary1); if (len > RARRAY_LEN(ary2)) { - len = RARRAY_LEN(ary2); + len = RARRAY_LEN(ary2); } for (i=0; i<len; i++) { - VALUE e1 = rb_ary_elt(ary1, i), e2 = rb_ary_elt(ary2, i); - VALUE v = rb_funcallv(e1, id_cmp, 1, &e2); - if (v != INT2FIX(0)) { - return v; - } + VALUE e1 = rb_ary_elt(ary1, i), e2 = rb_ary_elt(ary2, i); + VALUE v = rb_funcallv(e1, id_cmp, 1, &e2); + if (v != INT2FIX(0)) { + return v; + } } return Qundef; } /* * call-seq: - * ary <=> other_ary -> -1, 0, +1 or nil + * array <=> other_array -> -1, 0, or 1 + * + * Returns -1, 0, or 1 as +self+ is less than, equal to, or greater than +other_array+. + * For each index +i+ in +self+, evaluates <tt>result = self[i] <=> other_array[i]</tt>. + * + * Returns -1 if any result is -1: * - * Comparison --- Returns an integer (+-1+, +0+, or <code>+1</code>) if this - * array is less than, equal to, or greater than +other_ary+. + * [0, 1, 2] <=> [0, 1, 3] # => -1 * - * Each object in each array is compared (using the <=> operator). + * Returns 1 if any result is 1: * - * Arrays are compared in an "element-wise" manner; the first element of +ary+ - * is compared with the first one of +other_ary+ using the <=> operator, then - * each of the second elements, etc... - * As soon as the result of any such comparison is non zero (i.e. the two - * corresponding elements are not equal), that result is returned for the - * whole array comparison. + * [0, 1, 2] <=> [0, 1, 1] # => 1 * - * If all the elements are equal, then the result is based on a comparison of - * the array lengths. Thus, two arrays are "equal" according to Array#<=> if, - * and only if, they have the same length and the value of each element is - * equal to the value of the corresponding element in the other array. + * When all results are zero: * - * +nil+ is returned if the +other_ary+ is not an array or if the comparison - * of two elements returned +nil+. + * - Returns -1 if +array+ is smaller than +other_array+: * - * [ "a", "a", "c" ] <=> [ "a", "b", "c" ] #=> -1 - * [ 1, 2, 3, 4, 5, 6 ] <=> [ 1, 2 ] #=> +1 - * [ 1, 2 ] <=> [ 1, :two ] #=> nil + * [0, 1, 2] <=> [0, 1, 2, 3] # => -1 + * + * - Returns 1 if +array+ is larger than +other_array+: + * + * [0, 1, 2] <=> [0, 1] # => 1 + * + * - Returns 0 if +array+ and +other_array+ are the same size: + * + * [0, 1, 2] <=> [0, 1, 2] # => 0 * */ @@ -4073,7 +5509,7 @@ rb_ary_cmp(VALUE ary1, VALUE ary2) if (NIL_P(ary2)) return Qnil; if (ary1 == ary2) return INT2FIX(0); v = rb_exec_recursive_paired(recursive_cmp, ary1, ary2, ary2); - if (v != Qundef) return v; + if (!UNDEF_P(v)) return v; len = RARRAY_LEN(ary1) - RARRAY_LEN(ary2); if (len == 0) return INT2FIX(0); if (len > 0) return INT2FIX(1); @@ -4086,8 +5522,8 @@ ary_add_hash(VALUE hash, VALUE ary) long i; for (i=0; i<RARRAY_LEN(ary); i++) { - VALUE elt = RARRAY_AREF(ary, i); - rb_hash_add_new_element(hash, elt, elt); + VALUE elt = RARRAY_AREF(ary, i); + rb_hash_add_new_element(hash, elt, elt); } return hash; } @@ -4115,8 +5551,8 @@ ary_add_hash_by(VALUE hash, VALUE ary) long i; for (i = 0; i < RARRAY_LEN(ary); ++i) { - VALUE v = rb_ary_elt(ary, i), k = rb_yield(v); - rb_hash_add_new_element(hash, k, v); + VALUE v = rb_ary_elt(ary, i), k = rb_yield(v); + rb_hash_add_new_element(hash, k, v); } return hash; } @@ -4132,31 +5568,30 @@ static inline void ary_recycle_hash(VALUE hash) { assert(RBASIC_CLASS(hash) == 0); - if (RHASH(hash)->ntbl) { - st_table *tbl = RHASH(hash)->ntbl; - st_free_table(tbl); + if (RHASH_ST_TABLE_P(hash)) { + st_table *tbl = RHASH_ST_TABLE(hash); + st_free_table(tbl); + RHASH_ST_CLEAR(hash); } - rb_gc_force_recycle(hash); } /* * call-seq: - * ary - other_ary -> new_ary - * - * Array Difference - * - * Returns a new array that is a copy of the original array, removing any - * items that also appear in +other_ary+. The order is preserved from the - * original array. + * array - other_array -> new_array * - * It compares elements using their #hash and #eql? methods for efficiency. + * Returns a new \Array containing only those elements from +array+ + * that are not found in \Array +other_array+; + * items are compared using <tt>eql?</tt>; + * the order from +array+ is preserved: * - * [ 1, 1, 2, 2, 3, 3, 4, 5 ] - [ 1, 2, 4 ] #=> [ 3, 3, 5 ] + * [0, 1, 1, 2, 1, 1, 3, 1, 1] - [1] # => [0, 2, 3] + * [0, 1, 2, 3] - [3, 0] # => [1, 2] + * [0, 1, 2] - [4] # => [0, 1, 2] * - * If you need set-like behavior, see the library class Set. + * Related: Array#difference. */ -static VALUE +VALUE rb_ary_diff(VALUE ary1, VALUE ary2) { VALUE ary3; @@ -4164,21 +5599,22 @@ rb_ary_diff(VALUE ary1, VALUE ary2) long i; ary2 = to_ary(ary2); + if (RARRAY_LEN(ary2) == 0) { return ary_make_shared_copy(ary1); } ary3 = rb_ary_new(); if (RARRAY_LEN(ary1) <= SMALL_ARRAY_LEN || RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) { - for (i=0; i<RARRAY_LEN(ary1); i++) { - VALUE elt = rb_ary_elt(ary1, i); - if (rb_ary_includes_by_eql(ary2, elt)) continue; - rb_ary_push(ary3, elt); - } - return ary3; + for (i=0; i<RARRAY_LEN(ary1); i++) { + VALUE elt = rb_ary_elt(ary1, i); + if (rb_ary_includes_by_eql(ary2, elt)) continue; + rb_ary_push(ary3, elt); + } + return ary3; } hash = ary_make_hash(ary2); for (i=0; i<RARRAY_LEN(ary1); i++) { - if (st_lookup(rb_hash_tbl_raw(hash), RARRAY_AREF(ary1, i), 0)) continue; - rb_ary_push(ary3, rb_ary_elt(ary1, i)); + if (rb_hash_stlike_lookup(hash, RARRAY_AREF(ary1, i), NULL)) continue; + rb_ary_push(ary3, rb_ary_elt(ary1, i)); } ary_recycle_hash(hash); return ary3; @@ -4186,17 +5622,73 @@ rb_ary_diff(VALUE ary1, VALUE ary2) /* * call-seq: - * ary & other_ary -> new_ary + * array.difference(*other_arrays) -> new_array * - * Set Intersection --- Returns a new array containing unique elements common to the - * two arrays. The order is preserved from the original array. + * Returns a new \Array containing only those elements from +self+ + * that are not found in any of the Arrays +other_arrays+; + * items are compared using <tt>eql?</tt>; order from +self+ is preserved: * - * It compares elements using their #hash and #eql? methods for efficiency. + * [0, 1, 1, 2, 1, 1, 3, 1, 1].difference([1]) # => [0, 2, 3] + * [0, 1, 2, 3].difference([3, 0], [1, 3]) # => [2] + * [0, 1, 2].difference([4]) # => [0, 1, 2] * - * [ 1, 1, 3, 5 ] & [ 3, 2, 1 ] #=> [ 1, 3 ] - * [ 'a', 'b', 'b', 'z' ] & [ 'a', 'b', 'c' ] #=> [ 'a', 'b' ] + * Returns a copy of +self+ if no arguments given. * - * See also Array#uniq. + * Related: Array#-. + */ + +static VALUE +rb_ary_difference_multi(int argc, VALUE *argv, VALUE ary) +{ + VALUE ary_diff; + long i, length; + volatile VALUE t0; + bool *is_hash = ALLOCV_N(bool, t0, argc); + ary_diff = rb_ary_new(); + length = RARRAY_LEN(ary); + + for (i = 0; i < argc; i++) { + argv[i] = to_ary(argv[i]); + is_hash[i] = (length > SMALL_ARRAY_LEN && RARRAY_LEN(argv[i]) > SMALL_ARRAY_LEN); + if (is_hash[i]) argv[i] = ary_make_hash(argv[i]); + } + + for (i = 0; i < RARRAY_LEN(ary); i++) { + int j; + VALUE elt = rb_ary_elt(ary, i); + for (j = 0; j < argc; j++) { + if (is_hash[j]) { + if (rb_hash_stlike_lookup(argv[j], RARRAY_AREF(ary, i), NULL)) + break; + } + else { + if (rb_ary_includes_by_eql(argv[j], elt)) break; + } + } + if (j == argc) rb_ary_push(ary_diff, elt); + } + + ALLOCV_END(t0); + + return ary_diff; +} + + +/* + * call-seq: + * array & other_array -> new_array + * + * Returns a new \Array containing each element found in both +array+ and \Array +other_array+; + * duplicates are omitted; items are compared using <tt>eql?</tt>: + * + * [0, 1, 2, 3] & [1, 2] # => [1, 2] + * [0, 1, 0, 1] & [0, 1] # => [0, 1] + * + * Preserves order from +array+: + * + * [0, 1, 2] & [3, 2, 1, 0] # => [0, 1, 2] + * + * Related: Array#intersection. */ @@ -4204,39 +5696,70 @@ static VALUE rb_ary_and(VALUE ary1, VALUE ary2) { VALUE hash, ary3, v; - st_table *table; st_data_t vv; long i; ary2 = to_ary(ary2); ary3 = rb_ary_new(); - if (RARRAY_LEN(ary2) == 0) return ary3; + if (RARRAY_LEN(ary1) == 0 || RARRAY_LEN(ary2) == 0) return ary3; if (RARRAY_LEN(ary1) <= SMALL_ARRAY_LEN && RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) { - for (i=0; i<RARRAY_LEN(ary1); i++) { - v = RARRAY_AREF(ary1, i); - if (!rb_ary_includes_by_eql(ary2, v)) continue; - if (rb_ary_includes_by_eql(ary3, v)) continue; - rb_ary_push(ary3, v); - } - return ary3; + for (i=0; i<RARRAY_LEN(ary1); i++) { + v = RARRAY_AREF(ary1, i); + if (!rb_ary_includes_by_eql(ary2, v)) continue; + if (rb_ary_includes_by_eql(ary3, v)) continue; + rb_ary_push(ary3, v); + } + return ary3; } hash = ary_make_hash(ary2); - table = rb_hash_tbl_raw(hash); for (i=0; i<RARRAY_LEN(ary1); i++) { - v = RARRAY_AREF(ary1, i); - vv = (st_data_t)v; - if (st_delete(table, &vv, 0)) { - rb_ary_push(ary3, v); - } + v = RARRAY_AREF(ary1, i); + vv = (st_data_t)v; + if (rb_hash_stlike_delete(hash, &vv, 0)) { + rb_ary_push(ary3, v); + } } ary_recycle_hash(hash); return ary3; } +/* + * call-seq: + * array.intersection(*other_arrays) -> new_array + * + * Returns a new \Array containing each element found both in +self+ + * and in all of the given Arrays +other_arrays+; + * duplicates are omitted; items are compared using <tt>eql?</tt>: + * + * [0, 1, 2, 3].intersection([0, 1, 2], [0, 1, 3]) # => [0, 1] + * [0, 0, 1, 1, 2, 3].intersection([0, 1, 2], [0, 1, 3]) # => [0, 1] + * + * Preserves order from +self+: + * + * [0, 1, 2].intersection([2, 1, 0]) # => [0, 1, 2] + * + * Returns a copy of +self+ if no arguments given. + * + * Related: Array#&. + */ + +static VALUE +rb_ary_intersection_multi(int argc, VALUE *argv, VALUE ary) +{ + VALUE result = rb_ary_dup(ary); + int i; + + for (i = 0; i < argc; i++) { + result = rb_ary_and(result, argv[i]); + } + + return result; +} + static int ary_hash_orset(st_data_t *key, st_data_t *value, st_data_t arg, int existing) { @@ -4245,50 +5768,60 @@ ary_hash_orset(st_data_t *key, st_data_t *value, st_data_t arg, int existing) return ST_CONTINUE; } +static void +rb_ary_union(VALUE ary_union, VALUE ary) +{ + long i; + for (i = 0; i < RARRAY_LEN(ary); i++) { + VALUE elt = rb_ary_elt(ary, i); + if (rb_ary_includes_by_eql(ary_union, elt)) continue; + rb_ary_push(ary_union, elt); + } +} + +static void +rb_ary_union_hash(VALUE hash, VALUE ary2) +{ + long i; + for (i = 0; i < RARRAY_LEN(ary2); i++) { + VALUE elt = RARRAY_AREF(ary2, i); + if (!rb_hash_stlike_update(hash, (st_data_t)elt, ary_hash_orset, (st_data_t)elt)) { + RB_OBJ_WRITTEN(hash, Qundef, elt); + } + } +} + /* * call-seq: - * ary | other_ary -> new_ary - * - * Set Union --- Returns a new array by joining +ary+ with +other_ary+, - * excluding any duplicates and preserving the order from the given arrays. + * array | other_array -> new_array * - * It compares elements using their #hash and #eql? methods for efficiency. + * Returns the union of +array+ and \Array +other_array+; + * duplicates are removed; order is preserved; + * items are compared using <tt>eql?</tt>: * - * [ "a", "b", "c" ] | [ "c", "d", "a" ] #=> [ "a", "b", "c", "d" ] - * [ "c", "d", "a" ] | [ "a", "b", "c" ] #=> [ "c", "d", "a", "b" ] + * [0, 1] | [2, 3] # => [0, 1, 2, 3] + * [0, 1, 1] | [2, 2, 3] # => [0, 1, 2, 3] + * [0, 1, 2] | [3, 2, 1, 0] # => [0, 1, 2, 3] * - * See also Array#uniq. + * Related: Array#union. */ static VALUE rb_ary_or(VALUE ary1, VALUE ary2) { VALUE hash, ary3; - long i; ary2 = to_ary(ary2); if (RARRAY_LEN(ary1) + RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) { - ary3 = rb_ary_new(); - for (i=0; i<RARRAY_LEN(ary1); i++) { - VALUE elt = rb_ary_elt(ary1, i); - if (rb_ary_includes_by_eql(ary3, elt)) continue; - rb_ary_push(ary3, elt); - } - for (i=0; i<RARRAY_LEN(ary2); i++) { - VALUE elt = rb_ary_elt(ary2, i); - if (rb_ary_includes_by_eql(ary3, elt)) continue; - rb_ary_push(ary3, elt); - } - return ary3; + ary3 = rb_ary_new(); + rb_ary_union(ary3, ary1); + rb_ary_union(ary3, ary2); + return ary3; } hash = ary_make_hash(ary1); - for (i=0; i<RARRAY_LEN(ary2); i++) { - VALUE elt = RARRAY_AREF(ary2, i); - if (!st_update(RHASH_TBL_RAW(hash), (st_data_t)elt, ary_hash_orset, (st_data_t)elt)) { - RB_OBJ_WRITTEN(hash, Qundef, elt); - } - } + rb_ary_union_hash(hash, ary2); + ary3 = rb_hash_values(hash); ary_recycle_hash(hash); return ary3; @@ -4296,114 +5829,474 @@ rb_ary_or(VALUE ary1, VALUE ary2) /* * call-seq: - * ary.max -> obj - * ary.max { |a, b| block } -> obj - * ary.max(n) -> array - * ary.max(n) { |a, b| block } -> array + * array.union(*other_arrays) -> new_array + * + * Returns a new \Array that is the union of +self+ and all given Arrays +other_arrays+; + * duplicates are removed; order is preserved; items are compared using <tt>eql?</tt>: + * + * [0, 1, 2, 3].union([4, 5], [6, 7]) # => [0, 1, 2, 3, 4, 5, 6, 7] + * [0, 1, 1].union([2, 1], [3, 1]) # => [0, 1, 2, 3] + * [0, 1, 2, 3].union([3, 2], [1, 0]) # => [0, 1, 2, 3] * - * Returns the object in _ary_ with the maximum value. The - * first form assumes all objects implement <code>Comparable</code>; - * the second uses the block to return <em>a <=> b</em>. + * Returns a copy of +self+ if no arguments given. + * + * Related: Array#|. + */ + +static VALUE +rb_ary_union_multi(int argc, VALUE *argv, VALUE ary) +{ + int i; + long sum; + VALUE hash, ary_union; + + sum = RARRAY_LEN(ary); + for (i = 0; i < argc; i++) { + argv[i] = to_ary(argv[i]); + sum += RARRAY_LEN(argv[i]); + } + + if (sum <= SMALL_ARRAY_LEN) { + ary_union = rb_ary_new(); + + rb_ary_union(ary_union, ary); + for (i = 0; i < argc; i++) rb_ary_union(ary_union, argv[i]); + + return ary_union; + } + + hash = ary_make_hash(ary); + for (i = 0; i < argc; i++) rb_ary_union_hash(hash, argv[i]); + + ary_union = rb_hash_values(hash); + ary_recycle_hash(hash); + return ary_union; +} + +/* + * call-seq: + * ary.intersect?(other_ary) -> true or false * - * ary = %w(albatross dog horse) - * ary.max #=> "horse" - * ary.max { |a, b| a.length <=> b.length } #=> "albatross" + * Returns +true+ if the array and +other_ary+ have at least one element in + * common, otherwise returns +false+: * - * If the +n+ argument is given, maximum +n+ elements are returned - * as an array. + * a = [ 1, 2, 3 ] + * b = [ 3, 4, 5 ] + * c = [ 5, 6, 7 ] + * a.intersect?(b) #=> true + * a.intersect?(c) #=> false + * + */ + +static VALUE +rb_ary_intersect_p(VALUE ary1, VALUE ary2) +{ + VALUE hash, v, result, shorter, longer; + st_data_t vv; + long i; + + ary2 = to_ary(ary2); + if (RARRAY_LEN(ary1) == 0 || RARRAY_LEN(ary2) == 0) return Qfalse; + + if (RARRAY_LEN(ary1) <= SMALL_ARRAY_LEN && RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) { + for (i=0; i<RARRAY_LEN(ary1); i++) { + v = RARRAY_AREF(ary1, i); + if (rb_ary_includes_by_eql(ary2, v)) return Qtrue; + } + return Qfalse; + } + + shorter = ary1; + longer = ary2; + if (RARRAY_LEN(ary1) > RARRAY_LEN(ary2)) { + longer = ary1; + shorter = ary2; + } + + hash = ary_make_hash(shorter); + result = Qfalse; + + for (i=0; i<RARRAY_LEN(longer); i++) { + v = RARRAY_AREF(longer, i); + vv = (st_data_t)v; + if (rb_hash_stlike_lookup(hash, vv, 0)) { + result = Qtrue; + break; + } + } + ary_recycle_hash(hash); + + return result; +} + +static VALUE +ary_max_generic(VALUE ary, long i, VALUE vmax) +{ + RUBY_ASSERT(i > 0 && i < RARRAY_LEN(ary)); + + VALUE v; + for (; i < RARRAY_LEN(ary); ++i) { + v = RARRAY_AREF(ary, i); + + if (rb_cmpint(rb_funcallv(vmax, id_cmp, 1, &v), vmax, v) < 0) { + vmax = v; + } + } + + return vmax; +} + +static VALUE +ary_max_opt_fixnum(VALUE ary, long i, VALUE vmax) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(FIXNUM_P(vmax)); + + VALUE v; + for (; i < n; ++i) { + v = RARRAY_AREF(ary, i); + + if (FIXNUM_P(v)) { + if ((long)vmax < (long)v) { + vmax = v; + } + } + else { + return ary_max_generic(ary, i, vmax); + } + } + + return vmax; +} + +static VALUE +ary_max_opt_float(VALUE ary, long i, VALUE vmax) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(RB_FLOAT_TYPE_P(vmax)); + + VALUE v; + for (; i < n; ++i) { + v = RARRAY_AREF(ary, i); + + if (RB_FLOAT_TYPE_P(v)) { + if (rb_float_cmp(vmax, v) < 0) { + vmax = v; + } + } + else { + return ary_max_generic(ary, i, vmax); + } + } + + return vmax; +} + +static VALUE +ary_max_opt_string(VALUE ary, long i, VALUE vmax) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(STRING_P(vmax)); + + VALUE v; + for (; i < n; ++i) { + v = RARRAY_AREF(ary, i); + + if (STRING_P(v)) { + if (rb_str_cmp(vmax, v) < 0) { + vmax = v; + } + } + else { + return ary_max_generic(ary, i, vmax); + } + } + + return vmax; +} + +/* + * call-seq: + * array.max -> element + * array.max {|a, b| ... } -> element + * array.max(n) -> new_array + * array.max(n) {|a, b| ... } -> new_array + * + * Returns one of the following: + * + * - The maximum-valued element from +self+. + * - A new \Array of maximum-valued elements selected from +self+. + * + * When no block is given, each element in +self+ must respond to method <tt><=></tt> + * with an \Integer. + * + * With no argument and no block, returns the element in +self+ + * having the maximum value per method <tt><=></tt>: + * + * [0, 1, 2].max # => 2 + * + * With an argument \Integer +n+ and no block, returns a new \Array with at most +n+ elements, + * in descending order per method <tt><=></tt>: + * + * [0, 1, 2, 3].max(3) # => [3, 2, 1] + * [0, 1, 2, 3].max(6) # => [3, 2, 1, 0] + * + * When a block is given, the block must return an \Integer. + * + * With a block and no argument, calls the block <tt>self.size-1</tt> times to compare elements; + * returns the element having the maximum value per the block: + * + * ['0', '00', '000'].max {|a, b| a.size <=> b.size } # => "000" + * + * With an argument +n+ and a block, returns a new \Array with at most +n+ elements, + * in descending order per the block: + * + * ['0', '00', '000'].max(2) {|a, b| a.size <=> b.size } # => ["000", "00"] * - * ary = %w[albatross dog horse] - * ary.max(2) #=> ["horse", "dog"] - * ary.max(2) {|a, b| a.length <=> b.length } #=> ["albatross", "horse"] */ static VALUE rb_ary_max(int argc, VALUE *argv, VALUE ary) { - struct cmp_opt_data cmp_opt = { 0, 0 }; VALUE result = Qundef, v; VALUE num; long i; - rb_scan_args(argc, argv, "01", &num); - - if (!NIL_P(num)) + if (rb_check_arity(argc, 0, 1) && !NIL_P(num = argv[0])) return rb_nmin_run(ary, num, 0, 1, 1); + const long n = RARRAY_LEN(ary); if (rb_block_given_p()) { - for (i = 0; i < RARRAY_LEN(ary); i++) { - v = RARRAY_AREF(ary, i); - if (result == Qundef || rb_cmpint(rb_yield_values(2, v, result), v, result) > 0) { - result = v; - } - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + v = RARRAY_AREF(ary, i); + if (UNDEF_P(result) || rb_cmpint(rb_yield_values(2, v, result), v, result) > 0) { + result = v; + } + } } - else { - for (i = 0; i < RARRAY_LEN(ary); i++) { - v = RARRAY_AREF(ary, i); - if (result == Qundef || OPTIMIZED_CMP(v, result, cmp_opt) > 0) { - result = v; - } - } - } - if (result == Qundef) return Qnil; + else if (n > 0) { + result = RARRAY_AREF(ary, 0); + if (n > 1) { + if (FIXNUM_P(result) && CMP_OPTIMIZABLE(INTEGER)) { + return ary_max_opt_fixnum(ary, 1, result); + } + else if (STRING_P(result) && CMP_OPTIMIZABLE(STRING)) { + return ary_max_opt_string(ary, 1, result); + } + else if (RB_FLOAT_TYPE_P(result) && CMP_OPTIMIZABLE(FLOAT)) { + return ary_max_opt_float(ary, 1, result); + } + else { + return ary_max_generic(ary, 1, result); + } + } + } + if (UNDEF_P(result)) return Qnil; return result; } +static VALUE +ary_min_generic(VALUE ary, long i, VALUE vmin) +{ + RUBY_ASSERT(i > 0 && i < RARRAY_LEN(ary)); + + VALUE v; + for (; i < RARRAY_LEN(ary); ++i) { + v = RARRAY_AREF(ary, i); + + if (rb_cmpint(rb_funcallv(vmin, id_cmp, 1, &v), vmin, v) > 0) { + vmin = v; + } + } + + return vmin; +} + +static VALUE +ary_min_opt_fixnum(VALUE ary, long i, VALUE vmin) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(FIXNUM_P(vmin)); + + VALUE a; + for (; i < n; ++i) { + a = RARRAY_AREF(ary, i); + + if (FIXNUM_P(a)) { + if ((long)vmin > (long)a) { + vmin = a; + } + } + else { + return ary_min_generic(ary, i, vmin); + } + } + + return vmin; +} + +static VALUE +ary_min_opt_float(VALUE ary, long i, VALUE vmin) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(RB_FLOAT_TYPE_P(vmin)); + + VALUE a; + for (; i < n; ++i) { + a = RARRAY_AREF(ary, i); + + if (RB_FLOAT_TYPE_P(a)) { + if (rb_float_cmp(vmin, a) > 0) { + vmin = a; + } + } + else { + return ary_min_generic(ary, i, vmin); + } + } + + return vmin; +} + +static VALUE +ary_min_opt_string(VALUE ary, long i, VALUE vmin) +{ + const long n = RARRAY_LEN(ary); + RUBY_ASSERT(i > 0 && i < n); + RUBY_ASSERT(STRING_P(vmin)); + + VALUE a; + for (; i < n; ++i) { + a = RARRAY_AREF(ary, i); + + if (STRING_P(a)) { + if (rb_str_cmp(vmin, a) > 0) { + vmin = a; + } + } + else { + return ary_min_generic(ary, i, vmin); + } + } + + return vmin; +} + /* * call-seq: - * ary.min -> obj - * ary.min {| a,b | block } -> obj - * ary.min(n) -> array - * ary.min(n) {| a,b | block } -> array + * array.min -> element + * array.min { |a, b| ... } -> element + * array.min(n) -> new_array + * array.min(n) { |a, b| ... } -> new_array * - * Returns the object in _ary_ with the minimum value. The - * first form assumes all objects implement <code>Comparable</code>; - * the second uses the block to return <em>a <=> b</em>. + * Returns one of the following: * - * ary = %w(albatross dog horse) - * ary.min #=> "albatross" - * ary.min { |a, b| a.length <=> b.length } #=> "dog" + * - The minimum-valued element from +self+. + * - A new \Array of minimum-valued elements selected from +self+. * - * If the +n+ argument is given, minimum +n+ elements are returned - * as an array. + * When no block is given, each element in +self+ must respond to method <tt><=></tt> + * with an \Integer. + * + * With no argument and no block, returns the element in +self+ + * having the minimum value per method <tt><=></tt>: + * + * [0, 1, 2].min # => 0 + * + * With \Integer argument +n+ and no block, returns a new \Array with at most +n+ elements, + * in ascending order per method <tt><=></tt>: + * + * [0, 1, 2, 3].min(3) # => [0, 1, 2] + * [0, 1, 2, 3].min(6) # => [0, 1, 2, 3] + * + * When a block is given, the block must return an Integer. + * + * With a block and no argument, calls the block <tt>self.size-1</tt> times to compare elements; + * returns the element having the minimum value per the block: + * + * ['0', '00', '000'].min { |a, b| a.size <=> b.size } # => "0" + * + * With an argument +n+ and a block, returns a new \Array with at most +n+ elements, + * in ascending order per the block: + * + * ['0', '00', '000'].min(2) {|a, b| a.size <=> b.size } # => ["0", "00"] * - * ary = %w[albatross dog horse] - * ary.min(2) #=> ["albatross", "dog"] - * ary.min(2) {|a, b| a.length <=> b.length } #=> ["dog", "horse"] */ static VALUE rb_ary_min(int argc, VALUE *argv, VALUE ary) { - struct cmp_opt_data cmp_opt = { 0, 0 }; VALUE result = Qundef, v; VALUE num; long i; - rb_scan_args(argc, argv, "01", &num); - - if (!NIL_P(num)) + if (rb_check_arity(argc, 0, 1) && !NIL_P(num = argv[0])) return rb_nmin_run(ary, num, 0, 0, 1); + const long n = RARRAY_LEN(ary); if (rb_block_given_p()) { - for (i = 0; i < RARRAY_LEN(ary); i++) { - v = RARRAY_AREF(ary, i); - if (result == Qundef || rb_cmpint(rb_yield_values(2, v, result), v, result) < 0) { - result = v; - } - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + v = RARRAY_AREF(ary, i); + if (UNDEF_P(result) || rb_cmpint(rb_yield_values(2, v, result), v, result) < 0) { + result = v; + } + } } - else { - for (i = 0; i < RARRAY_LEN(ary); i++) { - v = RARRAY_AREF(ary, i); - if (result == Qundef || OPTIMIZED_CMP(v, result, cmp_opt) < 0) { - result = v; - } - } - } - if (result == Qundef) return Qnil; + else if (n > 0) { + result = RARRAY_AREF(ary, 0); + if (n > 1) { + if (FIXNUM_P(result) && CMP_OPTIMIZABLE(INTEGER)) { + return ary_min_opt_fixnum(ary, 1, result); + } + else if (STRING_P(result) && CMP_OPTIMIZABLE(STRING)) { + return ary_min_opt_string(ary, 1, result); + } + else if (RB_FLOAT_TYPE_P(result) && CMP_OPTIMIZABLE(FLOAT)) { + return ary_min_opt_float(ary, 1, result); + } + else { + return ary_min_generic(ary, 1, result); + } + } + } + if (UNDEF_P(result)) return Qnil; return result; } +/* + * call-seq: + * array.minmax -> [min_val, max_val] + * array.minmax {|a, b| ... } -> [min_val, max_val] + * + * Returns a new 2-element \Array containing the minimum and maximum values + * from +self+, either per method <tt><=></tt> or per a given block:. + * + * When no block is given, each element in +self+ must respond to method <tt><=></tt> + * with an \Integer; + * returns a new 2-element \Array containing the minimum and maximum values + * from +self+, per method <tt><=></tt>: + * + * [0, 1, 2].minmax # => [0, 2] + * + * When a block is given, the block must return an \Integer; + * the block is called <tt>self.size-1</tt> times to compare elements; + * returns a new 2-element \Array containing the minimum and maximum values + * from +self+, per the block: + * + * ['0', '00', '000'].minmax {|a, b| a.size <=> b.size } # => ["0", "000"] + * + */ +static VALUE +rb_ary_minmax(VALUE ary) +{ + if (rb_block_given_p()) { + return rb_call_super(0, NULL); + } + return rb_assoc_new(rb_ary_min(0, 0, ary), rb_ary_max(0, 0, ary)); +} + static int push_value(st_data_t key, st_data_t val, st_data_t ary) { @@ -4413,31 +6306,33 @@ push_value(st_data_t key, st_data_t val, st_data_t ary) /* * call-seq: - * ary.uniq! -> ary or nil - * ary.uniq! { |item| ... } -> ary or nil + * array.uniq! -> self or nil + * array.uniq! {|element| ... } -> self or nil * - * Removes duplicate elements from +self+. + * Removes duplicate elements from +self+, the first occurrence always being retained; + * returns +self+ if any elements removed, +nil+ otherwise. * - * If a block is given, it will use the return value of the block for - * comparison. + * With no block given, identifies and removes elements using method <tt>eql?</tt> + * to compare. * - * It compares values using their #hash and #eql? methods for efficiency. + * Returns +self+ if any elements removed: * - * +self+ is traversed in order, and the first occurrence is kept. + * a = [0, 0, 1, 1, 2, 2] + * a.uniq! # => [0, 1, 2] * - * Returns +nil+ if no changes are made (that is, no duplicates are found). + * Returns +nil+ if no elements removed. * - * a = [ "a", "a", "b", "b", "c" ] - * a.uniq! # => ["a", "b", "c"] + * With a block given, calls the block for each element; + * identifies (using method <tt>eql?</tt>) and removes + * elements for which the block returns duplicate values. * - * b = [ "a", "b", "c" ] - * b.uniq! # => nil + * Returns +self+ if any elements removed: * - * c = [["student","sam"], ["student","george"], ["teacher","matz"]] - * c.uniq! { |s| s.first } # => [["student", "sam"], ["teacher", "matz"]] + * a = ['a', 'aa', 'aaa', 'b', 'bb', 'bbb'] + * a.uniq! {|element| element.size } # => ['a', 'aa', 'aaa'] * + * Returns +nil+ if no elements removed. */ - static VALUE rb_ary_uniq_bang(VALUE ary) { @@ -4448,22 +6343,22 @@ rb_ary_uniq_bang(VALUE ary) if (RARRAY_LEN(ary) <= 1) return Qnil; if (rb_block_given_p()) - hash = ary_make_hash_by(ary); + hash = ary_make_hash_by(ary); else - hash = ary_make_hash(ary); + hash = ary_make_hash(ary); hash_size = RHASH_SIZE(hash); if (RARRAY_LEN(ary) == hash_size) { - return Qnil; + return Qnil; } rb_ary_modify_check(ary); ARY_SET_LEN(ary, 0); if (ARY_SHARED_P(ary) && !ARY_EMBED_P(ary)) { - rb_ary_unshare(ary); - FL_SET_EMBED(ary); + rb_ary_unshare(ary); + FL_SET_EMBED(ary); } ary_resize_capa(ary, hash_size); - st_foreach(rb_hash_tbl_raw(hash), push_value, ary); + rb_hash_foreach(hash, push_value, ary); ary_recycle_hash(hash); return ary; @@ -4471,22 +6366,24 @@ rb_ary_uniq_bang(VALUE ary) /* * call-seq: - * ary.uniq -> new_ary - * ary.uniq { |item| ... } -> new_ary - * - * Returns a new array by removing duplicate values in +self+. + * array.uniq -> new_array + * array.uniq {|element| ... } -> new_array * - * If a block is given, it will use the return value of the block for comparison. + * Returns a new \Array containing those elements from +self+ that are not duplicates, + * the first occurrence always being retained. * - * It compares values using their #hash and #eql? methods for efficiency. + * With no block given, identifies and omits duplicates using method <tt>eql?</tt> + * to compare: * - * +self+ is traversed in order, and the first occurrence is kept. + * a = [0, 0, 1, 1, 2, 2] + * a.uniq # => [0, 1, 2] * - * a = [ "a", "a", "b", "b", "c" ] - * a.uniq # => ["a", "b", "c"] + * With a block given, calls the block for each element; + * identifies (using method <tt>eql?</tt>) and omits duplicate values, + * that is, those elements for which the block returns the same value: * - * b = [["student","sam"], ["student","george"], ["teacher","matz"]] - * b.uniq { |s| s.first } # => [["student", "sam"], ["teacher", "matz"]] + * a = ['a', 'aa', 'aaa', 'b', 'bb', 'bbb'] + * a.uniq {|element| element.size } # => ["a", "aa", "aaa"] * */ @@ -4495,32 +6392,32 @@ rb_ary_uniq(VALUE ary) { VALUE hash, uniq; - if (RARRAY_LEN(ary) <= 1) - return rb_ary_dup(ary); - if (rb_block_given_p()) { - hash = ary_make_hash_by(ary); - uniq = rb_hash_values(hash); + if (RARRAY_LEN(ary) <= 1) { + hash = 0; + uniq = rb_ary_dup(ary); + } + else if (rb_block_given_p()) { + hash = ary_make_hash_by(ary); + uniq = rb_hash_values(hash); } else { - hash = ary_make_hash(ary); - uniq = rb_hash_values(hash); + hash = ary_make_hash(ary); + uniq = rb_hash_values(hash); + } + if (hash) { + ary_recycle_hash(hash); } - RBASIC_SET_CLASS(uniq, rb_obj_class(ary)); - ary_recycle_hash(hash); return uniq; } /* * call-seq: - * ary.compact! -> ary or nil + * array.compact! -> self or nil * - * Removes +nil+ elements from the array. + * Removes all +nil+ elements from +self+. * - * Returns +nil+ if no changes were made, otherwise returns the array. - * - * [ "a", nil, "b", nil, "c" ].compact! #=> [ "a", "b", "c" ] - * [ "a", "b", "c" ].compact! #=> nil + * Returns +self+ if any elements removed, otherwise +nil+. */ static VALUE @@ -4530,16 +6427,16 @@ rb_ary_compact_bang(VALUE ary) long n; rb_ary_modify(ary); - p = t = (VALUE *)RARRAY_CONST_PTR(ary); /* WB: no new reference */ + p = t = (VALUE *)RARRAY_CONST_PTR_TRANSIENT(ary); /* WB: no new reference */ end = p + RARRAY_LEN(ary); while (t < end) { - if (NIL_P(*t)) t++; - else *p++ = *t++; + if (NIL_P(*t)) t++; + else *p++ = *t++; } - n = p - RARRAY_CONST_PTR(ary); + n = p - RARRAY_CONST_PTR_TRANSIENT(ary); if (RARRAY_LEN(ary) == n) { - return Qnil; + return Qnil; } ary_resize_smaller(ary, n); @@ -4548,12 +6445,12 @@ rb_ary_compact_bang(VALUE ary) /* * call-seq: - * ary.compact -> new_ary + * array.compact -> new_array * - * Returns a copy of +self+ with all +nil+ elements removed. + * Returns a new \Array containing all non-+nil+ elements from +self+: * - * [ "a", nil, "b", nil, "c", nil ].compact - * #=> [ "a", "b", "c" ] + * a = [nil, 0, nil, 1, nil, 2, nil] + * a.compact # => [0, 1, 2] */ static VALUE @@ -4566,23 +6463,29 @@ rb_ary_compact(VALUE ary) /* * call-seq: - * ary.count -> int - * ary.count(obj) -> int - * ary.count { |item| block } -> int + * array.count -> an_integer + * array.count(obj) -> an_integer + * array.count {|element| ... } -> an_integer + * + * Returns a count of specified elements. * - * Returns the number of elements. + * With no argument and no block, returns the count of all elements: * - * If an argument is given, counts the number of elements which equal +obj+ - * using <code>==</code>. + * [0, 1, 2].count # => 3 + * [].count # => 0 * - * If a block is given, counts the number of elements for which the block - * returns a true value. + * With argument +obj+, returns the count of elements <tt>==</tt> to +obj+: * - * ary = [1, 2, 4, 2] - * ary.count #=> 4 - * ary.count(2) #=> 2 - * ary.count { |x| x%2 == 0 } #=> 3 + * [0, 1, 2, 0.0].count(0) # => 2 + * [0, 1, 2].count(3) # => 0 * + * With no argument and a block given, calls the block with each element; + * returns the count of elements for which the block returns a truthy value: + * + * [0, 1, 2, 3].count {|element| element > 1} # => 2 + * + * With argument +obj+ and a block given, issues a warning, ignores the block, + * and returns the count of elements <tt>==</tt> to +obj+. */ static VALUE @@ -4590,108 +6493,152 @@ rb_ary_count(int argc, VALUE *argv, VALUE ary) { long i, n = 0; - if (argc == 0) { - VALUE v; + if (rb_check_arity(argc, 0, 1) == 0) { + VALUE v; - if (!rb_block_given_p()) - return LONG2NUM(RARRAY_LEN(ary)); + if (!rb_block_given_p()) + return LONG2NUM(RARRAY_LEN(ary)); - for (i = 0; i < RARRAY_LEN(ary); i++) { - v = RARRAY_AREF(ary, i); - if (RTEST(rb_yield(v))) n++; - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + v = RARRAY_AREF(ary, i); + if (RTEST(rb_yield(v))) n++; + } } else { - VALUE obj; + VALUE obj = argv[0]; - rb_scan_args(argc, argv, "1", &obj); - if (rb_block_given_p()) { - rb_warn("given block not used"); - } - for (i = 0; i < RARRAY_LEN(ary); i++) { - if (rb_equal(RARRAY_AREF(ary, i), obj)) n++; - } + if (rb_block_given_p()) { + rb_warn("given block not used"); + } + for (i = 0; i < RARRAY_LEN(ary); i++) { + if (rb_equal(RARRAY_AREF(ary, i), obj)) n++; + } } return LONG2NUM(n); } static VALUE -flatten(VALUE ary, int level, int *modified) +flatten(VALUE ary, int level) { - long i = 0; - VALUE stack, result, tmp, elt; - st_table *memo; + long i; + VALUE stack, result, tmp = 0, elt, vmemo; + st_table *memo = 0; st_data_t id; - stack = ary_new(0, ARY_DEFAULT_SIZE); + for (i = 0; i < RARRAY_LEN(ary); i++) { + elt = RARRAY_AREF(ary, i); + tmp = rb_check_array_type(elt); + if (!NIL_P(tmp)) { + break; + } + } + if (i == RARRAY_LEN(ary)) { + return ary; + } + result = ary_new(0, RARRAY_LEN(ary)); - memo = st_init_numtable(); - st_insert(memo, (st_data_t)ary, (st_data_t)Qtrue); - *modified = 0; + ary_memcpy(result, 0, i, RARRAY_CONST_PTR_TRANSIENT(ary)); + ARY_SET_LEN(result, i); + + stack = ary_new(0, ARY_DEFAULT_SIZE); + rb_ary_push(stack, ary); + rb_ary_push(stack, LONG2NUM(i + 1)); + + if (level < 0) { + vmemo = rb_hash_new(); + RBASIC_CLEAR_CLASS(vmemo); + memo = st_init_numtable(); + rb_hash_st_table_set(vmemo, memo); + st_insert(memo, (st_data_t)ary, (st_data_t)Qtrue); + st_insert(memo, (st_data_t)tmp, (st_data_t)Qtrue); + } + + ary = tmp; + i = 0; while (1) { - while (i < RARRAY_LEN(ary)) { - elt = RARRAY_AREF(ary, i++); - if (level >= 0 && RARRAY_LEN(stack) / 2 >= level) { - rb_ary_push(result, elt); - continue; - } - tmp = rb_check_array_type(elt); - if (RBASIC(result)->klass) { - rb_raise(rb_eRuntimeError, "flatten reentered"); - } - if (NIL_P(tmp)) { - rb_ary_push(result, elt); - } - else { - *modified = 1; - id = (st_data_t)tmp; - if (st_lookup(memo, id, 0)) { - st_free_table(memo); - rb_raise(rb_eArgError, "tried to flatten recursive array"); - } - st_insert(memo, id, (st_data_t)Qtrue); - rb_ary_push(stack, ary); - rb_ary_push(stack, LONG2NUM(i)); - ary = tmp; - i = 0; - } - } - if (RARRAY_LEN(stack) == 0) { - break; - } - id = (st_data_t)ary; - st_delete(memo, &id, 0); - tmp = rb_ary_pop(stack); - i = NUM2LONG(tmp); - ary = rb_ary_pop(stack); - } - - st_free_table(memo); - - RBASIC_SET_CLASS(result, rb_obj_class(ary)); + while (i < RARRAY_LEN(ary)) { + elt = RARRAY_AREF(ary, i++); + if (level >= 0 && RARRAY_LEN(stack) / 2 >= level) { + rb_ary_push(result, elt); + continue; + } + tmp = rb_check_array_type(elt); + if (RBASIC(result)->klass) { + if (memo) { + RB_GC_GUARD(vmemo); + st_clear(memo); + } + rb_raise(rb_eRuntimeError, "flatten reentered"); + } + if (NIL_P(tmp)) { + rb_ary_push(result, elt); + } + else { + if (memo) { + id = (st_data_t)tmp; + if (st_is_member(memo, id)) { + st_clear(memo); + rb_raise(rb_eArgError, "tried to flatten recursive array"); + } + st_insert(memo, id, (st_data_t)Qtrue); + } + rb_ary_push(stack, ary); + rb_ary_push(stack, LONG2NUM(i)); + ary = tmp; + i = 0; + } + } + if (RARRAY_LEN(stack) == 0) { + break; + } + if (memo) { + id = (st_data_t)ary; + st_delete(memo, &id, 0); + } + tmp = rb_ary_pop(stack); + i = NUM2LONG(tmp); + ary = rb_ary_pop(stack); + } + + if (memo) { + st_clear(memo); + } + + RBASIC_SET_CLASS(result, rb_cArray); return result; } /* * call-seq: - * ary.flatten! -> ary or nil - * ary.flatten!