summaryrefslogtreecommitdiff
path: root/test/ruby
AgeCommit message (Collapse)Author
2023-12-16[Bug #20068] Encoding does not matter to empty stringsNobuyoshi Nakada
2023-12-15[PRISM] Compile CallTargetNodeMatt Valentine-House
2023-12-15[PRISM] Compile IndexTargetNodeMatt Valentine-House
2023-12-15Update error message in test_rubyoptionsKevin Newton
2023-12-15Introduce --parser runtime flagHParker
Introduce runtime flag for specifying the parser, ``` ruby --parser=prism ``` also update the description: ``` $ ruby --parser=prism --version ruby 3.3.0dev (2023-12-08T04:47:14Z add-parser-runtime.. 0616384c9f) +PRISM [x86_64-darwin23] ``` [Bug #20044]
2023-12-16[Bug #20062] Fixed numbered parameter syntax errorNobuyoshi Nakada
At the method definition, the local scope that saves the context of the numbered parameters needs to be pushed before saving.
2023-12-15[PRISM] Fix `compile_prism` when src is a fileeileencodes
`compile_prism` can take a source and file (and other arguments) or a file as the source. `compile` checks if the source is a file and if it is converts it. `compile_prism` is now doing the same thing. On the Ruby side `compile` handles a file [here](https://github.com/ruby/ruby/blob/master/iseq.c#L1159-L1162). Before: ``` "********* Ruby *************" == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(26,21)> local table (size: 1, argc: 0 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1]) [ 1] name@0 0000 putstring "Prism" ( 25)[Li] 0002 setlocal name@0, 0 0005 putself ( 26)[Li] 0006 putobject "hello, " 0008 getlocal name@0, 0 0011 dup 0012 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0014 anytostring 0015 concatstrings 2 0017 send <calldata!mid:puts, argc:1, FCALL|ARGS_SIMPLE>, nil 0020 leave hello, Prism "********* PRISM *************" ./test.rb:13:in `compile_prism': wrong argument type File (expected String) (TypeError) from ./test.rb:13:in `<main>' make: *** [run] Error 1 ``` After: ``` "********* Ruby *************" == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(26,21)> local table (size: 1, argc: 0 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1]) [ 1] name@0 0000 putstring "Prism" ( 25)[Li] 0002 setlocal name@0, 0 0005 putself ( 26)[Li] 0006 putobject "hello, " 0008 getlocal name@0, 0 0011 dup 0012 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0014 anytostring 0015 concatstrings 2 0017 send <calldata!mid:puts, argc:1, FCALL|ARGS_SIMPLE>, nil 0020 leave "********* PRISM *************" == disasm: #<ISeq:<compiled>@test_code.rb:24 (24,0)-(25,21)> local table (size: 1, argc: 0 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1]) [ 1] name@0 0000 putstring "Prism" ( 24)[Li] 0002 setlocal name@0, 0 0005 putself ( 25)[Li] 0006 putobject "hello, " 0008 getlocal name@0, 0 0011 dup 0012 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0014 anytostring 0015 concatstrings 2 0017 send <calldata!mid:puts, argc:1, FCALL|ARGS_SIMPLE>, nil 0020 leave ( 24) ``` Fixes ruby/prism#1609
2023-12-15add a testKoichi Sasada
proposed at https://bugs.ruby-lang.org/issues/20050#note-5
2023-12-14Remove unused variables in test_call_op_asgn_keywords_mutableJeremy Evans
2023-12-14[PRISM] Implement safe navigation in CallNodesJemma Issroff
This commit implements safe navigation for CallNodes, CallAndWriteNodes and CallOperatorWriteNodes
2023-12-14[PRISM] Fix keyword hash handling in method callsUfuk Kayserilioglu
* Start using the renamed `PM_KEYWORD_HASH_NODE_FLAGS_SYMBOL_KEYS` flag to check if all keys of the keyword hash node are symbols. * For arguments passed as a hash, start marking them as `KW_SPLAT_MUT` only if the number of entries in the hash is greater than 1 (which is what the old compiler used to do).
2023-12-14Pattern matchingKevin Newton
2023-12-14[PRISM] Account for multiple anonymous localsJemma Issroff
This commit adjusts the local table size to be consistent regardless of the number of anonymous locals.
