From af1c4167287b9353fec766f932fe4afe97116ad4 Mon Sep 17 00:00:00 2001 From: naruse Date: Mon, 4 Jun 2007 12:31:26 +0000 Subject: * lib/json.rb, lib/json, ext/json, test/json: import JSON library. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@12428 b2dd03c8-39d4-4d8f-98ff-823fe69b080e --- ChangeLog | 5 + ext/json/ext/generator/extconf.h | 3 + ext/json/ext/generator/extconf.rb | 9 + ext/json/ext/generator/generator.c | 728 ++++++++++++++++ ext/json/ext/generator/unicode.c | 184 +++++ ext/json/ext/generator/unicode.h | 53 ++ ext/json/ext/parser/extconf.h | 3 + ext/json/ext/parser/extconf.rb | 9 + ext/json/ext/parser/parser.c | 1601 ++++++++++++++++++++++++++++++++++++ ext/json/ext/parser/parser.rl | 569 +++++++++++++ ext/json/ext/parser/unicode.c | 156 ++++ ext/json/ext/parser/unicode.h | 58 ++ ext/nkf/nkf-utf8/nkf.c | 198 +++-- ext/nkf/nkf-utf8/utf8tbl.c | 247 ++++++ ext/nkf/nkf-utf8/utf8tbl.h | 3 + lib/json.rb | 210 +++++ lib/json/Array.xpm | 21 + lib/json/FalseClass.xpm | 21 + lib/json/Hash.xpm | 21 + lib/json/Key.xpm | 73 ++ lib/json/NilClass.xpm | 21 + lib/json/Numeric.xpm | 28 + lib/json/String.xpm | 96 +++ lib/json/TrueClass.xpm | 21 + lib/json/common.rb | 194 +++++ lib/json/editor.rb | 1295 +++++++++++++++++++++++++++++ lib/json/ext.rb | 13 + lib/json/json.xpm | 1499 +++++++++++++++++++++++++++++++++ lib/json/pure.rb | 75 ++ lib/json/pure/generator.rb | 321 ++++++++ lib/json/pure/parser.rb | 241 ++++++ lib/json/version.rb | 9 + test/json/fixtures/fail1.json | 1 + test/json/fixtures/fail10.json | 1 + test/json/fixtures/fail11.json | 1 + test/json/fixtures/fail12.json | 1 + test/json/fixtures/fail13.json | 1 + test/json/fixtures/fail14.json | 1 + test/json/fixtures/fail18.json | 1 + test/json/fixtures/fail19.json | 1 + test/json/fixtures/fail2.json | 1 + test/json/fixtures/fail20.json | 1 + test/json/fixtures/fail21.json | 1 + test/json/fixtures/fail22.json | 1 + test/json/fixtures/fail23.json | 1 + test/json/fixtures/fail24.json | 1 + test/json/fixtures/fail25.json | 1 + test/json/fixtures/fail27.json | 2 + test/json/fixtures/fail28.json | 2 + test/json/fixtures/fail3.json | 1 + test/json/fixtures/fail4.json | 1 + test/json/fixtures/fail5.json | 1 + test/json/fixtures/fail6.json | 1 + test/json/fixtures/fail7.json | 1 + test/json/fixtures/fail8.json | 1 + test/json/fixtures/fail9.json | 1 + test/json/fixtures/pass1.json | 56 ++ test/json/fixtures/pass15.json | 1 + test/json/fixtures/pass16.json | 1 + test/json/fixtures/pass17.json | 1 + test/json/fixtures/pass2.json | 1 + test/json/fixtures/pass26.json | 1 + test/json/fixtures/pass3.json | 6 + test/json/runner.rb | 24 + test/json/test_json.rb | 255 ++++++ test/json/test_json_addition.rb | 94 +++ test/json/test_json_fixtures.rb | 30 + test/json/test_json_generate.rb | 81 ++ test/json/test_json_unicode.rb | 59 ++ 69 files changed, 8551 insertions(+), 70 deletions(-) create mode 100644 ext/json/ext/generator/extconf.h create mode 100644 ext/json/ext/generator/extconf.rb create mode 100755 ext/json/ext/generator/generator.c create mode 100644 ext/json/ext/generator/unicode.c create mode 100755 ext/json/ext/generator/unicode.h create mode 100644 ext/json/ext/parser/extconf.h create mode 100644 ext/json/ext/parser/extconf.rb create mode 100644 ext/json/ext/parser/parser.c create mode 100644 ext/json/ext/parser/parser.rl create mode 100644 ext/json/ext/parser/unicode.c create mode 100755 ext/json/ext/parser/unicode.h create mode 100644 lib/json.rb create mode 100644 lib/json/Array.xpm create mode 100644 lib/json/FalseClass.xpm create mode 100644 lib/json/Hash.xpm create mode 100644 lib/json/Key.xpm create mode 100644 lib/json/NilClass.xpm create mode 100644 lib/json/Numeric.xpm create mode 100644 lib/json/String.xpm create mode 100644 lib/json/TrueClass.xpm create mode 100644 lib/json/common.rb create mode 100644 lib/json/editor.rb create mode 100644 lib/json/ext.rb create mode 100644 lib/json/json.xpm create mode 100644 lib/json/pure.rb create mode 100644 lib/json/pure/generator.rb create mode 100644 lib/json/pure/parser.rb create mode 100755 lib/json/version.rb create mode 100644 test/json/fixtures/fail1.json create mode 100644 test/json/fixtures/fail10.json create mode 100644 test/json/fixtures/fail11.json create mode 100644 test/json/fixtures/fail12.json create mode 100644 test/json/fixtures/fail13.json create mode 100644 test/json/fixtures/fail14.json create mode 100644 test/json/fixtures/fail18.json create mode 100644 test/json/fixtures/fail19.json create mode 100644 test/json/fixtures/fail2.json create mode 100644 test/json/fixtures/fail20.json create mode 100644 test/json/fixtures/fail21.json create mode 100644 test/json/fixtures/fail22.json create mode 100644 test/json/fixtures/fail23.json create mode 100644 test/json/fixtures/fail24.json create mode 100644 test/json/fixtures/fail25.json create mode 100644 test/json/fixtures/fail27.json create mode 100644 test/json/fixtures/fail28.json create mode 100644 test/json/fixtures/fail3.json create mode 100644 test/json/fixtures/fail4.json create mode 100644 test/json/fixtures/fail5.json create mode 100644 test/json/fixtures/fail6.json create