(level) -> ary or nil + * array.flatten! -> self or nil + * array.flatten!(level) -> self or nil + * + * Replaces each nested \Array in +self+ with the elements from that \Array; + * returns +self+ if any changes, +nil+ otherwise. + * + * With non-negative \Integer argument +level+, flattens recursively through +level+ levels: * - * Flattens +self+ in place. + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten!(1) # => [0, 1, [2, 3], 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten!(2) # => [0, 1, 2, 3, 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten!(3) # => [0, 1, 2, 3, 4, 5] + * [0, 1, 2].flatten!(1) # => nil * - * Returns +nil+ if no modifications were made (i.e., the array contains no - * subarrays.) + * With no argument, a +nil+ argument, or with negative argument +level+, flattens all levels: * - * The optional +level+ argument determines the level of recursion to flatten. + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten! # => [0, 1, 2, 3, 4, 5] + * [0, 1, 2].flatten! # => nil + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten!(-1) # => [0, 1, 2, 3, 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten!(-2) # => [0, 1, 2, 3, 4, 5] + * [0, 1, 2].flatten!(-1) # => nil * - * a = [ 1, 2, [3, [4, 5] ] ] - * a.flatten! #=> [1, 2, 3, 4, 5] - * a.flatten! #=> nil - * a #=> [1, 2, 3, 4, 5] - * a = [ 1, 2, [3, [4, 5] ] ] - * a.flatten!(1) #=> [1, 2, 3, [4, 5]] */ static VALUE @@ -4700,15 +6647,14 @@ rb_ary_flatten_bang(int argc, VALUE *argv, VALUE ary) int mod = 0, level = -1; VALUE result, lv; - rb_scan_args(argc, argv, "01", &lv); + lv = (rb_check_arity(argc, 0, 1) ? argv[0] : Qnil); rb_ary_modify_check(ary); if (!NIL_P(lv)) level = NUM2INT(lv); if (level == 0) return Qnil; - result = flatten(ary, level, &mod); - if (mod == 0) { - ary_discard(result); - return Qnil; + result = flatten(ary, level); + if (result == ary) { + return Qnil; } if (!(mod = ARY_EMBED_P(result))) rb_obj_freeze(result); rb_ary_replace(ary, result); @@ -4719,284 +6665,212 @@ rb_ary_flatten_bang(int argc, VALUE *argv, VALUE ary) /* * call-seq: - * ary.flatten -> new_ary - * ary.flatten(level) -> new_ary - * - * Returns a new array that is a one-dimensional flattening of +self+ - * (recursively). - * - * That is, for every element that is an array, extract its elements into - * the new array. + * array.flatten -> new_array + * array.flatten(level) -> new_array + * + * Returns a new \Array that is a recursive flattening of +self+: + * - Each non-Array element is unchanged. + * - Each \Array is replaced by its individual elements. + * + * With non-negative \Integer argument +level+, flattens recursively through +level+ levels: + * + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(0) # => [0, [1, [2, 3], 4], 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(1) # => [0, 1, [2, 3], 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(2) # => [0, 1, 2, 3, 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(3) # => [0, 1, 2, 3, 4, 5] + * + * With no argument, a +nil+ argument, or with negative argument +level+, flattens all levels: + * + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten # => [0, 1, 2, 3, 4, 5] + * [0, 1, 2].flatten # => [0, 1, 2] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(-1) # => [0, 1, 2, 3, 4, 5] + * a = [ 0, [ 1, [2, 3], 4 ], 5 ] + * a.flatten(-2) # => [0, 1, 2, 3, 4, 5] + * [0, 1, 2].flatten(-1) # => [0, 1, 2] * - * The optional +level+ argument determines the level of recursion to - * flatten. - * - * s = [ 1, 2, 3 ] #=> [1, 2, 3] - * t = [ 4, 5, 6, [7, 8] ] #=> [4, 5, 6, [7, 8]] - * a = [ s, t, 9, 10 ] #=> [[1, 2, 3], [4, 5, 6, [7, 8]], 9, 10] - * a.flatten #=> [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] - * a = [ 1, 2, [3, [4, 5] ] ] - * a.flatten(1) #=> [1, 2, 3, [4, 5]] */ static VALUE rb_ary_flatten(int argc, VALUE *argv, VALUE ary) { - int mod = 0, level = -1; - VALUE result, lv; + int level = -1; + VALUE result; - rb_scan_args(argc, argv, "01", &lv); - if (!NIL_P(lv)) level = NUM2INT(lv); - if (level == 0) return ary_make_shared_copy(ary); + if (rb_check_arity(argc, 0, 1) && !NIL_P(argv[0])) { + level = NUM2INT(argv[0]); + if (level == 0) return ary_make_shared_copy(ary); + } - result = flatten(ary, level, &mod); - OBJ_INFECT(result, ary); + result = flatten(ary, level); + if (result == ary) { + result = ary_make_shared_copy(ary); + } return result; } -#define OPTHASH_GIVEN_P(opts) \ - (argc > 0 && !NIL_P((opts) = rb_check_hash_type(argv[argc-1])) && (--argc, 1)) -static ID id_random; - #define RAND_UPTO(max) (long)rb_random_ulong_limited((randgen), (max)-1) -/* - * call-seq: - * ary.shuffle! -> ary - * ary.shuffle!(random: rng) -> ary - * - * Shuffles elements in +self+ in place. - * - * a = [ 1, 2, 3 ] #=> [1, 2, 3] - * a.shuffle! #=> [2, 3, 1] - * a #=> [2, 3, 1] - * - * The optional +rng+ argument will be used as the random number generator. - * - * a.shuffle!(random: Random.new(1)) #=> [1, 3, 2] - */ - static VALUE -rb_ary_shuffle_bang(int argc, VALUE *argv, VALUE ary) +rb_ary_shuffle_bang(rb_execution_context_t *ec, VALUE ary, VALUE randgen) { - VALUE opts, randgen = rb_cRandom; long i, len; - if (OPTHASH_GIVEN_P(opts)) { - VALUE rnd; - ID keyword_ids[1]; - - keyword_ids[0] = id_random; - rb_get_kwargs(opts, keyword_ids, 0, 1, &rnd); - if (rnd != Qundef) { - randgen = rnd; - } - } - rb_check_arity(argc, 0, 0); rb_ary_modify(ary); i = len = RARRAY_LEN(ary); RARRAY_PTR_USE(ary, ptr, { - while (i) { - long j = RAND_UPTO(i); - VALUE tmp; - if (len != RARRAY_LEN(ary) || ptr != RARRAY_CONST_PTR(ary)) { - rb_raise(rb_eRuntimeError, "modified during shuffle"); - } - tmp = ptr[--i]; - ptr[i] = ptr[j]; - ptr[j] = tmp; - } + while (i) { + long j = RAND_UPTO(i); + VALUE tmp; + if (len != RARRAY_LEN(ary) || ptr != RARRAY_CONST_PTR_TRANSIENT(ary)) { + rb_raise(rb_eRuntimeError, "modified during shuffle"); + } + tmp = ptr[--i]; + ptr[i] = ptr[j]; + ptr[j] = tmp; + } }); /* WB: no new reference */ return ary; } - -/* - * call-seq: - * ary.shuffle -> new_ary - * ary.shuffle(random: rng) -> new_ary - * - * Returns a new array with elements of +self+ shuffled. - * - * a = [ 1, 2, 3 ] #=> [1, 2, 3] - * a.shuffle #=> [2, 3, 1] - * a #=> [1, 2, 3] - * - * The optional +rng+ argument will be used as the random number generator. - * - * a.shuffle(random: Random.new(1)) #=> [1, 3, 2] - */ - static VALUE -rb_ary_shuffle(int argc, VALUE *argv, VALUE ary) +rb_ary_shuffle(rb_execution_context_t *ec, VALUE ary, VALUE randgen) { ary = rb_ary_dup(ary); - rb_ary_shuffle_bang(argc, argv, ary); + rb_ary_shuffle_bang(ec, ary, randgen); return ary; } - -/* - * call-seq: - * ary.sample -> obj - * ary.sample(random: rng) -> obj - * ary.sample(n) -> new_ary - * ary.sample(n, random: rng) -> new_ary - * - * Choose a random element or +n+ random elements from the array. - * - * The elements are chosen by using random and unique indices into the array - * in order to ensure that an element doesn't repeat itself unless the array - * already contained duplicate elements. - * - * If the array is empty the first form returns +nil+ and the second form - * returns an empty array. - * - * The optional +rng+ argument will be used as the random number generator. - * - * a = [ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 ] - * a.sample #=> 7 - * a.sample(4) #=> [6, 4, 2, 5] - */ - - static VALUE -rb_ary_sample(int argc, VALUE *argv, VALUE ary) +ary_sample(rb_execution_context_t *ec, VALUE ary, VALUE randgen, VALUE nv, VALUE to_array) { - VALUE nv, result; - VALUE opts, randgen = rb_cRandom; + VALUE result; long n, len, i, j, k, idx[10]; long rnds[numberof(idx)]; long memo_threshold; - if (OPTHASH_GIVEN_P(opts)) { - VALUE rnd; - ID keyword_ids[1]; - - keyword_ids[0] = id_random; - rb_get_kwargs(opts, keyword_ids, 0, 1, &rnd); - if (rnd != Qundef) { - randgen = rnd; - } - } len = RARRAY_LEN(ary); - if (argc == 0) { - if (len < 2) - i = 0; - else - i = RAND_UPTO(len); + if (!to_array) { + if (len < 2) + i = 0; + else + i = RAND_UPTO(len); - return rb_ary_elt(ary, i); + return rb_ary_elt(ary, i); } - rb_scan_args(argc, argv, "1", &nv); n = NUM2LONG(nv); if (n < 0) rb_raise(rb_eArgError, "negative sample number"); if (n > len) n = len; if (n <= numberof(idx)) { - for (i = 0; i < n; ++i) { - rnds[i] = RAND_UPTO(len - i); - } + for (i = 0; i < n; ++i) { + rnds[i] = RAND_UPTO(len - i); + } } k = len; len = RARRAY_LEN(ary); if (len < k && n <= numberof(idx)) { - for (i = 0; i < n; ++i) { - if (rnds[i] >= len) return rb_ary_new_capa(0); - } + for (i = 0; i < n; ++i) { + if (rnds[i] >= len) return rb_ary_new_capa(0); + } } if (n > len) n = len; switch (n) { case 0: - return rb_ary_new_capa(0); + return rb_ary_new_capa(0); case 1: - i = rnds[0]; - return rb_ary_new_from_values(1, &RARRAY_AREF(ary, i)); + i = rnds[0]; + return rb_ary_new_from_args(1, RARRAY_AREF(ary, i)); case 2: - i = rnds[0]; - j = rnds[1]; - if (j >= i) j++; - return rb_ary_new_from_args(2, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j)); + i = rnds[0]; + j = rnds[1]; + if (j >= i) j++; + return rb_ary_new_from_args(2, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j)); case 3: - i = rnds[0]; - j = rnds[1]; - k = rnds[2]; - { - long l = j, g = i; - if (j >= i) l = i, g = ++j; - if (k >= l && (++k >= g)) ++k; - } - return rb_ary_new_from_args(3, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j), RARRAY_AREF(ary, k)); + i = rnds[0]; + j = rnds[1]; + k = rnds[2]; + { + long l = j, g = i; + if (j >= i) l = i, g = ++j; + if (k >= l && (++k >= g)) ++k; + } + return rb_ary_new_from_args(3, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j), RARRAY_AREF(ary, k)); } memo_threshold = - len < 2560 ? len / 128 : - len < 5120 ? len / 64 : - len < 10240 ? len / 32 : - len / 16; + len < 2560 ? len / 128 : + len < 5120 ? len / 64 : + len < 10240 ? len / 32 : + len / 16; if (n <= numberof(idx)) { - long sorted[numberof(idx)]; - sorted[0] = idx[0] = rnds[0]; - for (i=1; i<n; i++) { - k = rnds[i]; - for (j = 0; j < i; ++j) { - if (k < sorted[j]) break; - ++k; - } - memmove(&sorted[j+1], &sorted[j], sizeof(sorted[0])*(i-j)); - sorted[j] = idx[i] = k; - } - result = rb_ary_new_capa(n); - RARRAY_PTR_USE(result, ptr_result, { - for (i=0; i<n; i++) { - ptr_result[i] = RARRAY_AREF(ary, idx[i]); - } - }); + long sorted[numberof(idx)]; + sorted[0] = idx[0] = rnds[0]; + for (i=1; i<n; i++) { + k = rnds[i]; + for (j = 0; j < i; ++j) { + if (k < sorted[j]) break; + ++k; + } + memmove(&sorted[j+1], &sorted[j], sizeof(sorted[0])*(i-j)); + sorted[j] = idx[i] = k; + } + result = rb_ary_new_capa(n); + RARRAY_PTR_USE_TRANSIENT(result, ptr_result, { + for (i=0; i<n; i++) { + ptr_result[i] = RARRAY_AREF(ary, idx[i]); + } + }); } else if (n <= memo_threshold / 2) { - long max_idx = 0; + long max_idx = 0; #undef RUBY_UNTYPED_DATA_WARNING #define RUBY_UNTYPED_DATA_WARNING 0 - VALUE vmemo = Data_Wrap_Struct(0, 0, st_free_table, 0); - st_table *memo = st_init_numtable_with_size(n); - DATA_PTR(vmemo) = memo; - result = rb_ary_new_capa(n); - RARRAY_PTR_USE(result, ptr_result, { - for (i=0; i<n; i++) { - long r = RAND_UPTO(len-i) + i; - ptr_result[i] = r; - if (r > max_idx) max_idx = r; - } - len = RARRAY_LEN(ary); - if (len <= max_idx) n = 0; - else if (n > len) n = len; - RARRAY_PTR_USE(ary, ptr_ary, { - for (i=0; i<n; i++) { - long j2 = j = ptr_result[i]; - long i2 = i; - st_data_t value; - if (st_lookup(memo, (st_data_t)i, &value)) i2 = (long)value; - if (st_lookup(memo, (st_data_t)j, &value)) j2 = (long)value; - st_insert(memo, (st_data_t)j, (st_data_t)i2); - ptr_result[i] = ptr_ary[j2]; - } - }); - }); - DATA_PTR(vmemo) = 0; - st_free_table(memo); + VALUE vmemo = Data_Wrap_Struct(0, 0, st_free_table, 0); + st_table *memo = st_init_numtable_with_size(n); + DATA_PTR(vmemo) = memo; + result = rb_ary_new_capa(n); + RARRAY_PTR_USE(result, ptr_result, { + for (i=0; i<n; i++) { + long r = RAND_UPTO(len-i) + i; + ptr_result[i] = r; + if (r > max_idx) max_idx = r; + } + len = RARRAY_LEN(ary); + if (len <= max_idx) n = 0; + else if (n > len) n = len; + RARRAY_PTR_USE_TRANSIENT(ary, ptr_ary, { + for (i=0; i<n; i++) { + long j2 = j = ptr_result[i]; + long i2 = i; + st_data_t value; + if (st_lookup(memo, (st_data_t)i, &value)) i2 = (long)value; + if (st_lookup(memo, (st_data_t)j, &value)) j2 = (long)value; + st_insert(memo, (st_data_t)j, (st_data_t)i2); + ptr_result[i] = ptr_ary[j2]; + } + }); + }); + DATA_PTR(vmemo) = 0; + st_free_table(memo); } else { - result = rb_ary_dup(ary); - RBASIC_CLEAR_CLASS(result); - RB_GC_GUARD(ary); - RARRAY_PTR_USE(result, ptr_result, { - for (i=0; i<n; i++) { - j = RAND_UPTO(len-i) + i; - nv = ptr_result[j]; - ptr_result[j] = ptr_result[i]; - ptr_result[i] = nv; - } - }); - RBASIC_SET_CLASS_RAW(result, rb_cArray); + result = rb_ary_dup(ary); + RBASIC_CLEAR_CLASS(result); + RB_GC_GUARD(ary); + RARRAY_PTR_USE(result, ptr_result, { + for (i=0; i<n; i++) { + j = RAND_UPTO(len-i) + i; + nv = ptr_result[j]; + ptr_result[j] = ptr_result[i]; + ptr_result[i] = nv; + } + }); + RBASIC_SET_CLASS_RAW(result, rb_cArray); } ARY_SET_LEN(result, n); @@ -5004,15 +6878,21 @@ rb_ary_sample(int argc, VALUE *argv, VALUE ary) } static VALUE +ary_sample0(rb_execution_context_t *ec, VALUE ary) +{ + return ary_sample(ec, ary, rb_cRandom, Qfalse, Qfalse); +} + +static VALUE rb_ary_cycle_size(VALUE self, VALUE args, VALUE eobj) { long mul; VALUE n = Qnil; if (args && (RARRAY_LEN(args) > 0)) { - n = RARRAY_AREF(args, 0); + n = RARRAY_AREF(args, 0); } if (RARRAY_LEN(self) == 0) return INT2FIX(0); - if (n == Qnil) return DBL2NUM(HUGE_VAL); + if (NIL_P(n)) return DBL2NUM(HUGE_VAL); mul = NUM2LONG(n); if (mul <= 0) return INT2FIX(0); n = LONG2FIX(mul); @@ -5021,38 +6901,50 @@ rb_ary_cycle_size(VALUE self, VALUE args, VALUE eobj) /* * call-seq: - * ary.cycle(n=nil) { |obj| block } -> nil - * ary.cycle(n=nil) -> Enumerator + * array.cycle {|element| ... } -> nil + * array.cycle(count) {|element| ... } -> nil + * array.cycle -> new_enumerator + * array.cycle(count) -> new_enumerator * - * Calls the given block for each element +n+ times or forever if +nil+ is - * given. + * When called with positive \Integer argument +count+ and a block, + * calls the block with each element, then does so again, + * until it has done so +count+ times; returns +nil+: * - * Does nothing if a non-positive number is given or the array is empty. + * output = [] + * [0, 1].cycle(2) {|element| output.push(element) } # => nil + * output # => [0, 1, 0, 1] * - * Returns +nil+ if the loop has finished without getting interrupted. + * If +count+ is zero or negative, does not call the block: * - * If no block is given, an Enumerator is returned instead. + * [0, 1].cycle(0) {|element| fail 'Cannot happen' } # => nil + * [0, 1].cycle(-1) {|element| fail 'Cannot happen' } # => nil * - * a = ["a", "b", "c"] - * a.cycle { |x| puts x } # print, a, b, c, a, b, c,.. forever. - * a.cycle(2) { |x| puts x } # print, a, b, c, a, b, c. + * When a block is given, and argument is omitted or +nil+, cycles forever: + * + * # Prints 0 and 1 forever. + * [0, 1].cycle {|element| puts element } + * [0, 1].cycle(nil) {|element| puts element } + * + * When no block is given, returns a new \Enumerator: + * + * [0, 1].cycle(2) # => #<Enumerator: [0, 1]:cycle(2)> + * [0, 1].cycle # => # => #<Enumerator: [0, 1]:cycle> + * [0, 1].cycle.first(5) # => [0, 1, 0, 1, 0] * */ - static VALUE rb_ary_cycle(int argc, VALUE *argv, VALUE ary) { long n, i; - VALUE nv = Qnil; - rb_scan_args(argc, argv, "01", &nv); + rb_check_arity(argc, 0, 1); RETURN_SIZED_ENUMERATOR(ary, argc, argv, rb_ary_cycle_size); - if (NIL_P(nv)) { + if (argc == 0 || NIL_P(argv[0])) { n = -1; } else { - n = NUM2LONG(nv); + n = NUM2LONG(argv[0]); if (n <= 0) return Qnil; } @@ -5064,11 +6956,6 @@ rb_ary_cycle(int argc, VALUE *argv, VALUE ary) return Qnil; } -#define tmpbuf(n, size) rb_str_tmp_new((n)*(size)) -#define tmpbuf_discard(s) (rb_str_resize((s), 0L), RBASIC_SET_CLASS_RAW(s, rb_cString)) -#define tmpary(n) rb_ary_tmp_new(n) -#define tmpary_discard(a) (ary_discard(a), RBASIC_SET_CLASS_RAW(a, rb_cArray)) - /* * Build a ruby array of the corresponding values and yield it to the * associated block. @@ -5078,11 +6965,9 @@ static int yield_indexed_values(const VALUE values, const long r, const long *const p) { const VALUE result = rb_ary_new2(r); - VALUE *const result_array = RARRAY_PTR(result); - const VALUE *const values_array = RARRAY_CONST_PTR(values); long i; - for (i = 0; i < r; i++) result_array[i] = values_array[p[i]]; + for (i = 0; i < r; i++) ARY_SET(result, i, RARRAY_AREF(values, p[i])); ARY_SET_LEN(result, r); rb_yield(result); return !RBASIC(values)->klass; @@ -5106,52 +6991,52 @@ permute0(const long n, const long r, long *const p, char *const used, const VALU long i = 0, index = 0; for (;;) { - const char *const unused = memchr(&used[i], 0, n-i); - if (!unused) { - if (!index) break; - i = p[--index]; /* pop index */ - used[i++] = 0; /* index unused */ - } - else { - i = unused - used; - p[index] = i; - used[i] = 1; /* mark index used */ - ++index; - if (index < r-1) { /* if not done yet */ - p[index] = i = 0; - continue; - } - for (i = 0; i < n; ++i) { - if (used[i]) continue; - p[index] = i; - if (!yield_indexed_values(values, r, p)) { - rb_raise(rb_eRuntimeError, "permute reentered"); - } - } - i = p[--index]; /* pop index */ - used[i] = 0; /* index unused */ - p[index] = ++i; - } + const char *const unused = memchr(&used[i], 0, n-i); + if (!unused) { + if (!index) break; + i = p[--index]; /* pop index */ + used[i++] = 0; /* index unused */ + } + else { + i = unused - used; + p[index] = i; + used[i] = 1; /* mark index used */ + ++index; + if (index < r-1) { /* if not done yet */ + p[index] = i = 0; + continue; + } + for (i = 0; i < n; ++i) { + if (used[i]) continue; + p[index] = i; + if (!yield_indexed_values(values, r, p)) { + rb_raise(rb_eRuntimeError, "permute reentered"); + } + } + i = p[--index]; /* pop index */ + used[i] = 0; /* index unused */ + p[index] = ++i; + } } } /* * Returns the product of from, from-1, ..., from - how_many + 1. - * http://en.wikipedia.org/wiki/Pochhammer_symbol + * https://en.wikipedia.org/wiki/Pochhammer_symbol */ static VALUE descending_factorial(long from, long how_many) { VALUE cnt; if (how_many > 0) { - cnt = LONG2FIX(from); - while (--how_many > 0) { - long v = --from; - cnt = rb_int_mul(cnt, LONG2FIX(v)); - } + cnt = LONG2FIX(from); + while (--how_many > 0) { + long v = --from; + cnt = rb_int_mul(cnt, LONG2FIX(v)); + } } else { - cnt = LONG2FIX(how_many == 0); + cnt = LONG2FIX(how_many == 0); } return cnt; } @@ -5162,18 +7047,18 @@ binomial_coefficient(long comb, long size) VALUE r; long i; if (comb > size-comb) { - comb = size-comb; + comb = size-comb; } if (comb < 0) { - return LONG2FIX(0); + return LONG2FIX(0); } else if (comb == 0) { - return LONG2FIX(1); + return LONG2FIX(1); } r = LONG2FIX(size); for (i = 1; i < comb; ++i) { - r = rb_int_mul(r, LONG2FIX(size - i)); - r = rb_int_idiv(r, LONG2FIX(i + 1)); + r = rb_int_mul(r, LONG2FIX(size - i)); + r = rb_int_idiv(r, LONG2FIX(i + 1)); } return r; } @@ -5189,66 +7074,118 @@ rb_ary_permutation_size(VALUE ary, VALUE args, VALUE eobj) /* * call-seq: - * ary.permutation { |p| block } -> ary - * ary.permutation -> Enumerator - * ary.permutation(n) { |p| block } -> ary - * ary.permutation(n) -> Enumerator + * array.permutation {|element| ... } -> self + * array.permutation(n) {|element| ... } -> self + * array.permutation -> new_enumerator + * array.permutation(n) -> new_enumerator + * + * When invoked with a block, yield all permutations of elements of +self+; returns +self+. + * The order of permutations is indeterminate. + * + * When a block and an in-range positive \Integer argument +n+ (<tt>0 < n <= self.size</tt>) + * are given, calls the block with all +n+-tuple permutations of +self+. + * + * Example: + * + * a = [0, 1, 2] + * a.permutation(2) {|permutation| p permutation } + * + * Output: + * + * [0, 1] + * [0, 2] + * [1, 0] + * [1, 2] + * [2, 0] + * [2, 1] + * + * Another example: + * + * a = [0, 1, 2] + * a.permutation(3) {|permutation| p permutation } + * + * Output: + * + * [0, 1, 2] + * [0, 2, 1] + * [1, 0, 2] + * [1, 2, 0] + * [2, 0, 1] + * [2, 1, 0] + * + * When +n+ is zero, calls the block once with a new empty \Array: + * + * a = [0, 1, 2] + * a.permutation(0) {|permutation| p permutation } + * + * Output: * - * When invoked with a block, yield all permutations of length +n+ of the - * elements of the array, then return the array itself. + * [] * - * If +n+ is not specified, yield all permutations of all elements. + * When +n+ is out of range (negative or larger than <tt>self.size</tt>), + * does not call the block: * - * The implementation makes no guarantees about the order in which the - * permutations are yielded. + * a = [0, 1, 2] + * a.permutation(-1) {|permutation| fail 'Cannot happen' } + * a.permutation(4) {|permutation| fail 'Cannot happen' } * - * If no block is given, an Enumerator is returned instead. + * When a block given but no argument, + * behaves the same as <tt>a.permutation(a.size)</tt>: * - * Examples: + * a = [0, 1, 2] + * a.permutation {|permutation| p permutation } + * + * Output: + * + * [0, 1, 2] + * [0, 2, 1] + * [1, 0, 2] + * [1, 2, 0] + * [2, 0, 1] + * [2, 1, 0] + * + * Returns a new \Enumerator if no block given: + * + * a = [0, 1, 2] + * a.permutation # => #<Enumerator: [0, 1, 2]:permutation> + * a.permutation(2) # => #<Enumerator: [0, 1, 2]:permutation(2)> * - * a = [1, 2, 3] - * a.permutation.to_a #=> [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]] - * a.permutation(1).to_a #=> [[1],[2],[3]] - * a.permutation(2).to_a #=> [[1,2],[1,3],[2,1],[2,3],[3,1],[3,2]] - * a.permutation(3).to_a #=> [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]] - * a.permutation(0).to_a #=> [[]] # one permutation of length 0 - * a.permutation(4).to_a #=> [] # no permutations of length 4 */ static VALUE rb_ary_permutation(int argc, VALUE *argv, VALUE ary) { - VALUE num; long r, n, i; n = RARRAY_LEN(ary); /* Array length */ RETURN_SIZED_ENUMERATOR(ary, argc, argv, rb_ary_permutation_size); /* Return enumerator if no block */ - rb_scan_args(argc, argv, "01", &num); - r = NIL_P(num) ? n : NUM2LONG(num); /* Permutation size from argument */ + r = n; + if (rb_check_arity(argc, 0, 1) && !NIL_P(argv[0])) + r = NUM2LONG(argv[0]); /* Permutation size from argument */ if (r < 0 || n < r) { - /* no permutations: yield nothing */ + /* no permutations: yield nothing */ } else if (r == 0) { /* exactly one permutation: the zero-length array */ - rb_yield(rb_ary_new2(0)); + rb_yield(rb_ary_new2(0)); } else if (r == 1) { /* this is a special, easy case */ - for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); + } } else { /* this is the general case */ - volatile VALUE t0; - long *p = ALLOCV_N(long, t0, r+roomof(n, sizeof(long))); - char *used = (char*)(p + r); - VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ - RBASIC_CLEAR_CLASS(ary0); + volatile VALUE t0; + long *p = ALLOCV_N(long, t0, r+roomof(n, sizeof(long))); + char *used = (char*)(p + r); + VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ + RBASIC_CLEAR_CLASS(ary0); - MEMZERO(used, char, n); /* initialize array */ + MEMZERO(used, char, n); /* initialize array */ - permute0(n, r, p, used, ary0); /* compute and yield permutations */ - ALLOCV_END(t0); - RBASIC_SET_CLASS_RAW(ary0, rb_cArray); + permute0(n, r, p, used, ary0); /* compute and yield permutations */ + ALLOCV_END(t0); + RBASIC_SET_CLASS_RAW(ary0, rb_cArray); } return ary; } @@ -5261,16 +7198,16 @@ combinate0(const long len, const long n, long *const stack, const VALUE values) MEMZERO(stack+1, long, n); stack[0] = -1; for (;;) { - for (lev++; lev < n; lev++) { - stack[lev+1] = stack[lev]+1; - } - if (!yield_indexed_values(values, n, stack+1)) { - rb_raise(rb_eRuntimeError, "combination reentered"); - } - do { - if (lev == 0) return; - stack[lev--]++; - } while (stack[lev+1]+n == len+lev+1); + for (lev++; lev < n; lev++) { + stack[lev+1] = stack[lev]+1; + } + if (!yield_indexed_values(values, n, stack+1)) { + rb_raise(rb_eRuntimeError, "combination reentered"); + } + do { + if (lev == 0) return; + stack[lev--]++; + } while (stack[lev+1]+n == len+lev+1); } } @@ -5285,26 +7222,55 @@ rb_ary_combination_size(VALUE ary, VALUE args, VALUE eobj) /* * call-seq: - * ary.combination(n) { |c| block } -> ary - * ary.combination(n) -> Enumerator + * array.combination(n) {|element| ... } -> self + * array.combination(n) -> new_enumerator + * + * Calls the block, if given, with combinations of elements of +self+; + * returns +self+. The order of combinations is indeterminate. + * + * When a block and an in-range positive \Integer argument +n+ (<tt>0 < n <= self.size</tt>) + * are given, calls the block with all +n+-tuple combinations of +self+. + * + * Example: + * + * a = [0, 1, 2] + * a.combination(2) {|combination| p combination } + * + * Output: * - * When invoked with a block, yields all combinations of length +n+ of elements - * from the array and then returns the array itself. + * [0, 1] + * [0, 2] + * [1, 2] * - * The implementation makes no guarantees about the order in which the - * combinations are yielded. + * Another example: * - * If no block is given, an Enumerator is returned instead. + * a = [0, 1, 2] + * a.combination(3) {|combination| p combination } * - * Examples: + * Output: * - * a = [1, 2, 3, 4] - * a.combination(1).to_a #=> [[1],[2],[3],[4]] - * a.combination(2).to_a #=> [[1,2],[1,3],[1,4],[2,3],[2,4],[3,4]] - * a.combination(3).to_a #=> [[1,2,3],[1,2,4],[1,3,4],[2,3,4]] - * a.combination(4).to_a #=> [[1,2,3,4]] - * a.combination(0).to_a #=> [[]] # one combination of length 0 - * a.combination(5).to_a #=> [] # no combinations of length 5 + * [0, 1, 2] + * + * When +n+ is zero, calls the block once with a new empty \Array: + * + * a = [0, 1, 2] + * a1 = a.combination(0) {|combination| p combination } + * + * Output: + * + * [] + * + * When +n+ is out of range (negative or larger than <tt>self.size</tt>), + * does not call the block: + * + * a = [0, 1, 2] + * a.combination(-1) {|combination| fail 'Cannot happen' } + * a.combination(4) {|combination| fail 'Cannot happen' } + * + * Returns a new \Enumerator if no block given: + * + * a = [0, 1, 2] + * a.combination(2) # => #<Enumerator: [0, 1, 2]:combination(2)> * */ @@ -5317,25 +7283,25 @@ rb_ary_combination(VALUE ary, VALUE num) RETURN_SIZED_ENUMERATOR(ary, 1, &num, rb_ary_combination_size); len = RARRAY_LEN(ary); if (n < 0 || len < n) { - /* yield nothing */ + /* yield nothing */ } else if (n == 0) { - rb_yield(rb_ary_new2(0)); + rb_yield(rb_ary_new2(0)); } else if (n == 1) { - for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); + } } else { - VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ - volatile VALUE t0; - long *stack = ALLOCV_N(long, t0, n+1); + VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ + volatile VALUE t0; + long *stack = ALLOCV_N(long, t0, n+1); - RBASIC_CLEAR_CLASS(ary0); - combinate0(len, n, stack, ary0); - ALLOCV_END(t0); - RBASIC_SET_CLASS_RAW(ary0, rb_cArray); + RBASIC_CLEAR_CLASS(ary0); + combinate0(len, n, stack, ary0); + ALLOCV_END(t0); + RBASIC_SET_CLASS_RAW(ary0, rb_cArray); } return ary; } @@ -5359,19 +7325,19 @@ rpermute0(const long n, const long r, long *const p, const VALUE values) p[index] = i; for (;;) { - if (++index < r-1) { - p[index] = i = 0; - continue; - } - for (i = 0; i < n; ++i) { - p[index] = i; - if (!yield_indexed_values(values, r, p)) { - rb_raise(rb_eRuntimeError, "repeated permute reentered"); - } - } - do { - if (index <= 0) return; - } while ((i = ++p[--index]) >= n); + if (++index < r-1) { + p[index] = i = 0; + continue; + } + for (i = 0; i < n; ++i) { + p[index] = i; + if (!yield_indexed_values(values, r, p)) { + rb_raise(rb_eRuntimeError, "repeated permute reentered"); + } + } + do { + if (index <= 0) return; + } while ((i = ++p[--index]) >= n); } } @@ -5382,37 +7348,79 @@ rb_ary_repeated_permutation_size(VALUE ary, VALUE args, VALUE eobj) long k = NUM2LONG(RARRAY_AREF(args, 0)); if (k < 0) { - return LONG2FIX(0); + return LONG2FIX(0); } if (n <= 0) { - return LONG2FIX(!k); + return LONG2FIX(!k); } return rb_int_positive_pow(n, (unsigned long)k); } /* * call-seq: - * ary.repeated_permutation(n) { |p| block } -> ary - * ary.repeated_permutation(n) -> Enumerator + * array.repeated_permutation(n) {|permutation| ... } -> self + * array.repeated_permutation(n) -> new_enumerator * - * When invoked with a block, yield all repeated permutations of length +n+ of - * the elements of the array, then return the array itself. + * Calls the block with each repeated permutation of length +n+ of the elements of +self+; + * each permutation is an \Array; + * returns +self+. The order of the permutations is indeterminate. * - * The implementation makes no guarantees about the order in which the repeated - * permutations are yielded. + * When a block and a positive \Integer argument +n+ are given, calls the block with each + * +n+-tuple repeated permutation of the elements of +self+. + * The number of permutations is <tt>self.size**n</tt>. * - * If no block is given, an Enumerator is returned instead. + * +n+ = 1: * - * Examples: + * a = [0, 1, 2] + * a.repeated_permutation(1) {|permutation| p permutation } + * + * Output: + * + * [0] + * [1] + * [2] + * + * +n+ = 2: + * + * a.repeated_permutation(2) {|permutation| p permutation } + * + * Output: + * + * [0, 0] + * [0, 1] + * [0, 2] + * [1, 0] + * [1, 1] + * [1, 2] + * [2, 0] + * [2, 1] + * [2, 2] + * + * If +n+ is zero, calls the block once with an empty \Array. + * + * If +n+ is negative, does not call the block: + * + * a.repeated_permutation(-1) {|permutation| fail 'Cannot happen' } + * + * Returns a new \Enumerator if no block given: + * + * a = [0, 1, 2] + * a.repeated_permutation(2) # => #<Enumerator: [0, 1, 2]:permutation(2)> + * + * Using Enumerators, it's convenient to show the permutations and counts + * for some values of +n+: + * + * e = a.repeated_permutation(0) + * e.size # => 1 + * e.to_a # => [[]] + * e = a.repeated_permutation(1) + * e.size # => 3 + * e.to_a # => [[0], [1], [2]] + * e = a.repeated_permutation(2) + * e.size # => 9 + * e.to_a # => [[0, 0], [0, 1], [0, 2], [1, 0], [1, 1], [1, 2], [2, 0], [2, 1], [2, 2]] * - * a = [1, 2] - * a.repeated_permutation(1).to_a #=> [[1], [2]] - * a.repeated_permutation(2).to_a #=> [[1,1],[1,2],[2,1],[2,2]] - * a.repeated_permutation(3).to_a #=> [[1,1,1],[1,1,2],[1,2,1],[1,2,2], - * # [2,1,1],[2,1,2],[2,2,1],[2,2,2]] - * a.repeated_permutation(0).to_a #=> [[]] # one permutation of length 0 */ - static VALUE rb_ary_repeated_permutation(VALUE ary, VALUE num) { @@ -5423,25 +7431,25 @@ rb_ary_repeated_permutation(VALUE ary, VALUE num) r = NUM2LONG(num); /* Permutation size from argument */ if (r < 0) { - /* no permutations: yield nothing */ + /* no permutations: yield nothing */ } else if (r == 0) { /* exactly one permutation: the zero-length array */ - rb_yield(rb_ary_new2(0)); + rb_yield(rb_ary_new2(0)); } else if (r == 1) { /* this is a special, easy case */ - for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); + } } else { /* this is the general case */ - volatile VALUE t0; - long *p = ALLOCV_N(long, t0, r); - VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ - RBASIC_CLEAR_CLASS(ary0); + volatile VALUE t0; + long *p = ALLOCV_N(long, t0, r); + VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ + RBASIC_CLEAR_CLASS(ary0); - rpermute0(n, r, p, ary0); /* compute and yield repeated permutations */ - ALLOCV_END(t0); - RBASIC_SET_CLASS_RAW(ary0, rb_cArray); + rpermute0(n, r, p, ary0); /* compute and yield repeated permutations */ + ALLOCV_END(t0); + RBASIC_SET_CLASS_RAW(ary0, rb_cArray); } return ary; } @@ -5453,19 +7461,19 @@ rcombinate0(const long n, const long r, long *const p, const long rest, const VA p[index] = i; for (;;) { - if (++index < r-1) { - p[index] = i; - continue; - } - for (; i < n; ++i) { - p[index] = i; - if (!yield_indexed_values(values, r, p)) { - rb_raise(rb_eRuntimeError, "repeated combination reentered"); - } - } - do { - if (index <= 0) return; - } while ((i = ++p[--index]) >= n); + if (++index < r-1) { + p[index] = i; + continue; + } + for (; i < n; ++i) { + p[index] = i; + if (!yield_indexed_values(values, r, p)) { + rb_raise(rb_eRuntimeError, "repeated combination reentered"); + } + } + do { + if (index <= 0) return; + } while ((i = ++p[--index]) >= n); } } @@ -5475,35 +7483,71 @@ rb_ary_repeated_combination_size(VALUE ary, VALUE args, VALUE eobj) long n = RARRAY_LEN(ary); long k = NUM2LONG(RARRAY_AREF(args, 0)); if (k == 0) { - return LONG2FIX(1); + return LONG2FIX(1); } return binomial_coefficient(k, n + k - 1); } /* * call-seq: - * ary.repeated_combination(n) { |c| block } -> ary - * ary.repeated_combination(n) -> Enumerator + * array.repeated_combination(n) {|combination| ... } -> self + * array.repeated_combination(n) -> new_enumerator + * + * Calls the block with each repeated combination of length +n+ of the elements of +self+; + * each combination is an \Array; + * returns +self+. The order of the combinations is indeterminate. + * + * When a block and a positive \Integer argument +n+ are given, calls the block with each + * +n+-tuple repeated combination of the elements of +self+. + * The number of combinations is <tt>(n+1)(n+2)/2</tt>. + * + * +n+ = 1: + * + * a = [0, 1, 2] + * a.repeated_combination(1) {|combination| p combination } + * + * Output: + * + * [0] + * [1] + * [2] + * + * +n+ = 2: + * + * a.repeated_combination(2) {|combination| p combination } * - * When invoked with a block, yields all repeated combinations of length +n+ of - * elements from the array and then returns the array itself. + * Output: * - * The implementation makes no guarantees about the order in which the repeated - * combinations are yielded. + * [0, 0] + * [0, 1] + * [0, 2] + * [1, 1] + * [1, 2] + * [2, 2] * - * If no block is given, an Enumerator is returned instead. + * If +n+ is zero, calls the block once with an empty \Array. * - * Examples: + * If +n+ is negative, does not call the block: * - * a = [1, 2, 3] - * a.repeated_combination(1).to_a #=> [[1], [2], [3]] - * a.repeated_combination(2).to_a #=> [[1,1],[1,2],[1,3],[2,2],[2,3],[3,3]] - * a.repeated_combination(3).to_a #=> [[1,1,1],[1,1,2],[1,1,3],[1,2,2],[1,2,3], - * # [1,3,3],[2,2,2],[2,2,3],[2,3,3],[3,3,3]] - * a.repeated_combination(4).to_a #=> [[1,1,1,1],[1,1,1,2],[1,1,1,3],[1,1,2,2],[1,1,2,3], - * # [1,1,3,3],[1,2,2,2],[1,2,2,3],[1,2,3,3],[1,3,3,3], - * # [2,2,2,2],[2,2,2,3],[2,2,3,3],[2,3,3,3],[3,3,3,3]] - * a.repeated_combination(0).to_a #=> [[]] # one combination of length 0 + * a.repeated_combination(-1) {|combination| fail 'Cannot happen' } + * + * Returns a new \Enumerator if no block given: + * + * a = [0, 1, 2] + * a.repeated_combination(2) # => #<Enumerator: [0, 1, 2]:combination(2)> + * + * Using Enumerators, it's convenient to show the combinations and counts + * for some values of +n+: + * + * e = a.repeated_combination(0) + * e.size # => 1 + * e.to_a # => [[]] + * e = a.repeated_combination(1) + * e.size # => 3 + * e.to_a # => [[0], [1], [2]] + * e = a.repeated_combination(2) + * e.size # => 6 + * e.to_a # => [[0, 0], [0, 1], [0, 2], [1, 1], [1, 2], [2, 2]] * */ @@ -5516,67 +7560,105 @@ rb_ary_repeated_combination(VALUE ary, VALUE num) RETURN_SIZED_ENUMERATOR(ary, 1, &num, rb_ary_repeated_combination_size); /* Return enumerator if no block */ len = RARRAY_LEN(ary); if (n < 0) { - /* yield nothing */ + /* yield nothing */ } else if (n == 0) { - rb_yield(rb_ary_new2(0)); + rb_yield(rb_ary_new2(0)); } else if (n == 1) { - for (i = 0; i < RARRAY_LEN(ary); i++) { - rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); - } + for (i = 0; i < RARRAY_LEN(ary); i++) { + rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i))); + } } else if (len == 0) { - /* yield nothing */ + /* yield nothing */ } else { - volatile VALUE t0; - long *p = ALLOCV_N(long, t0, n); - VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ - RBASIC_CLEAR_CLASS(ary0); + volatile VALUE t0; + long *p = ALLOCV_N(long, t0, n); + VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */ + RBASIC_CLEAR_CLASS(ary0); - rcombinate0(len, n, p, n, ary0); /* compute and yield repeated combinations */ - ALLOCV_END(t0); - RBASIC_SET_CLASS_RAW(ary0, rb_cArray); + rcombinate0(len, n, p, n, ary0); /* compute and yield repeated combinations */ + ALLOCV_END(t0); + RBASIC_SET_CLASS_RAW(ary0, rb_cArray); } return ary; } /* * call-seq: - * ary.product(other_ary, ...) -> new_ary - * ary.product(other_ary, ...) { |p| block } -> ary + * array.product(*other_arrays) -> new_array + * array.product(*other_arrays) {|combination| ... } -> self + * + * Computes and returns or yields all combinations of elements from all the Arrays, + * including both +self+ and +other_arrays+: + * + * - The number of combinations is the product of the sizes of all the arrays, + * including both +self+ and +other_arrays+. + * - The order of the returned combinations is indeterminate. + * + * When no block is given, returns the combinations as an \Array of Arrays: + * + * a = [0, 1, 2] + * a1 = [3, 4] + * a2 = [5, 6] + * p = a.product(a1) + * p.size # => 6 # a.size * a1.size + * p # => [[0, 3], [0, 4], [1, 3], [1, 4], [2, 3], [2, 4]] + * p = a.product(a1, a2) + * p.size # => 12 # a.size * a1.size * a2.size + * p # => [[0, 3, 5], [0, 3, 6], [0, 4, 5], [0, 4, 6], [1, 3, 5], [1, 3, 6], [1, 4, 5], [1, 4, 6], [2, 3, 5], [2, 3, 6], [2, 4, 5], [2, 4, 6]] + * + * If any argument is an empty \Array, returns an empty \Array. + * + * If no argument is given, returns an \Array of 1-element Arrays, + * each containing an element of +self+: + * + * a.product # => [[0], [1], [2]] + * + * When a block is given, yields each combination as an \Array; returns +self+: * - * Returns an array of all combinations of elements from all arrays. + * a.product(a1) {|combination| p combination } * - * The length of the returned array is the product of the length of +self+ and - * the argument arrays. + * Output: * - * If given a block, #product will yield all combinations and return +self+ - * instead. + * [0, 3] + * [0, 4] + * [1, 3] + * [1, 4] + * [2, 3] + * [2, 4] + * + * If any argument is an empty \Array, does not call the block: + * + * a.product(a1, a2, []) {|combination| fail 'Cannot happen' } + * + * If no argument is given, yields each element of +self+ as a 1-element \Array: + * + * a.product {|combination| p combination } + * + * Output: + * + * [0] + * [1] + * [2] * - * [1,2,3].product([4,5]) #=> [[1,4],[1,5],[2,4],[2,5],[3,4],[3,5]] - * [1,2].product([1,2]) #=> [[1,1],[1,2],[2,1],[2,2]] - * [1,2].product([3,4],[5,6]) #=> [[1,3,5],[1,3,6],[1,4,5],[1,4,6], - * # [2,3,5],[2,3,6],[2,4,5],[2,4,6]] - * [1,2].product() #=> [[1],[2]] - * [1,2].product([]) #=> [] */ static VALUE rb_ary_product(int argc, VALUE *argv, VALUE ary) { int n = argc+1; /* How many arrays we're operating on */ - volatile VALUE t0 = tmpary(n); - volatile VALUE t1 = tmpbuf(n, sizeof(int)); + volatile VALUE t0 = rb_ary_hidden_new(n); + volatile VALUE t1 = Qundef; VALUE *arrays = RARRAY_PTR(t0); /* The arrays we're computing the product of */ - int *counters = (int*)RSTRING_PTR(t1); /* The current position in each one */ + int *counters = ALLOCV_N(int, t1, n); /* The current position in each one */ VALUE result = Qnil; /* The array we'll be returning, when no block given */ long i,j; long resultlen = 1; RBASIC_CLEAR_CLASS(t0); - RBASIC_CLEAR_CLASS(t1); /* initialize the arrays of arrays */ ARY_SET_LEN(t0, n); @@ -5589,81 +7671,84 @@ rb_ary_product(int argc, VALUE *argv, VALUE ary) /* Otherwise, allocate and fill in an array of results */ if (rb_block_given_p()) { - /* Make defensive copies of arrays; exit if any is empty */ - for (i = 0; i < n; i++) { - if (RARRAY_LEN(arrays[i]) == 0) goto done; - arrays[i] = ary_make_shared_copy(arrays[i]); - } + /* Make defensive copies of arrays; exit if any is empty */ + for (i = 0; i < n; i++) { + if (RARRAY_LEN(arrays[i]) == 0) goto done; + arrays[i] = ary_make_shared_copy(arrays[i]); + } } else { - /* Compute the length of the result array; return [] if any is empty */ - for (i = 0; i < n; i++) { - long k = RARRAY_LEN(arrays[i]); - if (k == 0) { - result = rb_ary_new2(0); - goto done; - } + /* Compute the length of the result array; return [] if any is empty */ + for (i = 0; i < n; i++) { + long k = RARRAY_LEN(arrays[i]); + if (k == 0) { + result = rb_ary_new2(0); + goto done; + } if (MUL_OVERFLOW_LONG_P(resultlen, k)) - rb_raise(rb_eRangeError, "too big to product"); - resultlen *= k; - } - result = rb_ary_new2(resultlen); + rb_raise(rb_eRangeError, "too big to product"); + resultlen *= k; + } + result = rb_ary_new2(resultlen); } for (;;) { - int m; - /* fill in one subarray */ - VALUE subarray = rb_ary_new2(n); - for (j = 0; j < n; j++) { - rb_ary_push(subarray, rb_ary_entry(arrays[j], counters[j])); - } - - /* put it on the result array */ - if (NIL_P(result)) { - FL_SET(t0, FL_USER5); - rb_yield(subarray); - if (! FL_TEST(t0, FL_USER5)) { - rb_raise(rb_eRuntimeError, "product reentered"); - } - else { - FL_UNSET(t0, FL_USER5); - } - } - else { - rb_ary_push(result, subarray); - } - - /* - * Increment the last counter. If it overflows, reset to 0 - * and increment the one before it. - */ - m = n-1; - counters[m]++; - while (counters[m] == RARRAY_LEN(arrays[m])) { - counters[m] = 0; - /* If the first counter overflows, we are done */ - if (--m < 0) goto done; - counters[m]++; - } + int m; + /* fill in one subarray */ + VALUE subarray = rb_ary_new2(n); + for (j = 0; j < n; j++) { + rb_ary_push(subarray, rb_ary_entry(arrays[j], counters[j])); + } + + /* put it on the result array */ + if (NIL_P(result)) { + FL_SET(t0, RARRAY_SHARED_ROOT_FLAG); + rb_yield(subarray); + if (!FL_TEST(t0, RARRAY_SHARED_ROOT_FLAG)) { + rb_raise(rb_eRuntimeError, "product reentered"); + } + else { + FL_UNSET(t0, RARRAY_SHARED_ROOT_FLAG); + } + } + else { + rb_ary_push(result, subarray); + } + + /* + * Increment the last counter. If it overflows, reset to 0 + * and increment the one before it. + */ + m = n-1; + counters[m]++; + while (counters[m] == RARRAY_LEN(arrays[m])) { + counters[m] = 0; + /* If the first counter overflows, we are done */ + if (--m < 0) goto done; + counters[m]++; + } } + done: - tmpary_discard(t0); - tmpbuf_discard(t1); + ALLOCV_END(t1); return NIL_P(result) ? ary : result; } /* * call-seq: - * ary.take(n) -> new_ary - * - * Returns first +n+ elements from the array. + * array.take(n) -> new_array * - * If a negative number is given, raises an ArgumentError. + * Returns a new \Array containing the first +n+ element of +self+, + * where +n+ is a non-negative \Integer; + * does not modify +self+. * - * See also Array#drop + * Examples: * - * a = [1, 2, 3, 4, 5, 0] - * a.take(3) #=> [1, 2, 3] + * a = [0, 1, 2, 3, 4, 5] + * a.take(1) # => [0] + * a.take(2) # => [0, 1] + * a.take(50) # => [0, 1, 2, 3, 4, 5] + * a # => [0, 1, 2, 3, 4, 5] * */ @@ -5672,25 +7757,31 @@ rb_ary_take(VALUE obj, VALUE n) { long len = NUM2LONG(n); if (len < 0) { - rb_raise(rb_eArgError, "attempt to take negative size"); + rb_raise(rb_eArgError, "attempt to take negative size"); } return rb_ary_subseq(obj, 0, len); } /* * call-seq: - * ary.take_while { |obj| block } -> new_ary - * ary.take_while -> Enumerator + * array.take_while {|element| ... } -> new_array + * array.take_while -> new_enumerator * - * Passes elements to the block until the block returns +nil+ or +false+, then - * stops iterating and returns an array of all prior elements. + * Returns a new \Array containing zero or more leading elements of +self+; + * does not modify +self+. * - * If no block is given, an Enumerator is returned instead. + * With a block given, calls the block with each successive element of +self+; + * stops if the block returns +false+ or +nil+; + * returns a new \Array containing those elements for which the block returned a truthy value: * - * See also Array#drop_while + * a = [0, 1, 2, 3, 4, 5] + * a.take_while {|element| element < 3 } # => [0, 1, 2] + * a.take_while {|element| true } # => [0, 1, 2, 3, 4, 5] + * a # => [0, 1, 2, 3, 4, 5] * - * a = [1, 2, 3, 4, 5, 0] - * a.take_while { |i| i < 3 } #=> [1, 2] + * With no block given, returns a new \Enumerator: + * + * [0, 1].take_while # => #<Enumerator: [0, 1]:take_while> * */ @@ -5701,24 +7792,25 @@ rb_ary_take_while(VALUE ary) RETURN_ENUMERATOR(ary, 0, 0); for (i = 0; i < RARRAY_LEN(ary); i++) { - if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break; + if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break; } return rb_ary_take(ary, LONG2FIX(i)); } /* * call-seq: - * ary.drop(n) -> new_ary - * - * Drops first +n+ elements from +ary+ and returns the rest of the elements in - * an array. + * array.drop(n) -> new_array * - * If a negative number is given, raises an ArgumentError. + * Returns a new \Array containing all but the first +n+ element of +self+, + * where +n+ is a non-negative \Integer; + * does not modify +self+. * - * See also Array#take + * Examples: * - * a = [1, 2, 3, 4, 5, 0] - * a.drop(3) #=> [4, 5, 0] + * a = [0, 1, 2, 3, 4, 5] + * a.drop(0) # => [0, 1, 2, 3, 4, 5] + * a.drop(1) # => [1, 2, 3, 4, 5] + * a.drop(2) # => [2, 3, 4, 5] * */ @@ -5728,29 +7820,32 @@ rb_ary_drop(VALUE ary, VALUE n) VALUE result; long pos = NUM2LONG(n); if (pos < 0) { - rb_raise(rb_eArgError, "attempt to drop negative size"); + rb_raise(rb_eArgError, "attempt to drop negative size"); } result = rb_ary_subseq(ary, pos, RARRAY_LEN(ary)); - if (result == Qnil) result = rb_ary_new(); + if (NIL_P(result)) result = rb_ary_new(); return result; } /* * call-seq: - * ary.drop_while { |obj| block } -> new_ary - * ary.drop_while -> Enumerator + * array.drop_while {|element| ... } -> new_array + * array.drop_while -> new_enumerator + + * Returns a new \Array containing zero or more trailing elements of +self+; + * does not modify +self+. * - * Drops elements up to, but not including, the first element for which the - * block returns +nil+ or +false+ and returns an array containing the - * remaining elements. + * With a block given, calls the block with each successive element of +self+; + * stops if the block returns +false+ or +nil+; + * returns a new \Array _omitting_ those elements for which the block returned a truthy value: * - * If no block is given, an Enumerator is returned instead. + * a = [0, 1, 2, 3, 4, 5] + * a.drop_while {|element| element < 3 } # => [3, 4, 5] * - * See also Array#take_while + * With no block given, returns a new \Enumerator: * - * a = [1, 2, 3, 4, 5, 0] - * a.drop_while {|i| i < 3 } #=> [3, 4, 5, 0] + * [0, 1].drop_while # => # => #<Enumerator: [0, 1]:drop_while> * */ @@ -5761,59 +7856,286 @@ rb_ary_drop_while(VALUE ary) RETURN_ENUMERATOR(ary, 0, 0); for (i = 0; i < RARRAY_LEN(ary); i++) { - if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break; + if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break; } return rb_ary_drop(ary, LONG2FIX(i)); } /* * call-seq: - * ary.any? [{ |obj| block }] -> true or false + * array.any? -> true or false + * array.any? {|element| ... } -> true or false + * array.any?(obj) -> true or false * - * See also Enumerable#any? + * Returns +true+ if any element of +self+ meets a given criterion. + * + * With no block given and no argument, returns +true+ if +self+ has any truthy element, + * +false+ otherwise: + * + * [nil, 0, false].any? # => true + * [nil, false].any? # => false + * [].any? # => false + * + * With a block given and no argument, calls the block with each element in +self+; + * returns +true+ if the block returns any truthy value, +false+ otherwise: + * + * [0, 1, 2].any? {|element| element > 1 } # => true + * [0, 1, 2].any? {|element| element > 2 } # => false + * + * If argument +obj+ is given, returns +true+ if +obj+.<tt>===</tt> any element, + * +false+ otherwise: + * + * ['food', 'drink'].any?(/foo/) # => true + * ['food', 'drink'].any?(/bar/) # => false + * [].any?(/foo/) # => false + * [0, 1, 2].any?(1) # => true + * [0, 1, 2].any?(3) # => false + * + * Related: Enumerable#any? */ static VALUE rb_ary_any_p(int argc, VALUE *argv, VALUE ary) { long i, len = RARRAY_LEN(ary); - const VALUE *ptr = RARRAY_CONST_PTR(ary); rb_check_arity(argc, 0, 1); if (!len) return Qfalse; if (argc) { - for (i = 0; i < RARRAY_LEN(ary); ++i) { - if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qtrue; - } + if (rb_block_given_p()) { + rb_warn("given block not used"); + } + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qtrue; + } } else if (!rb_block_given_p()) { - for (i = 0; i < len; ++i) if (RTEST(ptr[i])) return Qtrue; + for (i = 0; i < len; ++i) { + if (RTEST(RARRAY_AREF(ary, i))) return Qtrue; + } } else { - for (i = 0; i < RARRAY_LEN(ary); ++i) { - if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qtrue; - } + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qtrue; + } } return Qfalse; } /* - * call-seq: - * ary.dig(idx, ...) -> object + * call-seq: + * array.all? -> true or false + * array.all? {|element| ... } -> true or false + * array.all?(obj) -> true or false * - * Extracts the nested value specified by the sequence of <i>idx</i> - * objects by calling +dig+ at each step, returning +nil+ if any - * intermediate step is +nil+. + * Returns +true+ if all elements of +self+ meet a given criterion. * - * a = [[1, [2, 3]]] + * With no block given and no argument, returns +true+ if +self+ contains only truthy elements, + * +false+ otherwise: * - * a.dig(0, 1, 1) #=> 3 - * a.dig(1, 2, 3) #=> nil - * a.dig(0, 0, 0) #=> TypeError: Integer does not have #dig method - * [42, {foo: :bar}].dig(1, :foo) #=> :bar + * [0, 1, :foo].all? # => true + * [0, nil, 2].all? # => false + * [].all? # => true + * + * With a block given and no argument, calls the block with each element in +self+; + * returns +true+ if the block returns only truthy values, +false+ otherwise: + * + * [0, 1, 2].all? { |element| element < 3 } # => true + * [0, 1, 2].all? { |element| element < 2 } # => false + * + * If argument +obj+ is given, returns +true+ if <tt>obj.===</tt> every element, +false+ otherwise: + * + * ['food', 'fool', 'foot'].all?(/foo/) # => true + * ['food', 'drink'].all?(/bar/) # => false + * [].all?(/foo/) # => true + * [0, 0, 0].all?(0) # => true + * [0, 1, 2].all?(1) # => false + * + * Related: Enumerable#all? */ -VALUE +static VALUE +rb_ary_all_p(int argc, VALUE *argv, VALUE ary) +{ + long i, len = RARRAY_LEN(ary); + + rb_check_arity(argc, 0, 1); + if (!len) return Qtrue; + if (argc) { + if (rb_block_given_p()) { + rb_warn("given block not used"); + } + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (!RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qfalse; + } + } + else if (!rb_block_given_p()) { + for (i = 0; i < len; ++i) { + if (!RTEST(RARRAY_AREF(ary, i))) return Qfalse; + } + } + else { + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qfalse; + } + } + return Qtrue; +} + +/* + * call-seq: + * array.none? -> true or false + * array.none? {|element| ... } -> true or false + * array.none?(obj) -> true or false + * + * Returns +true+ if no element of +self+ meet a given criterion. + * + * With no block given and no argument, returns +true+ if +self+ has no truthy elements, + * +false+ otherwise: + * + * [nil, false].none? # => true + * [nil, 0, false].none? # => false + * [].none? # => true + * + * With a block given and no argument, calls the block with each element in +self+; + * returns +true+ if the block returns no truthy value, +false+ otherwise: + * + * [0, 1, 2].none? {|element| element > 3 } # => true + * [0, 1, 2].none? {|element| element > 1 } # => false + * + * If argument +obj+ is given, returns +true+ if <tt>obj.===</tt> no element, +false+ otherwise: + * + * ['food', 'drink'].none?(/bar/) # => true + * ['food', 'drink'].none?(/foo/) # => false + * [].none?(/foo/) # => true + * [0, 1, 2].none?(3) # => true + * [0, 1, 2].none?(1) # => false + * + * Related: Enumerable#none? + */ + +static VALUE +rb_ary_none_p(int argc, VALUE *argv, VALUE ary) +{ + long i, len = RARRAY_LEN(ary); + + rb_check_arity(argc, 0, 1); + if (!len) return Qtrue; + if (argc) { + if (rb_block_given_p()) { + rb_warn("given block not used"); + } + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qfalse; + } + } + else if (!rb_block_given_p()) { + for (i = 0; i < len; ++i) { + if (RTEST(RARRAY_AREF(ary, i))) return Qfalse; + } + } + else { + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qfalse; + } + } + return Qtrue; +} + +/* + * call-seq: + * array.one? -> true or false + * array.one? {|element| ... } -> true or false + * array.one?