2023-12-14[PRISM] Fix bugs in compiling optional keyword parametersJemma Issroff
This PR fixes two bugs when compiling optional keyword parameters: - It moves keyword parameter compilation to STEP 5 in the parameters sequence, where the rest of compilation happens. This is important because keyword parameter compilation relies on the value of body->param.keyword->bits_start which gets set in an earlier step - It compiles array and hash values for keyword parameters, which it didn't previously
2023-12-14[PRISM] Use frozen flag on StringNodeJemma Issroff
2023-12-14Fix op asgn method calls passing mutable keyword splatsJeremy Evans
When passing the keyword splat to [], it cannot be mutable, because mutating the keyword splat inside [] would result in changes to the keyword splat passed to []=.
2023-12-14[PRISM] Add anon KW args to the block local tableMatt Valentine-House
2023-12-13[PRISM] Add a test with a non-static-literal hash keyUfuk Kayserilioglu
2023-12-13[Prism] Fix InterpolatedMatchLastLine Instructionseileencodes
I looked at this at RubyConf with Kevin, and we noticed that there was a `putobject` empty string missing from the instructions. I just got back around to implementing this and pushing a PR and while doing that noticed that we also have a `getspecial` when we want a `getglobal`. This change adds the `putobject` instruction and replaces the `getspecial` with a `getglobal`. If we look at the parsetree for the following code: ```ruby $pit = '.oo'; if /"#{$pit}"/mix; end ``` We can see it has a `NODE_GVAR` and the [Ruby compiler](https://github.com/ruby/ruby/blob/a4b43e92645e46ee5a8c9af42d3de57cd052e87c/compile.c#L10024-L10030) shows that should use `getglobal. ``` @ NODE_SCOPE (id: 14, line: 1, location: (1,0)-(22,36)) +- nd_tbl: (empty) +- nd_args: | (null node) +- nd_body: @ NODE_BLOCK (id: 12, line: 22, location: (22,0)-(22,36)) +- nd_head (1): | @ NODE_GASGN (id: 0, line: 22, location: (22,0)-(22,12))* | +- nd_vid: :$pit | +- nd_value: | @ NODE_STR (id: 1, line: 22, location: (22,7)-(22,12)) | +- nd_lit: ".oo" +- nd_head (2): @ NODE_IF (id: 11, line: 22, location: (22,14)-(22,36))* +- nd_cond: | @ NODE_MATCH2 (id: 10, line: 22, location: (22,14)-(22,36)) | +- nd_recv: | | @ NODE_DREGX (id: 2, line: 22, location: (22,17)-(22,31)) | | +- nd_lit: "\"" | | +- nd_next->nd_head: | | | @ NODE_EVSTR (id: 4, line: 22, location: (22,19)-(22,26)) | | | +- nd_body: | | | @ NODE_GVAR (id: 3, line: 22, location: (22,21)-(22,25)) | | | +- nd_vid: :$pit | | +- nd_next->nd_next: | | @ NODE_LIST (id: 7, line: 22, location: (22,26)-(22,27)) | | +- as.nd_alen: 1 | | +- nd_head: | | | @ NODE_STR (id: 6, line: 22, location: (22,26)-(22,27)) | | | +- nd_lit: "\"" | | +- nd_next: | | (null node) | +- nd_value: | @ NODE_GVAR (id: 9, line: 22, location: (22,14)-(22,36)) | +- nd_vid: :$_ +- nd_body: | @ NODE_BEGIN (id: 8, line: 22, location: (22,32)-(22,32)) | +- nd_body: | (null node) +- nd_else: (null node) ``` I'm struggling with writing a failing test, but the before/after instructions show that `getglobal` is correct here. I compared the instructions for the other `InterpolatedMatchLastLine` node tests and they also used `getglobal`. I've edited the existing `InterpolatedLastMatchLineNode` test so that it will use that instruction when copied out of the test. Without the quotes it thinks it's just a `MatchLastLineNode`. Incorrect instructions: ``` "********* Ruby *************" == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(22,36)> 0000 putstring ".oo" ( 22)[Li] 0002 setglobal :$pit 0004 putobject "\"" 0006 getglobal :$pit 0008 dup 0009 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0011 anytostring 0012 putobject "\"" 0014 toregexp 7, 3 0017 getglobal :$_ 0019 send <calldata!mid:=~, argc:1, ARGS_SIMPLE>, nil 