mode 100644 test/json/fixtures/fail7.json create mode 100644 test/json/fixtures/fail8.json create mode 100644 test/json/fixtures/fail9.json create mode 100644 test/json/fixtures/pass1.json create mode 100644 test/json/fixtures/pass15.json create mode 100644 test/json/fixtures/pass16.json create mode 100644 test/json/fixtures/pass17.json create mode 100644 test/json/fixtures/pass2.json create mode 100644 test/json/fixtures/pass26.json create mode 100644 test/json/fixtures/pass3.json create mode 100644 test/json/runner.rb create mode 100755 test/json/test_json.rb create mode 100755 test/json/test_json_addition.rb create mode 100755 test/json/test_json_fixtures.rb create mode 100755 test/json/test_json_generate.rb create mode 100755 test/json/test_json_unicode.rb diff --git a/ChangeLog b/ChangeLog index be313858fc..cd0ac15b09 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,8 @@ +Mon Jun 04 21:15:45 2007 NARUSE, Yui + + * lib/json.rb, lib/json, ext/json, test/json: + import JSON library. + Sat Jun 2 16:48:55 2007 Koichi Sasada * cont.c (Fiber#pass): rename to Fiber#yield. Block parameter diff --git a/ext/json/ext/generator/extconf.h b/ext/json/ext/generator/extconf.h new file mode 100644 index 0000000000..cda0cc8ea5 --- /dev/null +++ b/ext/json/ext/generator/extconf.h @@ -0,0 +1,3 @@ +#ifndef EXTCONF_H +#define EXTCONF_H +#endif diff --git a/ext/json/ext/generator/extconf.rb b/ext/json/ext/generator/extconf.rb new file mode 100644 index 0000000000..db721a92f8 --- /dev/null +++ b/ext/json/ext/generator/extconf.rb @@ -0,0 +1,9 @@ +require 'mkmf' +require 'rbconfig' + +if CONFIG['CC'] =~ /gcc/ + CONFIG['CC'] += ' -Wall -ggdb' + #CONFIG['CC'] += ' -Wall' +end + +create_makefile 'json/ext/generator' diff --git a/ext/json/ext/generator/generator.c b/ext/json/ext/generator/generator.c new file mode 100755 index 0000000000..60e09355f0 --- /dev/null +++ b/ext/json/ext/generator/generator.c @@ -0,0 +1,728 @@ +/* vim: set cin et sw=4 ts=4: */ + +#include +#include "ruby.h" +#include "st.h" +#include "unicode.h" + +static VALUE mJSON, mExt, mGenerator, cState, mGeneratorMethods, mObject, + mHash, mArray, mInteger, mFloat, mString, mString_Extend, + mTrueClass, mFalseClass, mNilClass, eGeneratorError, + eCircularDatastructure; + +static ID i_to_s, i_to_json, i_new, i_indent, i_space, i_space_before, + i_object_nl, i_array_nl, i_check_circular, i_pack, i_unpack, + i_create_id, i_extend; + +typedef struct JSON_Generator_StateStruct { + VALUE indent; + VALUE space; + VALUE space_before; + VALUE object_nl; + VALUE array_nl; + int check_circular; + VALUE seen; + VALUE memo; + VALUE depth; + int flag; +} JSON_Generator_State; + +#define GET_STATE(self) \ + JSON_Generator_State *state; \ + Data_Get_Struct(self, JSON_Generator_State, state); + +/* + * Document-module: JSON::Ext::Generator + * + * This is the JSON generator implemented as a C extension. It can be + * configured to be used by setting + * + * JSON.generator = JSON::Ext::Generator + * + * with the method generator= in JSON. + * + */ + +static int hash_to_json_state_i(VALUE key, VALUE value, VALUE Vstate) +{ + VALUE json, buf, Vdepth; + GET_STATE(Vstate); + buf = state->memo; + Vdepth = state->depth; + + if (key == Qundef) return ST_CONTINUE; + if (state->flag) { + state->flag = 0; + rb_str_buf_cat2(buf, ","); + if (RSTRING_LEN(state->object_nl)) rb_str_buf_append(buf, state->object_nl); + } + if (RSTRING_LEN(state->object_nl)) { + rb_str_buf_append(buf, rb_str_times(state->indent, Vdepth)); + } + json = rb_funcall(rb_funcall(key, i_to_s, 0), i_to_json, 2, Vstate, Vdepth); + rb_str_buf_append(buf, json); + OBJ_INFECT(buf, json); + if (RSTRING_LEN(state->space_before)) { + rb_str_buf_append(buf, state->space_before); + } + rb_str_buf_cat2(buf, ":"); + if (RSTRING_LEN(state->space)) rb_str_buf_append(buf, state->space); + json = rb_funcall(value, i_to_json, 2, Vstate, Vdepth); + state->flag = 1; + rb_str_buf_append(buf, json); + OBJ_INFECT(buf, json); + state->depth = Vdepth; + state->memo = buf; + return ST_CONTINUE; +} + +inline static VALUE mHash_json_transfrom(VALUE self, VALUE Vstate, VALUE Vdepth) { + long depth, len = RHASH(self)->tbl->num_entries; + VALUE result; + GET_STATE(Vstate); + + depth = 1 + FIX2LONG(Vdepth); + result = rb_str_buf_new(len); + state->memo = result; + state->depth = LONG2FIX(depth); + state->flag = 0; + rb_str_buf_cat2(result, "{"); + if (RSTRING_LEN(state->object_nl)) rb_str_buf_append(result, state->object_nl); + rb_hash_foreach(self, hash_to_json_state_i, Vstate); + if (RSTRING_LEN(state->object_nl)) rb_str_buf_append(result, state->object_nl); + if (RSTRING_LEN(state->object_nl)) { + rb_str_buf_append(result, rb_str_times(state->indent, Vdepth)); + } + rb_str_buf_cat2(result, "}"); + return result; +} + +static int hash_to_json_i(VALUE key, VALUE value, VALUE buf) +{ + VALUE tmp; + + if (key == Qundef) return ST_CONTINUE; + if (RSTRING_LEN(buf) > 1) rb_str_buf_cat2(buf, ","); + tmp = rb_funcall(rb_funcall(key, i_to_s, 0), i_to_json, 0); + rb_str_buf_append(buf, tmp); + OBJ_INFECT(buf, tmp); + rb_str_buf_cat2(buf, ":"); + tmp = rb_funcall(value, i_to_json, 0); + rb_str_buf_append(buf, tmp); + OBJ_INFECT(buf, tmp); + + return ST_CONTINUE; +} + +/* + * call-seq: to_json(state = nil, depth = 0) + * + * Returns a JSON string containing a JSON object, that is unparsed from + * this Hash instance. + * _state_ is a JSON::State object, that can also be used to configure the + * produced JSON string output further. + * _depth_ is used to find out nesting depth, to indent accordingly. + */ +static VALUE mHash_to_json(int argc, VALUE *argv, VALUE self) +{ + VALUE Vstate, Vdepth, result; + long depth; + + rb_scan_args(argc, argv, "02", &Vstate, &Vdepth); + depth = NIL_P(Vdepth) ? 