(obj) -> true or false + * + * Returns +true+ if exactly one element of +self+ meets a given criterion. + * + * With no block given and no argument, returns +true+ if +self+ has exactly one truthy element, + * +false+ otherwise: + * + * [nil, 0].one? # => true + * [0, 0].one? # => false + * [nil, nil].one? # => false + * [].one? # => false + * + * With a block given and no argument, calls the block with each element in +self+; + * returns +true+ if the block a truthy value for exactly one element, +false+ otherwise: + * + * [0, 1, 2].one? {|element| element > 0 } # => false + * [0, 1, 2].one? {|element| element > 1 } # => true + * [0, 1, 2].one? {|element| element > 2 } # => false + * + * If argument +obj+ is given, returns +true+ if <tt>obj.===</tt> exactly one element, + * +false+ otherwise: + * + * [0, 1, 2].one?(0) # => true + * [0, 0, 1].one?(0) # => false + * [1, 1, 2].one?(0) # => false + * ['food', 'drink'].one?(/bar/) # => false + * ['food', 'drink'].one?(/foo/) # => true + * [].one?(/foo/) # => false + * + * Related: Enumerable#one? + */ + +static VALUE +rb_ary_one_p(int argc, VALUE *argv, VALUE ary) +{ + long i, len = RARRAY_LEN(ary); + VALUE result = Qfalse; + + rb_check_arity(argc, 0, 1); + if (!len) return Qfalse; + if (argc) { + if (rb_block_given_p()) { + rb_warn("given block not used"); + } + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) { + if (result) return Qfalse; + result = Qtrue; + } + } + } + else if (!rb_block_given_p()) { + for (i = 0; i < len; ++i) { + if (RTEST(RARRAY_AREF(ary, i))) { + if (result) return Qfalse; + result = Qtrue; + } + } + } + else { + for (i = 0; i < RARRAY_LEN(ary); ++i) { + if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) { + if (result) return Qfalse; + result = Qtrue; + } + } + } + return result; +} + +/* + * call-seq: + * array.dig(index, *identifiers) -> object + * + * Finds and returns the object in nested objects + * that is specified by +index+ and +identifiers+. + * The nested objects may be instances of various classes. + * See {Dig Methods}[rdoc-ref:dig_methods.rdoc]. + * + * Examples: + * + * a = [:foo, [:bar, :baz, [:bat, :bam]]] + * a.dig(1) # => [:bar, :baz, [:bat, :bam]] + * a.dig(1, 2) # => [:bat, :bam] + * a.dig(1, 2, 0) # => :bat + * a.dig(1, 2, 3) # => nil + * + */ + +static VALUE rb_ary_dig(int argc, VALUE *argv, VALUE self) { rb_check_arity(argc, 1, UNLIMITED_ARGUMENTS); @@ -5828,14 +8150,8 @@ finish_exact_sum(long n, VALUE r, VALUE v, int z) { if (n != 0) v = rb_fix_plus(LONG2FIX(n), v); - if (r != Qundef) { - /* r can be an Integer when mathn is loaded */ - if (FIXNUM_P(r)) - v = rb_fix_plus(r, v); - else if (RB_TYPE_P(r, T_BIGNUM)) - v = rb_big_plus(r, v); - else - v = rb_rational_plus(r, v); + if (!UNDEF_P(r)) { + v = rb_rational_plus(r, v); } else if (!n && z) { v = rb_fix_plus(LONG2FIX(0), v); @@ -5845,43 +8161,41 @@ finish_exact_sum(long n, VALUE r, VALUE v, int z) /* * call-seq: - * ary.sum(init=0) -> number - * ary.sum(init=0) {|e| expr } -> number - * - * Returns the sum of elements. - * For example, [e1, e2, e3].sum returns init + e1 + e2 + e3. + * array.sum(init = 0) -> object + * array.sum(init = 0) {|element| ... } -> object * - * If a block is given, the block is applied to each element - * before addition. + * When no block is given, returns the object equivalent to: * - * If <i>ary</i> is empty, it returns <i>init</i>. + * sum = init + * array.each {|element| sum += element } + * sum * - * [].sum #=> 0 - * [].sum(0.0) #=> 0.0 - * [1, 2, 3].sum #=> 6 - * [3, 5.5].sum #=> 8.5 - * [2.5, 3.0].sum(0.0) {|e| e * e } #=> 15.25 - * [Object.new].sum #=> TypeError + * For example, <tt>[e1, e2, e3].sum</tt> returns <tt>init + e1 + e2 + e3</tt>. * - * The (arithmetic) mean value of an array can be obtained as follows. + * Examples: * - * mean = ary.sum(0.0) / ary.length + * a = [0, 1, 2, 3] + * a.sum # => 6 + * a.sum(100) # => 106 * - * This method can be used for non-numeric objects by - * explicit <i>init</i> argument. + * The elements need not be numeric, but must be <tt>+</tt>-compatible + * with each other and with +init+: * - * ["a", "b", "c"].sum("") #=> "abc" - * [[1], [[2]], [3]].sum([]) #=> [1, [2], 3] + * a = ['abc', 'def', 'ghi'] + * a.sum('jkl') # => "jklabcdefghi" * - * However, Array#join and Array#flatten is faster than Array#sum for - * array of strings and array of arrays. + * When a block is given, it is called with each element + * and the block's return value (instead of the element itself) is used as the addend: * - * ["a", "b", "c"].join #=> "abc" - * [[1], [[2]], [3]].flatten(1) #=> [1, [2], 3] + * a = ['zero', 1, :two] + * s = a.sum('Coerced and concatenated: ') {|element| element.to_s } + * s # => "Coerced and concatenated: zero1two" * + * Notes: * - * Array#sum method may not respect method redefinition of "+" methods - * such as Integer#+. + * - Array#join and Array#flatten may be faster than Array#sum + * for an \Array of Strings or an \Array of Arrays. + * - Array#sum method may not respect method redefinition of "+" methods such as Integer#+. * */ @@ -5892,8 +8206,7 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) long i, n; int block_given; - if (rb_scan_args(argc, argv, "01", &v) == 0) - v = LONG2FIX(0); + v = (rb_check_arity(argc, 0, 1) ? argv[0] : LONG2FIX(0)); block_given = rb_block_given_p(); @@ -5913,10 +8226,10 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) n = 0; } } - else if (RB_TYPE_P(e, T_BIGNUM)) + else if (RB_BIGNUM_TYPE_P(e)) v = rb_big_plus(e, v); else if (RB_TYPE_P(e, T_RATIONAL)) { - if (r == Qundef) + if (UNDEF_P(r)) r = e; else r = rb_rational_plus(r, e); @@ -5933,15 +8246,15 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) if (RB_FLOAT_TYPE_P(e)) { /* * Kahan-Babuska balancing compensated summation algorithm - * See http://link.springer.com/article/10.1007/s00607-005-0139-x + * See https://link.springer.com/article/10.1007/s00607-005-0139-x */ double f, c; + double x, t; f = NUM2DBL(v); c = 0.0; goto has_float_value; for (; i < RARRAY_LEN(ary); i++) { - double x, t; e = RARRAY_AREF(ary, i); if (block_given) e = rb_yield(e); @@ -5950,13 +8263,27 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) x = RFLOAT_VALUE(e); else if (FIXNUM_P(e)) x = FIX2LONG(e); - else if (RB_TYPE_P(e, T_BIGNUM)) + else if (RB_BIGNUM_TYPE_P(e)) x = rb_big2dbl(e); else if (RB_TYPE_P(e, T_RATIONAL)) x = rb_num2dbl(e); else goto not_float; + if (isnan(f)) continue; + if (isnan(x)) { + f = x; + continue; + } + if (isinf(x)) { + if (isinf(f) && signbit(x) != signbit(f)) + f = NAN; + else + f = x; + continue; + } + if (isinf(f)) continue; + t = f + x; if (fabs(f) >= fabs(x)) c += ((f - t) + x); @@ -5982,56 +8309,135 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) return v; } +static VALUE +rb_ary_deconstruct(VALUE ary) +{ + return ary; +} + /* - * Arrays are ordered, integer-indexed collections of any object. + * An \Array is an ordered, integer-indexed collection of objects, called _elements_. + * Any object (even another array) may be an array element, + * and an array can contain objects of different types. + * + * == \Array Indexes + * + * \Array indexing starts at 0, as in C or Java. + * + * A positive index is an offset from the first element: + * + * - Index 0 indicates the first element. + * - Index 1 indicates the second element. + * - ... + * + * A negative index is an offset, backwards, from the end of the array: + * + * - Index -1 indicates the last element. + * - Index -2 indicates the next-to-last element. + * - ... + * + * A non-negative index is <i>in range</i> if and only if it is smaller than + * the size of the array. For a 3-element array: + * + * - Indexes 0 through 2 are in range. + * - Index 3 is out of range. + * + * A negative index is <i>in range</i> if and only if its absolute value is + * not larger than the size of the array. For a 3-element array: + * + * - Indexes -1 through -3 are in range. + * - Index -4 is out of range. + * + * Although the effective index into an array is always an integer, + * some methods (both within and outside of class \Array) + * accept one or more non-integer arguments that are + * {integer-convertible objects}[rdoc-ref:implicit_conversion.rdoc@Integer-Convertible+Objects]. * - * Array indexing starts at 0, as in C or Java. A negative index is assumed - * to be relative to the end of the array---that is, an index of -1 indicates - * the last element of the array, -2 is the next to last element in the - * array, and so on. * * == Creating Arrays * - * A new array can be created by using the literal constructor - * <code>[]</code>. Arrays can contain different types of objects. For - * example, the array below contains an Integer, a String and a Float: + * You can create an \Array object explicitly with: + * + * - An {array literal}[rdoc-ref:literals.rdoc@Array+Literals]: + * + * [1, 'one', :one, [2, 'two', :two]] + * + * - A {%w or %W: string-array Literal}[rdoc-ref:literals.rdoc@25w+and+-25W-3A+String-Array+Literals]: + * + * %w[foo bar baz] # => ["foo", "bar", "baz"] + * %w[1 % *] # => ["1", "%", "*"] * - * ary = [1, "two", 3.0] #=> [1, "two", 3.0] + * - A {%i pr %I: symbol-array Literal}[rdoc-ref:literals.rdoc@25i+and+-25I-3A+Symbol-Array+Literals]: * - * An array can also be created by explicitly calling Array.new with zero, one - * (the initial size of the Array) or two arguments (the initial size and a - * default object). + * %i[foo bar baz] # => [:foo, :bar, :baz] + * %i[1 % *] # => [:"1", :%, :*] * - * ary = Array.new #=> [] - * Array.new(3) #=> [nil, nil, nil] - * Array.new(3, true) #=> [true, true, true] + * - \Method Kernel#Array: * - * Note that the second argument populates the array with references to the - * same object. Therefore, it is only recommended in cases when you need to - * instantiate arrays with natively immutable objects such as Symbols, - * numbers, true or false. + * Array(["a", "b"]) # => ["a", "b"] + * Array(1..5) # => [1, 2, 3, 4, 5] + * Array(key: :value) # => [[:key, :value]] + * Array(nil) # => [] + * Array(1) # => [1] + * Array({:a => "a", :b => "b"}) # => [[:a, "a"], [:b, "b"]] * - * To create an array with separate objects a block can be passed instead. - * This method is safe to use with mutable objects such as hashes, strings or - * other arrays: + * - \Method Array.new: * - * Array.new(4) { Hash.new } #=> [{}, {}, {}, {}] - * Array.new(4) {|i| i.to_s } #=> ["0", "1", "2", "3"] + * Array.new # => [] + * Array.new(3) # => [nil, nil, nil] + * Array.new(4) {Hash.new} # => [{}, {}, {}, {}] + * Array.new(3, true) # => [true, true, true] * - * This is also a quick way to build up multi-dimensional arrays: + * Note that the last example above populates the array + * with references to the same object. + * This is recommended only in cases where that object is a natively immutable object + * such as a symbol, a numeric, +nil+, +true+, or +false+. * - * empty_table = Array.new(3) { Array.new(3) } - * #=> [[nil, nil, nil], [nil, nil, nil], [nil, nil, nil]] + * Another way to create an array with various objects, using a block; + * this usage is safe for mutable objects such as hashes, strings or + * other arrays: * - * An array can also be created by using the Array() method, provided by - * Kernel, which tries to call #to_ary, then #to_a on its argument. + * Array.new(4) {|i| i.to_s } # => ["0", "1", "2", "3"] * - * Array({:a => "a", :b => "b"}) #=> [[:a, "a"], [:b, "b"]] + * Here is a way to create a multi-dimensional array: + * + * Array.new(3) {Array.new(3)} + * # => [[nil, nil, nil], [nil, nil, nil], [nil, nil, nil]] + * + * A number of Ruby methods, both in the core and in the standard library, + * provide instance method +to_a+, which converts an object to an array. + * + * - ARGF#to_a + * - Array#to_a + * - Enumerable#to_a + * - Hash#to_a + * - MatchData#to_a + * - NilClass#to_a + * - OptionParser#to_a + * - Range#to_a + * - Set#to_a + * - Struct#to_a + * - Time#to_a + * - Benchmark::Tms#to_a + * - CSV::Table#to_a + * - Enumerator::Lazy#to_a + * - Gem::List#to_a + * - Gem::NameTuple#to_a + * - Gem::Platform#to_a + * - Gem::RequestSet::Lockfile::Tokenizer#to_a + * - Gem::SourceList#to_a + * - OpenSSL::X509::Extension#to_a + * - OpenSSL::X509::Name#to_a + * - Racc::ISet#to_a + * - Rinda::RingFinger#to_a + * - Ripper::Lexer::Elem#to_a + * - RubyVM::InstructionSequence#to_a + * - YAML::DBM#to_a * * == Example Usage * * In addition to the methods it mixes in through the Enumerable module, the - * Array class has proprietary methods for accessing, searching and otherwise + * \Array class has proprietary methods for accessing, searching and otherwise * manipulating arrays. * * Some of the more common ones are illustrated below. @@ -6079,7 +8485,7 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * * arr.drop(3) #=> [4, 5, 6] * - * == Obtaining Information about an Array + * == Obtaining Information about an \Array * * Arrays keep track of their own length at all times. To query an array * about the number of elements it contains, use #length, #count or #size. @@ -6117,7 +8523,7 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * arr.insert(3, 'orange', 'pear', 'grapefruit') * #=> [0, 1, 2, "orange", "pear", "grapefruit", "apple", 3, 4, 5, 6] * - * == Removing Items from an Array + * == Removing Items from an \Array * * The method #pop removes the last element in an array and returns it: * @@ -6159,15 +8565,15 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * * == Iterating over Arrays * - * Like all classes that include the Enumerable module, Array has an each + * Like all classes that include the Enumerable module, \Array has an each * method, which defines what elements should be iterated over and how. In - * case of Array's #each, all elements in the Array instance are yielded to + * case of Array's #each, all elements in the \Array instance are yielded to * the supplied block in sequence. * * Note that this operation leaves the array unchanged. * * arr = [1, 2, 3, 4, 5] - * arr.each { |a| print a -= 10, " " } + * arr.each {|a| print a -= 10, " "} * # prints: -9 -8 -7 -6 -5 * #=> [1, 2, 3, 4, 5] * @@ -6176,18 +8582,19 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * * words = %w[first second third fourth fifth sixth] * str = "" - * words.reverse_each { |word| str += "#{word} " } + * words.reverse_each {|word| str += "#{word} "} * p str #=> "sixth fifth fourth third second first " * * The #map method can be used to create a new array based on the original * array, but with the values modified by the supplied block: * - * arr.map { |a| 2*a } #=> [2, 4, 6, 8, 10] + * arr.map {|a| 2*a} #=> [2, 4, 6, 8, 10] * arr #=> [1, 2, 3, 4, 5] - * arr.map! { |a| a**2 } #=> [1, 4, 9, 16, 25] + * arr.map! {|a| a**2} #=> [1, 4, 9, 16, 25] * arr #=> [1, 4, 9, 16, 25] * - * == Selecting Items from an Array + * + * == Selecting Items from an \Array * * Elements can be selected from an array according to criteria defined in a * block. The selection can happen in a destructive or a non-destructive @@ -6198,9 +8605,9 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * === Non-destructive Selection * * arr = [1, 2, 3, 4, 5, 6] - * arr.select { |a| a > 3 } #=> [4, 5, 6] - * arr.reject { |a| a < 3 } #=> [3, 4, 5, 6] - * arr.drop_while { |a| a < 4 } #=> [4, 5, 6] + * arr.select {|a| a > 3} #=> [4, 5, 6] + * arr.reject {|a| a < 3} #=> [3, 4, 5, 6] + * arr.drop_while {|a| a < 4} #=> [4, 5, 6] * arr #=> [1, 2, 3, 4, 5, 6] * * === Destructive Selection @@ -6211,25 +8618,205 @@ rb_ary_sum(int argc, VALUE *argv, VALUE ary) * Similar to #select vs. #reject, #delete_if and #keep_if have the exact * opposite result when supplied with the same block: * - * arr.delete_if { |a| a < 4 } #=> [4, 5, 6] + * arr.delete_if {|a| a < 4} #=> [4, 5, 6] * arr #=> [4, 5, 6] * * arr = [1, 2, 3, 4, 5, 6] - * arr.keep_if { |a| a < 4 } #=> [1, 2, 3] + * arr.keep_if {|a| a < 4} #=> [1, 2, 3] * arr #=> [1, 2, 3] * + * == What's Here + * + * First, what's elsewhere. \Class \Array: + * + * - Inherits from {class Object}[rdoc-ref:Object@What-27s+Here]. + * - Includes {module Enumerable}[rdoc-ref:Enumerable@What-27s+Here], + * which provides dozens of additional methods. + * + * Here, class \Array provides methods that are useful for: + * + * - {Creating an Array}[rdoc-ref:Array@Methods+for+Creating+an+Array] + * - {Querying}[rdoc-ref:Array@Methods+for+Querying] + * - {Comparing}[rdoc-ref:Array@Methods+for+Comparing] + * - {Fetching}[rdoc-ref:Array@Methods+for+Fetching] + * - {Assigning}[rdoc-ref:Array@Methods+for+Assigning] + * - {Deleting}[rdoc-ref:Array@Methods+for+Deleting] + * - {Combining}[rdoc-ref:Array@Methods+for+Combining] + * - {Iterating}[rdoc-ref:Array@Methods+for+Iterating] + * - {Converting}[rdoc-ref:Array@Methods+for+Converting] + * - {And more....}[rdoc-ref:Array@Other+Methods] + * + * === Methods for Creating an \Array + * + * - ::[]: Returns a new array populated with given objects. + * - ::new: Returns a new array. + * - ::try_convert: Returns a new array created from a given object. + * + * === Methods for Querying + * + * - #length, #size: Returns the count of elements. + * - #include?: Returns whether any element <tt>==</tt> a given object. + * - #empty?: Returns whether there are no elements. + * - #all?: Returns whether all elements meet a given criterion. + * - #any?: Returns whether any element meets a given criterion. + * - #none?: Returns whether no element <tt>==</tt> a given object. + * - #one?: Returns whether exactly one element <tt>==</tt> a given object. + * - #count: Returns the count of elements that meet a given criterion. + * - #find_index, #index: Returns the index of the first element that meets a given criterion. + * - #rindex: Returns the index of the last element that meets a given criterion. + * - #hash: Returns the integer hash code. + * + * === Methods for Comparing + * + * - #<=>: Returns -1, 0, or 1 * as +self+ is less than, equal to, or + * greater than a given object. + * - #==: Returns whether each element in +self+ is <tt>==</tt> to the corresponding element + * in a given object. + * - #eql?: Returns whether each element in +self+ is <tt>eql?