0022 branchunless 30 0024 jump 26 0026 putnil 0027 jump 31 0029 pop 0030 putnil 0031 leave "********* PRISM *************" == disasm: #<ISeq:<compiled>@<compiled>:21 (21,0)-(21,36)> 0000 putstring ".oo" ( 21)[Li] 0002 setglobal :$pit 0004 putobject "\"" 0006 getglobal :$pit 0008 dup 0009 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0011 anytostring 0012 putobject "\"" 0014 toregexp 7, 3 0017 getspecial 0, 0 0020 send <calldata!mid:=~, argc:1, ARGS_SIMPLE>, nil 0023 branchunless 31 0025 jump 27 0027 putnil 0028 jump 32 0030 pop 0031 putnil 0032 leave ``` Fixed instructions: ``` == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(22,36)> 0000 putstring ".oo" ( 22)[Li] 0002 setglobal :$pit 0004 putobject "\"" 0006 getglobal :$pit 0008 dup 0009 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0011 anytostring 0012 putobject "\"" 0014 toregexp 7, 3 0017 getglobal :$_ 0019 send <calldata!mid:=~, argc:1, ARGS_SIMPLE>, nil 0022 branchunless 30 0024 jump 26 0026 putnil 0027 jump 31 0029 pop 0030 putnil 0031 leave "********* PRISM *************" == disasm: #<ISeq:<compiled>@<compiled>:21 (21,0)-(21,36)> 0000 putstring ".oo" ( 21)[Li] 0002 setglobal :$pit 0004 putobject "\"" 0006 getglobal :$pit 0008 dup 0009 objtostring <calldata!mid:to_s, argc:0, FCALL|ARGS_SIMPLE> 0011 anytostring 0012 putobject "\"" 0014 toregexp 7, 3 0017 getglobal :$_ 0019 send <calldata!mid:=~, argc:1, ARGS_SIMPLE>, nil 0022 branchunless 30 0024 jump 26 0026 putnil 0027 jump 31 0029 pop 0030 putnil 0031 leave ```
2023-12-13Fix memory leak in Hash#compare_by_identityAlan Wu
We didn't free the old ST before overwriting it which caused a leak. Found with RUBY_FREE_ON_EXIT. Co-authored-by: Peter Zhu <peter@peterzhu.ca>
2023-12-13Make String#chomp! raise ArgumentError for 2+ arguments if string is emptyJeremy Evans
String#chomp! returned nil without checking the number of passed arguments in this case.
2023-12-13[PRISM] Generate instruction for when redo_label is setMatt Valentine-House
2023-12-13[PRISM] If receiver on CallNode is SelfNode, use FCALL flagsJemma Issroff
2023-12-13[PRISM] Fix attrset edge caseseileencodes
In some cases code may look like an attrset ID but should actually return the value of the method, not the passed values. In ruby/prism#2051 a flag was added when we have a attribute write call. I used this flag to add the proper instructions when we have a real attrset instead of using `rb_is_attrset_id` which was kind of hacky anyway. The value that should be returned in the newly added test is 42, not 2. Previously the changes we had made returned 2. Related to ruby/prism#1715
2023-12-12Skip a GC test for RJITTakashi Kokubun
It randomly fails like this: https://github.com/ruby/ruby/actions/runs/7191443542/job/19586164973
2023-12-12[PRISM] Correctly parse non-base 10 integers in PrismJemma Issroff
This commit passes an `end` to rb_int_parse_cstr which allows us to correctly parse non-base 10 integers which are enclosed in parenthesis. Prior to this commit, we were getting a putobject nil when compiling `(0o0)` for example.
2023-12-12[PRISM] Fix segv with regex once flageileencodes
When you have an interpolated regex with a `once` flag and local variable is outside the block created by the `once` flag, Prism would see a segv. This is because it was not taking the depth into account. To fix this, we need to add 1 to the `local_depth_offset` on the `scope`. Fixes: ruby/prism#2047
2023-12-12[PRISM] Compile NextNode argumentsMatt Valentine-House
2023-12-12[PRISM] Implementing forwarding of args for ForwardingSuperNodeJemma Issroff
ForwardingSuperNodes need to actually forward any applicable arguments. This commit implements that logic, by using the data stored on the local iseq about the parameters to forward the appropriate arguments.