0 : FIX2LONG(Vdepth); + if (NIL_P(Vstate)) { + long len = RHASH(self)->tbl->num_entries; + result = rb_str_buf_new(len); + rb_str_buf_cat2(result, "{"); + rb_hash_foreach(self, hash_to_json_i, result); + rb_str_buf_cat2(result, "}"); + } else { + GET_STATE(Vstate); + if (state->check_circular) { + VALUE self_id = rb_obj_id(self); + if (RTEST(rb_hash_aref(state->seen, self_id))) { + rb_raise(eCircularDatastructure, + "circular data structures not supported!"); + } + rb_hash_aset(state->seen, self_id, Qtrue); + result = mHash_json_transfrom(self, Vstate, LONG2FIX(depth)); + rb_hash_delete(state->seen, self_id); + } else { + result = mHash_json_transfrom(self, Vstate, LONG2FIX(depth)); + } + } + OBJ_INFECT(result, self); + return result; +} + +inline static VALUE mArray_json_transfrom(VALUE self, VALUE Vstate, VALUE Vdepth) { + long i, len = RARRAY_LEN(self); + VALUE shift, result; + long depth = NIL_P(Vdepth) ? 0 : FIX2LONG(Vdepth); + VALUE delim = rb_str_new2(","); + GET_STATE(Vstate); + + if (state->check_circular) { + VALUE self_id = rb_obj_id(self); + rb_hash_aset(state->seen, self_id, Qtrue); + result = rb_str_buf_new(len); + if (RSTRING_LEN(state->array_nl)) rb_str_append(delim, state->array_nl); + shift = rb_str_times(state->indent, LONG2FIX(depth + 1)); + + rb_str_buf_cat2(result, "["); + rb_str_buf_append(result, state->array_nl); + for (i = 0; i < len; i++) { + VALUE element = RARRAY_PTR(self)[i]; + if (RTEST(rb_hash_aref(state->seen, rb_obj_id(element)))) { + rb_raise(eCircularDatastructure, + "circular data structures not supported!"); + } + OBJ_INFECT(result, element); + if (i > 0) rb_str_buf_append(result, delim); + rb_str_buf_append(result, shift); + rb_str_buf_append(result, rb_funcall(element, i_to_json, 2, Vstate, LONG2FIX(depth + 1))); + } + if (RSTRING_LEN(state->array_nl)) { + rb_str_buf_append(result, state->array_nl); + rb_str_buf_append(result, rb_str_times(state->indent, LONG2FIX(depth))); + } + rb_str_buf_cat2(result, "]"); + rb_hash_delete(state->seen, self_id); + } else { + result = rb_str_buf_new(len); + if (RSTRING_LEN(state->array_nl)) rb_str_append(delim, state->array_nl); + shift = rb_str_times(state->indent, LONG2FIX(depth + 1)); + + rb_str_buf_cat2(result, "["); + rb_str_buf_append(result, state->array_nl); + for (i = 0; i < len; i++) { + VALUE element = RARRAY_PTR(self)[i]; + OBJ_INFECT(result, element); + if (i > 0) rb_str_buf_append(result, delim); + rb_str_buf_append(result, shift); + rb_str_buf_append(result, rb_funcall(element, i_to_json, 2, Vstate, LONG2FIX(depth + 1))); + } + rb_str_buf_append(result, state->array_nl); + if (RSTRING_LEN(state->array_nl)) { + rb_str_buf_append(result, rb_str_times(state->indent, LONG2FIX(depth))); + } + rb_str_buf_cat2(result, "]"); + } + return result; +} + +/* + * call-seq: to_json(state = nil, depth = 0) + * + * Returns a JSON string containing a JSON array, that is unparsed from + * this Array instance. + * _state_ is a JSON::State object, that can also be used to configure the + * produced JSON string output further. + * _depth_ is used to find out nesting depth, to indent accordingly. + */ +static VALUE mArray_to_json(int argc, VALUE *argv, VALUE self) { + VALUE Vstate, Vdepth, result; + + rb_scan_args(argc, argv, "02", &Vstate, &Vdepth); + if (NIL_P(Vstate)) { + long i, len = RARRAY_LEN(self); + result = rb_str_buf_new(2 + 2 * len); + rb_str_buf_cat2(result, "["); + for (i = 0; i < len; i++) { + VALUE element = RARRAY_PTR(self)[i]; + OBJ_INFECT(result, element); + if (i > 0) rb_str_buf_cat2(result, ","); + rb_str_buf_append(result, rb_funcall(element, i_to_json, 0)); + } + rb_str_buf_cat2(result, "]"); + } else { + result = mArray_json_transfrom(self, Vstate, Vdepth); + } + OBJ_INFECT(result, self); + return result; +} + +/* + * call-seq: to_json(*) + * + * Returns a JSON string representation for this Integer number. + */ +static VALUE mInteger_to_json(int argc, VALUE *argv, VALUE self) +{ + return rb_funcall(self, i_to_s, 0); +} + +/* + * call-seq: to_json(*) + * + * Returns a JSON string representation for this Float number. + */ +static VALUE mFloat_to_json(int argc, VALUE *argv, VALUE self) +{ + return rb_funcall(self, i_to_s, 0); +} + +/* + * call-seq: String.included(modul) + * + * Extends _modul_ with the String::Extend module. + */ +static VALUE mString_included_s(VALUE self, VALUE modul) { + return rb_funcall(modul, i_extend, 1, mString_Extend); +} + +/* + * call-seq: to_json(*) + * + * This string should be encoded with UTF-8 A call to this method + * returns