</tt> to the corresponding + * element in a given object. + + * === Methods for Fetching + * + * These methods do not modify +self+. + * + * - #[]: Returns one or more elements. + * - #fetch: Returns the element at a given offset. + * - #first: Returns one or more leading elements. + * - #last: Returns one or more trailing elements. + * - #max: Returns one or more maximum-valued elements, + * as determined by <tt><=></tt> or a given block. + * - #min: Returns one or more minimum-valued elements, + * as determined by <tt><=></tt> or a given block. + * - #minmax: Returns the minimum-valued and maximum-valued elements, + * as determined by <tt><=></tt> or a given block. + * - #assoc: Returns the first element that is an array + * whose first element <tt>==</tt> a given object. + * - #rassoc: Returns the first element that is an array + * whose second element <tt>==</tt> a given object. + * - #at: Returns the element at a given offset. + * - #values_at: Returns the elements at given offsets. + * - #dig: Returns the object in nested objects + * that is specified by a given index and additional arguments. + * - #drop: Returns trailing elements as determined by a given index. + * - #take: Returns leading elements as determined by a given index. + * - #drop_while: Returns trailing elements as determined by a given block. + * - #take_while: Returns leading elements as determined by a given block. + * - #slice: Returns consecutive elements as determined by a given argument. + * - #sort: Returns all elements in an order determined by <tt><=></tt> or a given block. + * - #reverse: Returns all elements in reverse order. + * - #compact: Returns an array containing all non-+nil+ elements. + * - #select, #filter: Returns an array containing elements selected by a given block. + * - #uniq: Returns an array containing non-duplicate elements. + * - #rotate: Returns all elements with some rotated from one end to the other. + * - #bsearch: Returns an element selected via a binary search + * as determined by a given block. + * - #bsearch_index: Returns the index of an element selected via a binary search + * as determined by a given block. + * - #sample: Returns one or more random elements. + * - #shuffle: Returns elements in a random order. + * + * === Methods for Assigning + * + * These methods add, replace, or reorder elements in +self+. + * + * - #[]=: Assigns specified elements with a given object. + * - #push, #append, #<<: Appends trailing elements. + * - #unshift, #prepend: Prepends leading elements. + * - #insert: Inserts given objects at a given offset; does not replace elements. + * - #concat: Appends all elements from given arrays. + * - #fill: Replaces specified elements with specified objects. + * - #replace: Replaces the content of +self+ with the content of a given array. + * - #reverse!: Replaces +self+ with its elements reversed. + * - #rotate!: Replaces +self+ with its elements rotated. + * - #shuffle!: Replaces +self+ with its elements in random order. + * - #sort!: Replaces +self+ with its elements sorted, + * as determined by <tt><=></tt> or a given block. + * - #sort_by!: Replaces +self+ with its elements sorted, as determined by a given block. + * + * === Methods for Deleting + * + * Each of these methods removes elements from +self+: + * + * - #pop: Removes and returns the last element. + * - #shift: Removes and returns the first element. + * - #compact!: Removes all +nil+ elements. + * - #delete: Removes elements equal to a given object. + * - #delete_at: Removes the element at a given offset. + * - #delete_if: Removes elements specified by a given block. + * - #keep_if: Removes elements not specified by a given block. + * - #reject!: Removes elements specified by a given block. + * - #select!, #filter!: Removes elements not specified by a given block. + * - #slice!: Removes and returns a sequence of elements. + * - #uniq!: Removes duplicates. + * + * === Methods for Combining + * + * - #&: Returns an array containing elements found both in +self+ and a given array. + * - #intersection: Returns an array containing elements found both in +self+ + * and in each given array. + * - #+: Returns an array containing all elements of +self+ followed by all elements of a given array. + * - #-: Returns an array containing all elements of +self+ that are not found in a given array. + * - #|: Returns an array containing all elements of +self+ and all elements of a given array, + * duplicates removed. + * - #union: Returns an array containing all elements of +self+ and all elements of given arrays, + * duplicates removed. + * - #difference: Returns an array containing all elements of +self+ that are not found + * in any of the given arrays.. + * - #product: Returns or yields all combinations of elements from +self+ and given arrays. + * + * === Methods for Iterating + * + * - #each: Passes each element to a given block. + * - #reverse_each: Passes each element, in reverse order, to a given block. + * - #each_index: Passes each element index to a given block. + * - #cycle: Calls a given block with each element, then does so again, + * for a specified number of times, or forever. + * - #combination: Calls a given block with combinations of elements of +self+; + * a combination does not use the same element more than once. + * - #permutation: Calls a given block with permutations of elements of +self+; + * a permutation does not use the same element more than once. + * - #repeated_combination: Calls a given block with combinations of elements of +self+; + * a combination may use the same element more than once. + * - #repeated_permutation: Calls a given block with permutations of elements of +self+; + * a permutation may use the same element more than once. + * + * === Methods for Converting + * + * - #map, #collect: Returns an array containing the block return-value for each element. + * - #map!, #collect!: Replaces each element with a block return-value. + * - #flatten: Returns an array that is a recursive flattening of +self+. + * - #flatten!: Replaces each nested array in +self+ with the elements from that array. + * - #inspect, #to_s: Returns a new String containing the elements. + * - #join: Returns a newsString containing the elements joined by the field separator. + * - #to_a: Returns +self+ or a new array containing all elements. + * - #to_ary: Returns +self+. + * - #to_h: Returns a new hash formed from the elements. + * - #transpose: Transposes +self+, which must be an array of arrays. + * - #zip: Returns a new array of arrays containing +self+ and given arrays; + * follow the link for details. + * + * === Other Methods + * + * - #*: Returns one of the following: + * + * - With integer argument +n+, a new array that is the concatenation + * of +n+ copies of +self+. + * - With string argument +field_separator+, a new string that is equivalent to + * <tt>join(field_separator)</tt>. + * + * - #abbrev: Returns a hash of unambiguous abbreviations for elements. + * - #pack: Packs the elements into a binary sequence. + * - #sum: Returns a sum of elements according to either <tt>+</tt> or a given block. */ void Init_Array(void) { -#undef rb_intern -#define rb_intern(str) rb_intern_const(str) - rb_cArray = rb_define_class("Array", rb_cObject); rb_include_module(rb_cArray, rb_mEnumerable); rb_define_alloc_func(rb_cArray, empty_ary_alloc); + rb_define_singleton_method(rb_cArray, "new", rb_ary_s_new, -1); rb_define_singleton_method(rb_cArray, "[]", rb_ary_s_create, -1); rb_define_singleton_method(rb_cArray, "try_convert", rb_ary_s_try_convert, 1); rb_define_method(rb_cArray, "initialize", rb_ary_initialize, -1); @@ -6240,7 +8827,6 @@ Init_Array(void) rb_define_method(rb_cArray, "to_a", rb_ary_to_a, 0); rb_define_method(rb_cArray, "to_h", rb_ary_to_h, 0); rb_define_method(rb_cArray, "to_ary", rb_ary_to_ary_m, 0); - rb_define_method(rb_cArray, "frozen?", rb_ary_frozen_p, 0); rb_define_method(rb_cArray, "==", rb_ary_equal, 1); rb_define_method(rb_cArray, "eql?", rb_ary_eql, 1); @@ -6253,6 +8839,10 @@ Init_Array(void) rb_define_method(rb_cArray, "first", rb_ary_first, -1); rb_define_method(rb_cArray, "last", rb_ary_last, -1); rb_define_method(rb_cArray, "concat", rb_ary_concat_multi, -1); + rb_define_method(rb_cArray, "union", rb_ary_union_multi, -1); + rb_define_method(rb_cArray, "difference", rb_ary_difference_multi, -1); + rb_define_method(rb_cArray, "intersection", rb_ary_intersection_multi, -1); + rb_define_method(rb_cArray, "intersect?", rb_ary_intersect_p, 1); rb_define_method(rb_cArray, "<<", rb_ary_push, 1); rb_define_method(rb_cArray, "push", rb_ary_push_m, -1); rb_define_alias(rb_cArray, "append", "push"); @@ -6265,7 +8855,7 @@ Init_Array(void) rb_define_method(rb_cArray, "each_index", rb_ary_each_index, 0); rb_define_method(rb_cArray, "reverse_each", rb_ary_reverse_each, 0); rb_define_method(rb_cArray, "length", rb_ary_length, 0); - rb_define_alias(rb_cArray, "size", "length"); + rb_define_method(rb_cArray, "size", rb_ary_length, 0); rb_define_method(rb_cArray, "empty?", rb_ary_empty_p, 0); rb_define_method(rb_cArray, "find_index", rb_ary_index, -1); rb_define_method(rb_cArray, "index", rb_ary_index, -1); @@ -6284,6 +8874,8 @@ Init_Array(void) rb_define_method(rb_cArray, "map!", rb_ary_collect_bang, 0); rb_define_method(rb_cArray, "select", rb_ary_select, 0); rb_define_method(rb_cArray, "select!", rb_ary_select_bang, 0); + rb_define_method(rb_cArray, "filter", rb_ary_select, 0); + rb_define_method(rb_cArray, "filter!", rb_ary_select_bang, 0); rb_define_method(rb_cArray, "keep_if", rb_ary_keep_if, 0); rb_define_method(rb_cArray, "values_at", rb_ary_values_at, -1); rb_define_method(rb_cArray, "delete", rb_ary_delete, 1); @@ -6314,6 +8906,7 @@ Init_Array(void) rb_define_method(rb_cArray, "max", rb_ary_max, -1); rb_define_method(rb_cArray, "min", rb_ary_min, -1); + rb_define_method(rb_cArray, "minmax", rb_ary_minmax, 0); rb_define_method(rb_cArray, "uniq", rb_ary_uniq, 0); rb_define_method(rb_cArray, "uniq!", rb_ary_uniq_bang, 0); @@ -6322,9 +8915,6 @@ Init_Array(void) rb_define_method(rb_cArray, "flatten", rb_ary_flatten, -1); rb_define_method(rb_cArray, "flatten!", rb_ary_flatten_bang, -1); rb_define_method(rb_cArray, "count", rb_ary_count, -1); - rb_define_method(rb_cArray, "shuffle!", rb_ary_shuffle_bang, -1); - rb_define_method(rb_cArray, "shuffle", rb_ary_shuffle, -1); - rb_define_method(rb_cArray, "sample", rb_ary_sample, -1); rb_define_method(rb_cArray, "cycle", rb_ary_cycle, -1); rb_define_method(rb_cArray, "permutation", rb_ary_permutation, -1); rb_define_method(rb_cArray, "combination", rb_ary_combination, 1); @@ -6339,8 +8929,13 @@ Init_Array(void) rb_define_method(rb_cArray, "bsearch", rb_ary_bsearch, 0); rb_define_method(rb_cArray, "bsearch_index", rb_ary_bsearch_index, 0); rb_define_method(rb_cArray, "any?", rb_ary_any_p, -1); + rb_define_method(rb_cArray, "all?", rb_ary_all_p, -1); + rb_define_method(rb_cArray, "none?", rb_ary_none_p, -1); + rb_define_method(rb_cArray, "one?", rb_ary_one_p, -1); rb_define_method(rb_cArray, "dig", rb_ary_dig, -1); rb_define_method(rb_cArray, "sum", rb_ary_sum, -1); - id_random = rb_intern("random"); + rb_define_method(rb_cArray, "deconstruct", rb_ary_deconstruct, 0); } + +#include "array.rbinc" diff --git a/array.rb b/array.rb new file mode 100644 index 0000000000..b9fa9844e6 --- /dev/null +++ b/array.rb @@ -0,0 +1,69 @@ +class Array + # call-seq: + # array.shuffle!(random: Random) -> array + # + # Shuffles the elements of +self+ in place. + # a = [1, 2, 3] #=> [1, 2, 3] + # a.shuffle! #=> [2, 3, 1] + # a #=> [2, 3, 1] + # + # The optional +random+ argument will be used as the random number generator: + # a.shuffle!(random: Random.new(1)) #=> [1, 3, 2] + def shuffle!(random: Random) + Primitive.rb_ary_shuffle_bang(random) + end + + # call-seq: + # array.shuffle(random: Random) -> new_ary + # + # Returns a new array with elements of +self+ shuffled. + # a = [1, 2, 3] #=> [1, 2, 3] + # a.shuffle #=> [2, 3, 1] + # a #=> [1, 2, 3] + # + # The optional +random+ argument will be used as the random number generator: + # a.shuffle(random: Random.new(1)) #=> [1, 3, 2] + def shuffle(random: Random) + Primitive.rb_ary_shuffle(random) + end + + # call-seq: + # array.sample(random: Random) -> object + # array.sample(n, random: Random) -> new_ary + # + # Returns random elements from +self+. + # + # When no arguments are given, returns a random element from +self+: + # a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] + # a.sample # => 3 + # a.sample # => 8 + # If +self+ is empty, returns +nil+. + # + # When argument +n+ is given, returns a new \Array containing +n+ random + # elements from +self+: + # a.sample(3) # => [8, 9, 2] + # a.sample(6) # => [9, 6, 10, 3, 1, 4] + # Returns no more than <tt>a.size</tt> elements + # (because no new duplicates are introduced): + # a.sample(a.size * 2) # => [6, 4, 1, 8, 5, 9, 10, 2, 3, 7] + # But +self+ may contain duplicates: + # a = [1, 1, 1, 2, 2, 3] + # a.sample(a.size * 2) # => [1, 1, 3, 2, 1, 2] + # The argument +n+ must be a non-negative numeric value. + # The order of the result array is unrelated to the order of +self+. + # Returns a new empty \Array if +self+ is empty. + # + # The optional +random+ argument will be used as the random number generator: + # a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] + # a.sample(random: Random.new(1)) #=> 6 + # a.sample(4, random: Random.new(1)) #=> [6, 10, 9, 2] + def sample(n = (ary = false), random: Random) + if Primitive.mandatory_only? + # Primitive.cexpr! %{ rb_ary_sample(self, rb_cRandom, Qfalse, Qfalse) } + Primitive.ary_sample0 + else + # Primitive.cexpr! %{ rb_ary_sample(self, random, n, ary) } + Primitive.ary_sample(random, n, ary) + end + end +end @@ -0,0 +1,773 @@ +/* indent-tabs-mode: nil */ +#include "internal.h" +#include "internal/parse.h" +#include "internal/symbol.h" +#include "internal/warnings.h" +#include "iseq.h" +#include "node.h" +#include "ruby.h" +#include "ruby/encoding.h" +#include "ruby/util.h" +#include "vm_core.h" + +#include "builtin.h" + +static VALUE rb_mAST; +static VALUE rb_cNode; + +struct ASTNodeData { + rb_ast_t *ast; + const NODE *node; +}; + +static void +node_gc_mark(void *ptr) +{ + struct ASTNodeData *data = (struct ASTNodeData *)ptr; + rb_gc_mark((VALUE)data->ast); +} + +static size_t +node_memsize(const void *ptr) +{ + struct ASTNodeData *data = (struct ASTNodeData *)ptr; + return rb_ast_memsize(data->ast); +} + +static const rb_data_type_t rb_node_type = { + "AST/node", + {node_gc_mark, RUBY_TYPED_DEFAULT_FREE, node_memsize,}, + 0, 0, + RUBY_TYPED_FREE_IMMEDIATELY, +}; + +static VALUE rb_ast_node_alloc(VALUE klass); + +static void +setup_node(VALUE obj, rb_ast_t *ast, const NODE *node) +{ + struct ASTNodeData *data; + + TypedData_Get_Struct(obj, struct ASTNodeData, &rb_node_type, data); + data->ast = ast; + data->node = node; +} + +static VALUE +ast_new_internal(rb_ast_t *ast, const NODE *node) +{ + VALUE obj; + + obj = rb_ast_node_alloc(rb_cNode); + setup_node(obj, ast, node); + + return obj; +} + +static VALUE rb_ast_parse_str(VALUE str, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens); +static VALUE rb_ast_parse_file(VALUE path, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens); + +static VALUE +ast_parse_new(void) +{ + return rb_parser_set_context(rb_parser_new(), NULL, 0); +} + +static VALUE +ast_parse_done(rb_ast_t *ast) +{ + if (!ast->body.root) { + rb_ast_dispose(ast); + rb_exc_raise(GET_EC()->errinfo); + } + + return ast_new_internal(ast, (NODE *)ast->body.root); +} + +static VALUE +ast_s_parse(rb_execution_context_t *ec, VALUE module, VALUE str, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + return rb_ast_parse_str(str, keep_script_lines, error_tolerant, keep_tokens); +} + +static VALUE +rb_ast_parse_str(VALUE str, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + rb_ast_t *ast = 0; + + StringValue(str); + VALUE vparser = ast_parse_new(); + if (RTEST(keep_script_lines)) rb_parser_keep_script_lines(vparser); + if (RTEST(error_tolerant)) rb_parser_error_tolerant(vparser); + if (RTEST(keep_tokens)) rb_parser_keep_tokens(vparser); + ast = rb_parser_compile_string_path(vparser, Qnil, str, 1); + return ast_parse_done(ast); +} + +static VALUE +ast_s_parse_file(rb_execution_context_t *ec, VALUE module, VALUE path, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + return rb_ast_parse_file(path, keep_script_lines, error_tolerant, keep_tokens); +} + +static VALUE +rb_ast_parse_file(VALUE path, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + VALUE f; + rb_ast_t *ast = 0; + rb_encoding *enc = rb_utf8_encoding(); + + FilePathValue(path); + f = rb_file_open_str(path, "r"); + rb_funcall(f, rb_intern("set_encoding"), 2, rb_enc_from_encoding(enc), rb_str_new_cstr("-")); + VALUE vparser = ast_parse_new(); + if (RTEST(keep_script_lines)) rb_parser_keep_script_lines(vparser); + if (RTEST(error_tolerant)) rb_parser_error_tolerant(vparser); + if (RTEST(keep_tokens)) rb_parser_keep_tokens(vparser); + ast = rb_parser_compile_file_path(vparser, Qnil, f, 1); + rb_io_close(f); + return ast_parse_done(ast); +} + +static VALUE +lex_array(VALUE array, int index) +{ + VALUE str = rb_ary_entry(array, index); + if (!NIL_P(str)) { + StringValue(str); + if (!rb_enc_asciicompat(rb_enc_get(str))) { + rb_raise(rb_eArgError, "invalid source encoding"); + } + } + return str; +} + +static VALUE +rb_ast_parse_array(VALUE array, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + rb_ast_t *ast = 0; + + array = rb_check_array_type(array); + VALUE vparser = ast_parse_new(); + if (RTEST(keep_script_lines)) rb_parser_keep_script_lines(vparser); + if (RTEST(error_tolerant)) rb_parser_error_tolerant(vparser); + if (RTEST(keep_tokens)) rb_parser_keep_tokens(vparser); + ast = rb_parser_compile_generic(vparser, lex_array, Qnil, array, 1); + return ast_parse_done(ast); +} + +static VALUE node_children(rb_ast_t*, const NODE*); + +static VALUE +node_find(VALUE self, const int node_id) +{ + VALUE ary; + long i; + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + if (nd_node_id(data->node) == node_id) return self; + + ary = node_children(data->ast, data->node); + + for (i = 0; i < RARRAY_LEN(ary); i++) { + VALUE child = RARRAY_AREF(ary, i); + + if (CLASS_OF(child) == rb_cNode) { + VALUE result = node_find(child, node_id); + if (RTEST(result)) return result; + } + } + + return Qnil; +} + +extern VALUE rb_e_script; + +static VALUE +script_lines(VALUE path) +{ + VALUE hash, lines; + ID script_lines; + CONST_ID(script_lines, "SCRIPT_LINES__"); + if (!rb_const_defined_at(rb_cObject, script_lines)) return Qnil; + hash = rb_const_get_at(rb_cObject, script_lines); + if (!RB_TYPE_P(hash, T_HASH)) return Qnil; + lines = rb_hash_lookup(hash, path); + if (!RB_TYPE_P(lines, T_ARRAY)) return Qnil; + return lines; +} + +static VALUE +node_id_for_backtrace_location(rb_execution_context_t *ec, VALUE module, VALUE location) +{ + int node_id; + + if (!rb_frame_info_p(location)) { + rb_raise(rb_eTypeError, "Thread::Backtrace::Location object expected"); + } + + node_id = rb_get_node_id_from_frame_info(location); + if (node_id == -1) { + return Qnil; + } + + return INT2NUM(node_id); +} + +static VALUE +ast_s_of(rb_execution_context_t *ec, VALUE module, VALUE body, VALUE keep_script_lines, VALUE error_tolerant, VALUE keep_tokens) +{ + VALUE node, lines = Qnil; + const rb_iseq_t *iseq; + int node_id; + + if (rb_frame_info_p(body)) { + iseq = rb_get_iseq_from_frame_info(body); + node_id = rb_get_node_id_from_frame_info(body); + } + else { + iseq = NULL; + + if (rb_obj_is_proc(body)) { + iseq = vm_proc_iseq(body); + + if (!rb_obj_is_iseq((VALUE)iseq)) return Qnil; + } + else { + iseq = rb_method_iseq(body); + } + if (iseq) { + node_id = ISEQ_BODY(iseq)->location.node_id; + } + } + + if (!iseq) { + return Qnil; + } + lines = ISEQ_BODY(iseq)->variable.script_lines; + + VALUE path = rb_iseq_path(iseq); + int e_option = RSTRING_LEN(path) == 2 && memcmp(RSTRING_PTR(path), "-e", 2) == 0; + + if (NIL_P(lines) && rb_iseq_from_eval_p(iseq) && !e_option) { + rb_raise(rb_eArgError, "cannot get AST for method defined in eval"); + } + + if (!NIL_P(lines) || !NIL_P(lines = script_lines(path))) { + node = rb_ast_parse_array(lines, keep_script_lines, error_tolerant, keep_tokens); + } + else if (e_option) { + node = rb_ast_parse_str(rb_e_script, keep_script_lines, error_tolerant, keep_tokens); + } + else { + node = rb_ast_parse_file(path, keep_script_lines, error_tolerant, keep_tokens); + } + + return node_find(node, node_id); +} + +static VALUE +rb_ast_node_alloc(VALUE klass) +{ + struct ASTNodeData *data; + VALUE obj = TypedData_Make_Struct(klass, struct ASTNodeData, &rb_node_type, data); + + return obj; +} + +static const char* +node_type_to_str(const NODE *node) +{ + return (ruby_node_name(nd_type(node)) + rb_strlen_lit("NODE_")); +} + +static VALUE +ast_node_type(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return rb_sym_intern_ascii_cstr(node_type_to_str(data->node)); +} + +static VALUE +ast_node_node_id(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return INT2FIX(nd_node_id(data->node)); +} + +#define NEW_CHILD(ast, node) node ? ast_new_internal(ast, node) : Qnil + +static VALUE +rb_ary_new_from_node_args(rb_ast_t *ast, long n, ...) +{ + va_list ar; + VALUE ary; + long i; + + ary = rb_ary_new2(n); + + va_start(ar, n); + for (i=0; i<n; i++) { + NODE *node; + node = va_arg(ar, NODE *); + rb_ary_push(ary, NEW_CHILD(ast, node)); + } + va_end(ar); + return ary; +} + +static VALUE +dump_block(rb_ast_t *ast, const NODE *node) +{ + VALUE ary = rb_ary_new(); + do { + rb_ary_push(ary, NEW_CHILD(ast, node->nd_head)); + } while (node->nd_next && + nd_type_p(node->nd_next, NODE_BLOCK) && + (node = node->nd_next, 1)); + if (node->nd_next) { + rb_ary_push(ary, NEW_CHILD(ast, node->nd_next)); + } + + return ary; +} + +static VALUE +dump_array(rb_ast_t *ast, const NODE *node) +{ + VALUE ary = rb_ary_new(); + rb_ary_push(ary, NEW_CHILD(ast, node->nd_head)); + + while (node->nd_next && nd_type_p(node->nd_next, NODE_LIST)) { + node = node->nd_next; + rb_ary_push(ary, NEW_CHILD(ast, node->nd_head)); + } + rb_ary_push(ary, NEW_CHILD(ast, node->nd_next)); + + return ary; +} + +static VALUE +var_name(ID id) +{ + if (!id) return Qnil; + if (!rb_id2str(id)) return Qnil; + return ID2SYM(id); +} + +static VALUE +no_name_rest(void) +{ + ID rest; + CONST_ID(rest, "NODE_SPECIAL_NO_NAME_REST"); + return ID2SYM(rest); +} + +static VALUE +rest_arg(rb_ast_t *ast, const NODE *rest_arg) +{ + return NODE_NAMED_REST_P(rest_arg) ? NEW_CHILD(ast, rest_arg) : no_name_rest(); +} + +static VALUE +node_children(rb_ast_t *ast, const NODE *node) +{ + char name[DECIMAL_SIZE_OF_BITS(sizeof(long) * CHAR_BIT) + 2]; /* including '$' */ + + enum node_type type = nd_type(node); + switch (type) { + case NODE_BLOCK: + return dump_block(ast, node); + case NODE_IF: + return rb_ary_new_from_node_args(ast, 3, node->nd_cond, node->nd_body, node->nd_else); + case NODE_UNLESS: + return rb_ary_new_from_node_args(ast, 3, node->nd_cond, node->nd_body, node->nd_else); + case NODE_CASE: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_CASE2: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_CASE3: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_WHEN: + return rb_ary_new_from_node_args(ast, 3, node->nd_head, node->nd_body, node->nd_next); + case NODE_IN: + return rb_ary_new_from_node_args(ast, 3, node->nd_head, node->nd_body, node->nd_next); + case NODE_WHILE: + case NODE_UNTIL: + return rb_ary_push(rb_ary_new_from_node_args(ast, 2, node->nd_cond, node->nd_body), + RBOOL(node->nd_state)); + case NODE_ITER: + case NODE_FOR: + return rb_ary_new_from_node_args(ast, 2, node->nd_iter, node->nd_body); + case NODE_FOR_MASGN: + return rb_ary_new_from_node_args(ast, 1, node->nd_var); + case NODE_BREAK: + case NODE_NEXT: + case NODE_RETURN: + return rb_ary_new_from_node_args(ast, 1, node->nd_stts); + case NODE_REDO: + return rb_ary_new_from_node_args(ast, 0); + case NODE_RETRY: + return rb_ary_new_from_node_args(ast, 0); + case NODE_BEGIN: + return rb_ary_new_from_node_args(ast, 1, node->nd_body); + case NODE_RESCUE: + return rb_ary_new_from_node_args(ast, 3, node->nd_head, node->nd_resq, node->nd_else); + case NODE_RESBODY: + return rb_ary_new_from_node_args(ast, 3, node->nd_args, node->nd_body, node->nd_head); + case NODE_ENSURE: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_ensr); + case NODE_AND: + case NODE_OR: + { + VALUE ary = rb_ary_new(); + + while (1) { + rb_ary_push(ary, NEW_CHILD(ast, node->nd_1st)); + if (!node->nd_2nd || !nd_type_p(node->nd_2nd, type)) + break; + node = node->nd_2nd; + } + rb_ary_push(ary, NEW_CHILD(ast, node->nd_2nd)); + return ary; + } + case NODE_MASGN: + if (NODE_NAMED_REST_P(node->nd_args)) { + return rb_ary_new_from_node_args(ast, 3, node->nd_value, node->nd_head, node->nd_args); + } + else { + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_value), + NEW_CHILD(ast, node->nd_head), + no_name_rest()); + } + case NODE_LASGN: + case NODE_DASGN: + case NODE_IASGN: + case NODE_CVASGN: + case NODE_GASGN: + if (NODE_REQUIRED_KEYWORD_P(node)) { + return rb_ary_new_from_args(2, var_name(node->nd_vid), ID2SYM(rb_intern("NODE_SPECIAL_REQUIRED_KEYWORD"))); + } + return rb_ary_new_from_args(2, var_name(node->nd_vid), NEW_CHILD(ast, node->nd_value)); + case NODE_CDECL: + if (node->nd_vid) { + return rb_ary_new_from_args(2, ID2SYM(node->nd_vid), NEW_CHILD(ast, node->nd_value)); + } + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_else), ID2SYM(node->nd_else->nd_mid), NEW_CHILD(ast, node->nd_value)); + case NODE_OP_ASGN1: + return rb_ary_new_from_args(4, NEW_CHILD(ast, node->nd_recv), + ID2SYM(node->nd_mid), + NEW_CHILD(ast, node->nd_args->nd_head), + NEW_CHILD(ast, node->nd_args->nd_body)); + case NODE_OP_ASGN2: + return rb_ary_new_from_args(5, NEW_CHILD(ast, node->nd_recv), + RBOOL(node->nd_next->nd_aid), + ID2SYM(node->nd_next->nd_vid), + ID2SYM(node->nd_next->nd_mid), + NEW_CHILD(ast, node->nd_value)); + case NODE_OP_ASGN_AND: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head), ID2SYM(idANDOP), + NEW_CHILD(ast, node->nd_value)); + case NODE_OP_ASGN_OR: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head), ID2SYM(idOROP), + NEW_CHILD(ast, node->nd_value)); + case NODE_OP_CDECL: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head), + ID2SYM(node->nd_aid), + NEW_CHILD(ast, node->nd_value)); + case NODE_CALL: + case NODE_OPCALL: + case NODE_QCALL: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv), + ID2SYM(node->nd_mid), + NEW_CHILD(ast, node->nd_args)); + case NODE_FCALL: + return rb_ary_new_from_args(2, ID2SYM(node->nd_mid), + NEW_CHILD(ast, node->nd_args)); + case NODE_VCALL: + return rb_ary_new_from_args(1, ID2SYM(node->nd_mid)); + case NODE_SUPER: + return rb_ary_new_from_node_args(ast, 1, node->nd_args); + case NODE_ZSUPER: + return rb_ary_new_from_node_args(ast, 0); + case NODE_LIST: + case NODE_VALUES: + return dump_array(ast, node); + case NODE_ZLIST: + return rb_ary_new_from_node_args(ast, 0); + case NODE_HASH: + return rb_ary_new_from_node_args(ast, 1, node->nd_head); + case NODE_YIELD: + return rb_ary_new_from_node_args(ast, 1, node->nd_head); + case NODE_LVAR: + case NODE_DVAR: + return rb_ary_new_from_args(1, var_name(node->nd_vid)); + case NODE_IVAR: + case NODE_CONST: + case NODE_CVAR: + case NODE_GVAR: + return rb_ary_new_from_args(1, ID2SYM(node->nd_vid)); + case NODE_NTH_REF: + snprintf(name, sizeof(name), "$%ld", node->nd_nth); + return rb_ary_new_from_args(1, ID2SYM(rb_intern(name))); + case NODE_BACK_REF: + name[0] = '$'; + name[1] = (char)node->nd_nth; + name[2] = '\0'; + return rb_ary_new_from_args(1, ID2SYM(rb_intern(name))); + case NODE_MATCH2: + if (node->nd_args) { + return rb_ary_new_from_node_args(ast, 3, node->nd_recv, node->nd_value, node->nd_args); + } + return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_value); + case NODE_MATCH3: + return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_value); + case NODE_MATCH: + case NODE_LIT: + case NODE_STR: + case NODE_XSTR: + return rb_ary_new_from_args(1, node->nd_lit); + case NODE_ONCE: + return rb_ary_new_from_node_args(ast, 1, node->nd_body); + case NODE_DSTR: + case NODE_DXSTR: + case NODE_DREGX: + case NODE_DSYM: + { + NODE *n = node->nd_next; + VALUE head = Qnil, next = Qnil; + if (n) { + head = NEW_CHILD(ast, n->nd_head); + next = NEW_CHILD(ast, n->nd_next); + } + return rb_ary_new_from_args(3, node->nd_lit, head, next); + } + case NODE_EVSTR: + return rb_ary_new_from_node_args(ast, 1, node->nd_body); + case NODE_ARGSCAT: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_ARGSPUSH: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_SPLAT: + return rb_ary_new_from_node_args(ast, 1, node->nd_head); + case NODE_BLOCK_PASS: + return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body); + case NODE_DEFN: + return rb_ary_new_from_args(2, ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_defn)); + case NODE_DEFS: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv), ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_defn)); + case NODE_ALIAS: + return rb_ary_new_from_node_args(ast, 2, node->nd_1st, node->nd_2nd); + case NODE_VALIAS: + return rb_ary_new_from_args(2, ID2SYM(node->nd_alias), ID2SYM(node->nd_orig)); + case NODE_UNDEF: + return rb_ary_new_from_node_args(ast, 1, node->nd_undef); + case NODE_CLASS: + return rb_ary_new_from_node_args(ast, 3, node->nd_cpath, node->nd_super, node->nd_body); + case NODE_MODULE: + return rb_ary_new_from_node_args(ast, 2, node->nd_cpath, node->nd_body); + case NODE_SCLASS: + return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_body); + case NODE_COLON2: + return rb_ary_new_from_args(2, NEW_CHILD(ast, node->nd_head), ID2SYM(node->nd_mid)); + case NODE_COLON3: + return rb_ary_new_from_args(1, ID2SYM(node->nd_mid)); + case NODE_DOT2: + case NODE_DOT3: + case NODE_FLIP2: + case NODE_FLIP3: + return rb_ary_new_from_node_args(ast, 2, node->nd_beg, node->nd_end); + case NODE_SELF: + return rb_ary_new_from_node_args(ast, 0); + case NODE_NIL: + return rb_ary_new_from_node_args(ast, 0); + case NODE_TRUE: + return rb_ary_new_from_node_args(ast, 0); + case NODE_FALSE: + return rb_ary_new_from_node_args(ast, 0); + case NODE_ERRINFO: + return rb_ary_new_from_node_args(ast, 0); + case NODE_DEFINED: + return rb_ary_new_from_node_args(ast, 1, node->nd_head); + case NODE_POSTEXE: + return rb_ary_new_from_node_args(ast, 1, node->nd_body); + case NODE_ATTRASGN: + return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv), ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_args)); + case NODE_LAMBDA: + return rb_ary_new_from_node_args(ast, 1, node->nd_body); + case NODE_OPT_ARG: + return rb_ary_new_from_node_args(ast, 2, node->nd_body, node->nd_next); + case NODE_KW_ARG: + return rb_ary_new_from_node_args(ast, 2, node->nd_body, node->nd_next); + case NODE_POSTARG: + if (NODE_NAMED_REST_P(node->nd_1st)) { + return rb_ary_new_from_node_args(ast, 2, node->nd_1st, node->nd_2nd); + } + return rb_ary_new_from_args(2, no_name_rest(), + NEW_CHILD(ast, node->nd_2nd)); + case NODE_ARGS: + { + struct rb_args_info *ainfo = node->nd_ainfo; + return rb_ary_new_from_args(10, + INT2NUM(ainfo->pre_args_num), + NEW_CHILD(ast, ainfo->pre_init), + NEW_CHILD(ast, ainfo->opt_args), + var_name(ainfo->first_post_arg), + INT2NUM(ainfo->post_args_num), + NEW_CHILD(ast, ainfo->post_init), + (ainfo->rest_arg == NODE_SPECIAL_EXCESSIVE_COMMA + ? ID2SYM(rb_intern("NODE_SPECIAL_EXCESSIVE_COMMA")) + : var_name(ainfo->rest_arg)), + (ainfo->no_kwarg ? Qfalse : NEW_CHILD(ast, ainfo->kw_args)), + (ainfo->no_kwarg ? Qfalse : NEW_CHILD(ast, ainfo->kw_rest_arg)), + var_name(ainfo->block_arg)); + } + case NODE_SCOPE: + { + rb_ast_id_table_t *tbl = node->nd_tbl; + int i, size = tbl ? tbl->size : 0; + VALUE locals = rb_ary_new_capa(size); + for (i = 0; i < size; i++) { + rb_ary_push(locals, var_name(tbl->ids[i])); + } + return rb_ary_new_from_args(3, locals, NEW_CHILD(ast, node->nd_args), NEW_CHILD(ast, node->nd_body)); + } + case NODE_ARYPTN: + { + struct rb_ary_pattern_info *apinfo = node->nd_apinfo; + VALUE rest = rest_arg(ast, apinfo->rest_arg); + return rb_ary_new_from_args(4, + NEW_CHILD(ast, node->nd_pconst), + NEW_CHILD(ast, apinfo->pre_args), + rest, + NEW_CHILD(ast, apinfo->post_args)); + } + case NODE_FNDPTN: + { + struct rb_fnd_pattern_info *fpinfo = node->nd_fpinfo; + VALUE pre_rest = rest_arg(ast, fpinfo->pre_rest_arg); + VALUE post_rest = rest_arg(ast, fpinfo->post_rest_arg); + return rb_ary_new_from_args(4, + NEW_CHILD(ast, node->nd_pconst), + pre_rest, + NEW_CHILD(ast, fpinfo->args), + post_rest); + } + case NODE_HSHPTN: + { + VALUE kwrest = node->nd_pkwrestarg == NODE_SPECIAL_NO_REST_KEYWORD ? ID2SYM(rb_intern("NODE_SPECIAL_NO_REST_KEYWORD")) : + NEW_CHILD(ast, node->nd_pkwrestarg); + + return rb_ary_new_from_args(3, + NEW_CHILD(ast, node->nd_pconst), + NEW_CHILD(ast, node->nd_pkwargs), + kwrest); + } + case NODE_ERROR: + return rb_ary_new_from_node_args(ast, 0); + case NODE_ARGS_AUX: + case NODE_LAST: + break; + } + + rb_bug("node_children: unknown node: %s", ruby_node_name(type)); +} + +static VALUE +ast_node_children(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return node_children(data->ast, data->node); +} + +static VALUE +ast_node_first_lineno(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return INT2NUM(nd_first_lineno(data->node)); +} + +static VALUE +ast_node_first_column(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return INT2NUM(nd_first_column(data->node)); +} + +static VALUE +ast_node_last_lineno(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return INT2NUM(nd_last_lineno(data->node)); +} + +static VALUE +ast_node_last_column(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return INT2NUM(nd_last_column(data->node)); +} + +static VALUE +ast_node_all_tokens(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + return rb_ast_tokens(data->ast); +} + +static VALUE +ast_node_inspect(rb_execution_context_t *ec, VALUE self) +{ + VALUE str; + VALUE cname; + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + + cname = rb_class_path(rb_obj_class(self)); + str = rb_str_new2("#<"); + + rb_str_append(str, cname); + rb_str_catf(str, ":%s@%d:%d-%d:%d>", + node_type_to_str(data->node), + nd_first_lineno(data->node), nd_first_column(data->node), + nd_last_lineno(data->node), nd_last_column(data->node)); + + return str; +} + +static VALUE +ast_node_script_lines(rb_execution_context_t *ec, VALUE self) +{ + struct ASTNodeData *data; + TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data); + VALUE ret = data->ast->body.script_lines; + if (!RB_TYPE_P(ret, T_ARRAY)) return Qnil; + return ret; +} + +#include "ast.rbinc" + +void +Init_ast(void) +{ + rb_mAST = rb_define_module_under(rb_cRubyVM, "AbstractSyntaxTree"); + rb_cNode = rb_define_class_under(rb_mAST, "Node", rb_cObject); + rb_undef_alloc_func(rb_cNode); +} diff --git a/ast.rb b/ast.rb new file mode 100644 |