2023-12-12[Bug #19114] Fix for multiple calls of TracePoint#enableKouhei Yanagita
2023-12-12Fix op asgn calls with keywordsJeremy Evans
Examples of such calls: ```ruby obj[kw: 1] += fo obj[**kw] &&= bar ``` Before this patch, literal keywords would segfault in the compiler, and keyword splat usage would result in TypeError. This handles all cases I can think of: * literal keywords * keyword splats * combined with positional arguments * combined with regular splats * both with and without blocks * both popped and non-popped cases This also makes sure that to_hash is only called once on the keyword splat argument, instead of twice, and make sure it is called before calling to_proc on a passed block. Fixes [Bug #20051] Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
2023-12-12Make WeakKeyMap safe for compaction during allocationPeter Zhu
During allocation, the table may not have been allocated yet which would crash in the st_foreach.
2023-12-12Make WeakMap safe for compaction during allocationPeter Zhu
During allocation, the table may not have been allocated yet which would crash in the st_foreach.
2023-12-11Make Dir#chdir never yield args, and return block return valueJeremy Evans
If no block is given, return 0 instead of nil for consistency with Dir.chdir and Dir.fchdir.
2023-12-12Prevent a warning: method redefinedYusuke Endoh
2023-12-12Prevent warning: possibly useless use of a literal in void contextYusuke Endoh
2023-12-11[PRISM] Properly compile MultiTargetNodes within parametersJemma Issroff
If there are MultiTargetNodes within parameters, we need to iterate over them and compile them individually correctly, once the locals are all in the correct spaces. We need to add one getlocal for the hidden variable, and then can recurse into the MultiTargetNodes themselves
2023-12-11[PRISM] Fixed rest in MultiTargetNodesJemma Issroff
2023-12-11[PRISM] Define and use a pm_add_ensure_iseqJemma Issroff
Prior to this commit, we were using `add_ensure_iseq` which compiled a node as if it was a CRuby node. This commit defines `pm_add_ensure_iseq` which compiles the Prism node appropriately.
2023-12-11[PRISM] Fix ElseNode within CaseNodeJemma Issroff
The logic within the consequent for the CaseNodes in popped cases was incorrect as it wouldn't emit consequent instructions for a popped CaseNode. This commit fixes that.
2023-12-11Add handling of implicit hash argumentsUfuk Kayserilioglu
Arguments that are passed as a hash need special consideration since in certain case they are not treated as keyword arguments. For example, if a call is passing `"a" => 1` as an argument, this will be turned into an implicit hash argument and not a keyword argument. The existing compiler checks to see if all hash nodes can be treated as keyword arguments. If they can, then it will treat them as keyword arguments. If not, then it will treat them as implicit hash arguments. This commit implements the same logic inside the Prism compiler.
2023-12-09Ensure super(**kw, &block) calls kw.to_hash before block.to_procJeremy Evans
Similar as previous commit, but handles the super case with explicit arguments.
2023-12-09Ensure f(**kw, &block) calls kw.to_hash before block.to_procJeremy Evans
Previously, block.to_proc was called first, by vm_caller_setup_arg_block. kw.to_hash was called later inside CALLER_SETUP_ARG or setup_parameters_complex. This adds a splatkw instruction that is inserted before sends with ARGS_BLOCKARG and KW_SPLAT and without KW_SPLAT_MUT. This is not needed in the KW_SPLAT_MUT case, because then you know the value is a hash, and you don't need to call to_hash on it. The splatkw instruction checks whether the second to top block is a hash, and if not, replaces it with the value of calling to_hash on it (using rb_to_hash_type). As it is always before a send with ARGS_BLOCKARG and KW_SPLAT, second to top is the keyword splat, and top is the passed block.