a JSON string encoded with UTF16 big endian characters as + * \u????. + */ +static VALUE mString_to_json(int argc, VALUE *argv, VALUE self) +{ + VALUE result = rb_str_buf_new(RSTRING_LEN(self)); + rb_str_buf_cat2(result, "\""); + JSON_convert_UTF8_to_JSON(result, self, strictConversion); + rb_str_buf_cat2(result, "\""); + return result; +} + +/* + * call-seq: to_json_raw_object() + * + * This method creates a raw object hash, that can be nested into + * other data structures and will be unparsed as a raw string. This + * method should be used, if you want to convert raw strings to JSON + * instead of UTF-8 strings, e. g. binary data. + */ +static VALUE mString_to_json_raw_object(VALUE self) { + VALUE ary; + VALUE result = rb_hash_new(); + rb_hash_aset(result, rb_funcall(mJSON, i_create_id, 0), rb_class_name(rb_obj_class(self))); + ary = rb_funcall(self, i_unpack, 1, rb_str_new2("C*")); + rb_hash_aset(result, rb_str_new2("raw"), ary); + return result; +} + +/* + * call-seq: to_json_raw(*args) + * + * This method creates a JSON text from the result of a call to + * to_json_raw_object of this String. + */ +static VALUE mString_to_json_raw(int argc, VALUE *argv, VALUE self) { + VALUE obj = mString_to_json_raw_object(self); + Check_Type(obj, T_HASH); + return mHash_to_json(argc, argv, obj); +} + +/* + * call-seq: json_create(o) + * + * Raw Strings are JSON Objects (the raw bytes are stored in an array for the + * key "raw"). The Ruby String can be created by this module method. + */ +static VALUE mString_Extend_json_create(VALUE self, VALUE o) { + VALUE ary; + Check_Type(o, T_HASH); + ary = rb_hash_aref(o, rb_str_new2("raw")); + return rb_funcall(ary, i_pack, 1, rb_str_new2("C*")); +} + +/* + * call-seq: to_json(state = nil, depth = 0) + * + * Returns a JSON string for true: 'true'. + */ +static VALUE mTrueClass_to_json(int argc, VALUE *argv, VALUE self) +{ + return rb_str_new2("true"); +} + +/* + * call-seq: to_json(state = nil, depth = 0) + * + * Returns a JSON string for false: 'false'. + */ +static VALUE mFalseClass_to_json(int argc, VALUE *argv, VALUE self) +{ + return rb_str_new2("false"); +} + +/* + * call-seq: to_json(state = nil, depth = 0) + * + */ +static VALUE mNilClass_to_json(int argc, VALUE *argv, VALUE self) +{ + return rb_str_new2("null"); +} + +/* + * call-seq: to_json(*) + * + * Converts this object to a string (calling #to_s), converts + * it to a JSON string, and returns the result. This is a fallback, if no + * special method #to_json was defined for some object. + */ +static VALUE mObject_to_json(int argc, VALUE *argv, VALUE self) +{ + VALUE string = rb_funcall(self, i_to_s, 0); + Check_Type(string, T_STRING); + return mString_to_json(argc, argv, string); +} + +/* + * Document-class: JSON::Ext::Generator::State + * + * This class is used to create State instances, that are use to hold data + * while generating a JSON text from a a Ruby data structure. + */ + +static void State_mark(JSON_Generator_State *state) +{ + rb_gc_mark_maybe(state->indent); + rb_gc_mark_maybe(state->space); + rb_gc_mark_maybe(state->space_before); + rb_gc_mark_maybe(state->object_nl); + rb_gc_mark_maybe(state->array_nl); + rb_gc_mark_maybe(state->seen); + rb_gc_mark_maybe(state->memo); + rb_gc_mark_maybe(state->depth); +} + +static JSON_Generator_State *State_allocate() +{ + JSON_Generator_State *state = ALLOC(JSON_Generator_State); + return state; +} + +static VALUE cState_s_allocate(VALUE klass) +{ + JSON_Generator_State *state = State_allocate(); + return Data_Wrap_Struct(klass, State_mark, -1, state); +} + +/* + * call-seq: new(opts = {}) + * + * Instantiates a new State object, configured by _opts_. + * + * _opts_ can have the following keys: + * + * * *indent*: a string used to indent levels (default: ''), + * * *space*: a string that is put after, a : or , delimiter (default: ''), + * * *space_before*: a string that is put before a : pair delimiter (default: ''), + * * *object_nl*: a string that is put at the end of a JSON object (default: ''), + * * *array_nl*: a string that is put at the end of a JSON array (default: ''), + * * *check_circular*: true if checking for circular data structures + * should be done, false (the default) otherwise. + */ +static VALUE cState_initialize(int argc, VALUE *argv, VALUE self) +{ + VALUE opts; + GET_STATE(self); + + rb_scan_args(argc, argv, "01", &opts); + if (NIL_P(opts)) { + state->indent = rb_str_new2(""); + state->space = rb_str_new2(""); + state->space_before = rb_str_new2(""); + state->array_nl = rb_str_new2(""); + state->object_nl = rb_str_new2(""); + state->check_circular = 0; + } else { + VALUE tmp; + opts = rb_convert_type(opts, T_HASH, "Hash", "to_hash"); + tmp = rb_hash_aref(opts, ID2SYM(i_indent)); + if (RTEST(tmp)) { + Check_Type(tmp, T_STRING); + state->indent = tmp; + } else { + state->indent = rb_str_new2(""); + } + tmp = rb_hash_aref(opts, ID2SYM(i_space)); + if (RTEST(tmp)) { + Check_Type(tmp, T_STRING); + state->space = tmp; + } else { + state->space = rb_str_new2(""); + } + tmp = rb_hash_aref(opts, ID2SYM(i_space_before)); + if (RTEST(tmp)) { + Check_Type(tmp, T_STRING); + state->space_before = tmp; + } else { + state->space_before = rb_str_new2(""); + } + tmp = rb_hash_aref(opts, ID2SYM(i_array_nl)); + if (RTEST(tmp)) { + Check_Type(tmp, T_STRING); + state->array_nl = tmp; + } else { + state->array_nl = rb_str_new2(""); + } + tmp = rb_hash_aref(opts, ID2SYM(i_object_nl)); + if (RTEST(tmp)) { + Check_Type(tmp, T_STRING); + state->object_nl = tmp; + } else { + state->object_nl = rb_str_new2(""); + } + tmp = rb_hash_aref(opts, ID2SYM(i_check_circular)); + state->check_circular = RTEST(tmp); + } + state->seen = rb_hash_new(); + state->memo = Qnil; + state->depth = INT2FIX(0); + return self; +} + +/* + * call-seq: from_state(opts) + * + * Creates a State object from _opts_, which ought to be Hash to create a + * new State instance configured by _opts_, something else to create an + * unconfigured instance. If _opts_ is a State object, it is just returned. + */ +static VALUE cState_from_state_s(VALUE self, VALUE opts) +{ + if (rb_obj_is_kind_of(opts, self)) { + return opts; + } else if (rb_obj_is_kind_of(opts, rb_cHash)) { + return rb_funcall(self, i_new, 1, opts); + } else { + return rb_funcall(self, i_new, 0); + } +} + +/* + * call-seq: indent() + * + * This string is used to indent levels in the JSON text. + */ +static VALUE cState_indent(VALUE self) +{ + GET_STATE(self); + return state->indent; +} + +/* + * call-seq: indent=(indent) + * + * This string is used to indent levels in the JSON text. + */ +static VALUE cState_indent_set(VALUE self, VALUE indent) +{ + GET_STATE(self); + Check_Type(indent, T_STRING); + return state->indent = indent; +} + +/* + * call-seq: space() + * + * This string is used to insert a space between the tokens in a JSON + * string. + */ +static VALUE cState_space(VALUE self) +{ + GET_STATE(self); + return state->space; +} + +/* + * call-seq: space=(space) + * + * This string is used to insert a space between the tokens in a JSON + * string. + */ +static VALUE cState_space_set(VALUE self, VALUE space) +{ + GET_STATE(self); + Check_Type(space, T_STRING); + return state->space = space; +} + +/* + * call-seq: space_before() + * + * This string is used to insert a space before the ':' in JSON objects. + */ +static VALUE cState_space_before(VALUE self) +{ + GET_STATE(self); + return state->space_before; +} + +/* + * call-seq: space_before=(space_before) + * + * This string is used to insert a space before the ':' in JSON objects. + */ +static VALUE cState_space_before_set(VALUE self, VALUE space_before) +{ + GET_STATE(self); + Check_Type(space_before, T_STRING); + return state->space_before = space_before; +} + +/* + * call-seq: object_nl() + * + * This string is put at the end of a line that holds a JSON object (or + * Hash). + */ +static VALUE cState_object_nl(VALUE self) +{ + GET_STATE(self); + return state->object_nl; +} + +/* + * call-seq: object_nl=(object_nl) + * + * This string is put at the end of a line that holds a JSON object (or + * Hash). + */ +static VALUE cState_object_nl_set(VALUE self, VALUE object_nl) +{ + GET_STATE(self); + Check_Type(object_nl, T_STRING); + return state->object_nl = object_nl; +} + +/* + * call-seq: array_nl() + * + * This string is put at the end of a line that holds a JSON array. + */ +static VALUE cState_array_nl(VALUE self) +{ + GET_STATE(self); + return state->array_nl; +} + +/* + * call-seq: array_nl=(array_nl) + * + * This string is put at the end of a line that holds a JSON array. + */ +static VALUE cState_array_nl_set(VALUE self, VALUE array_nl) +{ + GET_STATE(self); + Check_Type(array_nl, T_STRING); + return state->array_nl = array_nl; +} + +/* + * call-seq: check_circular?(object) + * + * Returns true, if circular data structures should be checked, + * otherwise returns false. + */ +static VALUE cState_check_circular_p(VALUE self) +{ + GET_STATE(self); + return state->check_circular ? Qtrue : Qfalse; +} + +/* + * call-seq: seen?(object) + * + * Returns _true_, if _object_ was already seen during this generating run. + */ +static VALUE cState_seen_p(VALUE self, VALUE object) +{ + GET_STATE(self); + return rb_hash_aref(state->seen, rb_obj_id(object)); +} + +/* + * call-seq: remember(object) + * + * Remember _object_, to find out if it was already encountered (if a cyclic + * data structure is rendered). + */ +static VALUE cState_remember(VALUE self, VALUE object) +{ + GET_STATE(self); + return rb_hash_aset(state->seen, rb_obj_id(object), Qtrue); +} + +/* + * call-seq: forget(object) + * + * Forget _object_ for this generating run. + */ +static VALUE cState_forget(VALUE self, VALUE object) +{ + GET_STATE(self); + return rb_hash_delete(state->seen, rb_obj_id(object)); +} + +void Init_generator() +{ + mJSON = rb_define_module("JSON"); + mExt = rb_define_module_under(mJSON, "Ext"); + mGenerator = rb_define_module_under(mExt, "Generator"); + eGeneratorError = rb_path2class("JSON::GeneratorError"); + eCircularDatastructure = rb_path2class("JSON::CircularDatastructure"); + cState = rb_define_class_under(mGenerator, "State", rb_cObject); + rb_define_alloc_func(cState, cState_s_allocate); + rb_define_singleton_method(cState, "from_state", cState_from_state_s, 1); + rb_define_method(cState, "initialize", cState_initialize, -1); + + rb_define_method(cState, "indent", cState_indent, 0); + rb_define_method(cState, "indent=", cState_indent_set, 1); + rb_define_method(cState, "space", cState_space, 0); + rb_define_method(cState, "space=", cState_space_set, 1); + rb_define_method(cState, "space_before", cState_space_before, 0); + rb_define_method(cState, "space_before=", cState_space_before_set, 1); + rb_define_method(cState, "object_nl", cState_object_nl, 0); + rb_define_method(cState, "object_nl=", cState_object_nl_set, 1); + rb_define_method(cState, "array_nl", cState_array_nl, 0); + rb_define_method(cState, "array_nl=", cState_array_nl_set, 1); + rb_define_method(cState, "check_circular?", cState_check_circular_p, 0); + rb_define_method(cState, "seen?", cState_seen_p, 1); + rb_define_method(cState, "remember", cState_remember, 1); + rb_define_method(cState, "forget", cState_forget, 1); + mGeneratorMethods = rb_define_module_under(mGenerator, "GeneratorMethods"); + mObject = rb_define_module_under(mGeneratorMethods, "Object"); + rb_define_method(mObject, "to_json", mObject_to_json, -1); + mHash = rb_define_module_under(mGeneratorMethods, "Hash"); + rb_define_method(mHash, "to_json", mHash_to_json, -1); + mArray = rb_define_module_under(mGeneratorMethods, "Array"); + rb_define_method(mArray, "to_json", mArray_to_json, -1); + mInteger = rb_define_module_under(mGeneratorMethods, "Integer"); + rb_define_method(mInteger, "to_json", mInteger_to_json, -1); + mFloat = rb_define_module_under(mGeneratorMethods, "Float"); + rb_define_method(mFloat, "to_json", mFloat_to_json, -1); + mString = rb_define_module_under(mGeneratorMethods, "String"); + rb_define_singleton_method(mString, "included", mString_included_s, 1); + rb_define_method(mString, "to_json", mString_to_json, -1); + rb_define_method(mString, "to_json_raw", mString_to_json_raw, -1); + rb_define_method(mString, "to_json_raw_object", mString_to_json_raw_object, 0); + mString_Extend = rb_define_module_under(mString, "Extend"); + rb_define_method(mString_Extend, "json_create", mString_Extend_json_create, 1); + mTrueClass = rb_define_module_under(mGeneratorMethods, "TrueClass"); + rb_define_method(mTrueClass, "to_json", mTrueClass_to_json, -1); + mFalseClass = rb_define_module_under(mGeneratorMethods, "FalseClass"); + rb_define_method(mFalseClass, "to_json", mFalseClass_to_json, -1); + mNilClass = rb_define_module_under(mGeneratorMethods, "NilClass"); + rb_define_method(mNilClass, "to_json", mNilClass_to_json, -1); + + i_to_s = rb_intern("to_s"); + i_to_json = rb_intern("to_json"); + i_new = rb_intern("new"); + i_indent = rb_intern("indent"); + i_space = rb_intern("space"); + i_space_before = rb_intern("space_before"); + i_object_nl = rb_intern("object_nl"); + i_array_nl = rb_intern("array_nl"); + i_check_circular = rb_intern("check_circular"); + i_pack = rb_intern("pack"); + i_unpack = rb_intern("unpack"); + i_create_id = rb_intern("create_id"); + i_extend = rb_intern("extend"); +} diff --git a/ext/json/ext/generator/unicode.c b/ext/json/ext/generator/unicode.c new file mode 100644 index 0000000000..44e1f41f98 --- /dev/null +++ b/ext/json/ext/generator/unicode.c @@ -0,0 +1,184 @@ +/* vim: set cin et sw=4 ts=4: */ + +#include "unicode.h" + +#define unicode_escape(buffer, character) \ + snprintf(buf, 7, "\\u%04x", (unsigned int) (character)); \ + rb_str_buf_cat(buffer, buf, 6); + +/* + * Copyright 2001-2004 Unicode, Inc. + * + * 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. + * + * 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. + */ + +/* + * Index into the table below with the first byte of a UTF-8 sequence to + * get the number of trailing bytes that are supposed to follow it. + * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is + * left as-is for anyone who may want to do such conversion, which was + * allowed in earlier algorithms. + */ +static const char trailingBytesForUTF8[256] = { + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 +}; + +/* + * Magic values subtracted from a buffer value during UTF8 conversion. + * This table contains as many values as there might be trailing bytes + * in a UTF-8 sequence. + */ +static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, + 0x03C82080UL, 0xFA082080UL, 0x82082080UL }; + +/* + * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed + * into the first byte, depending on how many bytes follow. There are + * as many entries in this table as there are UTF-8 sequence types. + * (I.e., one byte sequence, two byte... etc.). Remember that sequencs + * for *legal* UTF-8 will be 4 or fewer bytes total. + */ +static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC }; + +/* + * Utility routine to tell whether a sequence of bytes is legal UTF-8. + * This must be called with the length pre-determined by the first byte. + * If not calling this from ConvertUTF8to*, then the length can be set by: + * length = trailingBytesForUTF8[*source]+1; + * and the sequence is illegal right away if there aren't that many bytes + * available. + * If presented with a length > 4, this returns 0. The Unicode + * definition