2023-12-09Add a fudge factor to the GC compaction move up/down testsKJ Tsanaktsidis
There seems to be another manifestation of bug #20021, where some of the compaction tests are failing on i686 for unrelated PR's because of fake "live" references to moved objects on the machine stack. We _could_ solve this by counting how many objects are pinned during compaction, but doing that involves pushing down the mark & pin bitset merge into gc_compact_plane and out of gc_compact_page, which I thought was pretty ugly. Now that we've solved bug #20022 though, we're able to compact arbitrarily many objects with GC.verify_compaction_references, so the number of objects we're moving is now 50,000 instead of 500. Since that's now much larger than the number of objects likely to be pinned, I think it's safe enough to just add a fudge-factor to the tests. Any _other_ change in GC.verify_compaction_references that breaks compaction is now highly likely to break the assertion by more than 10 objects, since it's operating on so many more in the first place. [Bug #20021]
2023-12-08[PRISM] Fix `PM_CALL_NODE` assignmenteileencodes
This PR fixes ruby/prism#1963. Array and variable assignment was broken for call nodes. The change checks if the `method_id` is related to assignment and if is adds a `putnil`, `setn` and a `pop`. The incorrect instructions meant that in some cases (demonstrated in tests) the wrong value would be returned. I verified that this fixes the test mentioned in the issue (run: `RUBY_ISEQ_DUMP_DEBUG=prism make test/-ext-/st/test_numhash.rb`) Incorrect instructions: ``` "********* Ruby *************" == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(4,10)> 0000 putnil ( 4)[Li] 0001 putself 0002 send <calldata!mid:tbl, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0005 putself 0006 send <calldata!mid:i, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0009 putself 0010 send <calldata!mid:j, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0013 setn 3 0015 send <calldata!mid:[]=, argc:2, ARGS_SIMPLE>, nil 0018 pop 0019 leave "********* PRISM *************" == disasm: #<ISeq:<compiled>@<compiled>:3 (3,0)-(3,10)> 0000 putself ( 3)[Li] 0001 send <calldata!mid:tbl, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0004 putself 0005 send <calldata!mid:i, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0008 putself 0009 send <calldata!mid:j, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0012 send <calldata!mid:[]=, argc:2, ARGS_SIMPLE>, nil 0015 leave ``` Fixed instructions: ``` "********* Ruby *************" == disasm: #<ISeq:<compiled>@<compiled>:1 (1,0)-(4,10)> 0000 putnil ( 4)[Li] 0001 putself 0002 send <calldata!mid:tbl, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0005 putself 0006 send <calldata!mid:i, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0009 putself 0010 send <calldata!mid:j, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0013 setn 3 0015 send <calldata!mid:[]=, argc:2, ARGS_SIMPLE>, nil 0018 pop 0019 leave "********* PRISM *************" == disasm: #<ISeq:<compiled>@<compiled>:3 (3,0)-(3,10)> 0000 putnil ( 3)[Li] 0001 putself 0002 send <calldata!mid:tbl, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0005 putself 0006 send <calldata!mid:i, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0009 putself 0010 send <calldata!mid:j, argc:0, FCALL|VCALL|ARGS_SIMPLE>, nil 0013 setn 3 0015 send <calldata!mid:[]=, argc:2, ARGS_SIMPLE>, nil 0018 pop 0019 leave ``` Fixes ruby/prism#1963
2023-12-08[Bug #19877] Flip-flop needs to be direct conditionNobuyoshi Nakada
2023-12-07YJIT: Fix on-stack ISEQ comparison for auto_compact (#9164)Takashi Kokubun
2023-12-07Support method calls inside `defined?`Aaron Patterson
This commit supports all kinds of method calls (including methods with parameters) inside `defined?` calls.
2023-12-07Free everything at shutdownAdam Hess
when the RUBY_FREE_ON_SHUTDOWN environment variable is set, manually free memory at shutdown. Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org> Co-authored-by: Peter Zhu <peter@peterzhu.ca>
2023-12-07[PRISM] Account for multiple arguments when compiling ArgumentsNodeJemma Issroff
BreakNode, ReturnNode and NextNode all compile the ArgumentsNode directly, but we weren't accounting for multiple arguments. If there is more than one argument, we need to also emit a newarray instruction to put the arguments onto the stack
generated by cgit v1.2.3 (git 2.25.1) at 2026-01-19 21:32:25 +0000