of UTF-8 goes up to 4-byte sequences. + */ + +inline static unsigned char isLegalUTF8(const UTF8 *source, int length) +{ + UTF8 a; + const UTF8 *srcptr = source+length; + switch (length) { + default: return 0; + /* Everything else falls through when "1"... */ + case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return 0; + case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return 0; + case 2: if ((a = (*--srcptr)) > 0xBF) return 0; + + switch (*source) { + /* no fall-through in this inner switch */ + case 0xE0: if (a < 0xA0) return 0; break; + case 0xED: if (a > 0x9F) return 0; break; + case 0xF0: if (a < 0x90) return 0; break; + case 0xF4: if (a > 0x8F) return 0; break; + default: if (a < 0x80) return 0; + } + + case 1: if (*source >= 0x80 && *source < 0xC2) return 0; + } + if (*source > 0xF4) return 0; + return 1; +} + +void JSON_convert_UTF8_to_JSON(VALUE buffer, VALUE string, ConversionFlags flags) +{ + char buf[7]; + const UTF8* source = (UTF8 *) RSTRING_PTR(string); + const UTF8* sourceEnd = source + RSTRING_LEN(string); + + while (source < sourceEnd) { + UTF32 ch = 0; + unsigned short extraBytesToRead = trailingBytesForUTF8[*source]; + if (source + extraBytesToRead >= sourceEnd) { + rb_raise(rb_path2class("JSON::GeneratorError"), + "partial character in source, but hit end"); + } + if (!isLegalUTF8(source, extraBytesToRead+1)) { + rb_raise(rb_path2class("JSON::GeneratorError"), + "source sequence is illegal/malformed"); + } + /* + * The cases all fall through. See "Note A" below. + */ + switch (extraBytesToRead) { + case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */ + case 3: ch += *source++; ch <<= 6; + case 2: ch += *source++; ch <<= 6; + case 1: ch += *source++; ch <<= 6; + case 0: ch += *source++; + } + ch -= offsetsFromUTF8[extraBytesToRead]; + + if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */ + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the illegal value itself */ + rb_raise(rb_path2class("JSON::GeneratorError"), + "source sequence is illegal/malformed"); + } else { + unicode_escape(buffer, UNI_REPLACEMENT_CHAR); + } + } else { + /* normal case */ + if (ch == '"') { + rb_str_buf_cat2(buffer, "\\\""); + } else if (ch == '\\') { + rb_str_buf_cat2(buffer, "\\\\"); + } else if (ch == '/') { + rb_str_buf_cat2(buffer, "\\/"); + } else if (ch >= 0x20 && ch <= 0x7f) { + rb_str_buf_cat(buffer, (char *) source - 1, 1); + } else if (ch == '\n') { + rb_str_buf_cat2(buffer, "\\n"); + } else if (ch == '\r') { + rb_str_buf_cat2(buffer, "\\r"); + } else if (ch == '\t') { + rb_str_buf_cat2(buffer, "\\t"); + } else if (ch == '\f') { + rb_str_buf_cat2(buffer, "\\f"); + } else if (ch == '\b') { + rb_str_buf_cat2(buffer, "\\b"); + } else if (ch < 0x20) { + unicode_escape(buffer, (UTF16) ch); + } else { + unicode_escape(buffer, (UTF16) ch); + } + } + } else if (ch > UNI_MAX_UTF16) { + if (flags == strictConversion) { + source -= (extraBytesToRead+1); /* return to the start */ + rb_raise(rb_path2class("JSON::GeneratorError"), + "source sequence is illegal/malformed"); + } else { + unicode_escape(buffer, UNI_REPLACEMENT_CHAR); + } + } else { + /* target is a character in range 0xFFFF - 0x10FFFF. */ + ch -= halfBase; + unicode_escape(buffer, (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START)); + unicode_escape(buffer, (UTF16)((ch & halfMask) + UNI_SUR_LOW_START)); + } + } +} diff --git a/ext/json/ext/generator/unicode.h b/ext/json/ext/generator/unicode.h new file mode 100755 index 0000000000..841474bcea --- /dev/null +++ b/ext/json/ext/generator/unicode.h @@ -0,0 +1,53 @@ +#include "ruby.h" + +#ifndef _GENERATOR_UNICODE_H_ +#define _GENERATOR_UNICODE_H_ + +typedef enum { + conversionOK = 0, /* conversion successful */ + sourceExhausted, /* partial character in source, but hit end */ + targetExhausted, /* insuff. room in target for conversion */ + sourceIllegal /* source sequence is illegal/malformed */ +} ConversionResult; + +typedef enum { + strictConversion = 0, + lenientConversion +} ConversionFlags; + +typedef unsigned long UTF32; /* at least 32 bits */ +typedef unsigned short UTF16; /* at least 16 bits */ +typedef unsigned char UTF8; /* typically 8 bits */ + +#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD +#define UNI_MAX_BMP (UTF32)0x0000FFFF +#define UNI_MAX_UTF16 (UTF32)0x0010FFFF +#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF +#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF + +#define UNI_SUR_HIGH_START (UTF32)0xD800 +#define UNI_SUR_HIGH_END (UTF32)0xDBFF +#define UNI_SUR_LOW_START (UTF32)0xDC00 +#define UNI_SUR_LOW_END (UTF32)0xDFFF + +static const int halfShift = 10; /* used for shifting by 10 bits */ + +static const UTF32 halfBase = 0x0010000UL; +static const UTF32 halfMask = 0x3FFUL; + +void JSON_convert_UTF8_to_JSON(VALUE buffer, VALUE string, ConversionFlags flags); + +#ifndef RARRAY_PTR +#define RARRAY_PTR(ARRAY) RARRAY(ARRAY)->ptr +#endif +#ifndef RARRAY_LEN +#define RARRAY_LEN(ARRAY) RARRAY(ARRAY)->len +#endif +#ifndef RSTRING_PTR +#define RSTRING_PTR(string) RSTRING(string)->ptr +#endif +#ifndef RSTRING_LEN +#define RSTRING_LEN(string) RSTRING(string)->len +#endif + +#endif diff --git a/ext/json/ext/parser/extconf.h b/ext/json/ext/parser/extconf.h new file mode 100644 index 0000000000..cda0cc8ea5 --- /dev/null +++ b/ext/json/ext/parser/extconf.h @@ -0,0 +1,3 @@ +#ifndef EXTCONF_H +#define EXTCONF_H +#endif diff --git a/ext/json/ext/parser/extconf.rb b/ext/json/ext/parser/extconf.rb new file mode 100644 index 0000000000..085c8d060c --- /dev/null +++ b/ext/json/ext/parser/extconf.rb @@ -0,0 +1,9 @@ +require 'mkmf' +require 'rbconfig' + +if CONFIG['CC'] =~ /gcc/ + #CONFIG['CC'] += ' -Wall -ggdb' + CONFIG['CC'] += ' -Wall' +end + +create_makefile 'json/ext/parser' diff --git a/ext/json/ext/parser/parser.c b/ext/json/ext/parser/parser.c new file mode 100644 index 0000000000..7448e5fb7a --- /dev/null +++ b/ext/json/ext/parser/parser.c @@ -0,0 +1,1601 @@ +#line 1 "parser.rl" +/* vim: set cin et sw=4 ts=4: */ + +#include "ruby.h" +#include "re.h" +#include "st.h" +#include "unicode.h" + +#define EVIL 0x666 + +static VALUE mJSON, mExt, cParser, eParserError, eNestingError; + +static ID i_json_creatable_p, i_json_create, i_create_id, i_chr, i_max_nesting; + +typedef struct JSON_ParserStruct { + VALUE Vsource; + char *source; + long len; + char *memo; + VALUE create_id; + int max_nesting; + int current_nesting; +} JSON_Parser; + +static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_string(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result); + +#define GET_STRUCT \ + JSON_Parser *json; \ + Data_Get_Struct(self, JSON_Parser, json); + +#line 58 "parser.rl" + + + +#line 41 "parser.c" +static const int JSON_object_start = 1; +static const int JSON_object_first_final = 27; +static const int JSON_object_error = 0; + +static const int JSON_object_en_main = 1; + +#line 91 "parser.rl" + + +static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + VALUE last_name = Qnil; + + if (json->max_nesting && json->current_nesting > json->max_nesting) { + rb_raise(eNestingError, "nesting of %d is to deep", json->current_nesting); + } + + *result = rb_hash_new(); + + +#line 63 "parser.c" + { + cs = JSON_object_start; + } +#line 105 "parser.rl" + +#line 69 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + if ( (*p) == 123 ) + goto st2; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + switch( (*p) ) { + case 13: goto st2; + case 32: goto st2; + case 34: goto tr2; + case 47: goto st23; + case 125: goto tr4; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st2; + goto st0; +tr2: +#line 77 "parser.rl" + { + char *np = JSON_parse_string(json, p, pe, &last_name); + if (np == NULL) goto _out3; else {p = (( np))-1;} + } + goto st3; +st3: + if ( ++p == pe ) + goto _out3; +case 3: +#line 106 "parser.c" + switch( (*p) ) { + case 13: goto st3; + case 32: goto st3; + case 47: goto st4; + case 58: goto st8; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st3; + goto st0; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + switch( (*p) ) { + case 42: goto st5; + case 47: goto st7; + } + goto st0; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + if ( (*p) == 42 ) + goto st6; + goto st5; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + switch( (*p) ) { + case 42: goto st6; + case 47: goto st3; + } + goto st5; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + if ( (*p) == 10 ) + goto st3; + goto st7; +st8: + if ( ++p == pe ) + goto _out8; +case 8: + switch( (*p) ) { + case 13: goto st8; + case 32: goto st8; + case 34: goto tr11; + case 45: goto tr11; + case 47: goto st19; + case 91: goto tr11; + case 102: goto tr11; + case 110: goto tr11; + case 116: goto tr11; + case 123: goto tr11; + } + if ( (*p) > 10 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto tr11; + } else if ( (*p) >= 9 ) + goto st8; + goto st0; +tr11: +#line 66 "parser.rl" + { + VALUE v = Qnil; + char *np = JSON_parse_value(json, p, pe, &v); + if (np == NULL) { + goto _out9; + } else { + rb_hash_aset(*result, last_name, v); + {p = (( np))-1;} + } + } + goto st9; +st9: + if ( ++p == pe ) + goto _out9; +case 9: +#line 187 "parser.c" + switch( (*p) ) { + case 13: goto st9; + case 32: goto st9; + case 44: goto st10; + case 47: goto st15; + case 125: goto tr4; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st9; + goto st0; +st10: + if ( ++p == pe ) + goto _out10; +case 10: + switch( (*p) ) { + case 13: goto st10; + case 32: goto st10; + case 34: goto tr2; + case 47: goto st11; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st10; + goto st0; +st11: + if ( ++p == pe ) + goto _out11; +case 11: + switch( (*p) ) { + case 42: goto st12; + case 47: goto st14; + } + goto st0; +st12: + if ( ++p == pe ) + goto _out12; +case 12: + if ( (*p) == 42 ) + goto st13; + goto st12; +st13: + if ( ++p == pe ) + goto _out13; +case 13: + switch( (*p) ) { + case 42: goto st13; + case 47: goto st10; + } + goto st12; +st14: + if ( ++p == pe ) + goto _out14; +case 14: + if ( (*p) == 10 ) + goto st10; + goto st14; +st15: + if ( ++p == pe ) + goto _out15; +case 15: + switch( (*p) ) { + case 42: goto st16; + case 47: goto st18; + } + goto st0; +st16: + if ( ++p == pe ) + goto _out16; +case 16: + if ( (*p) == 42 ) + goto st17; + goto st16; +st17: + if ( ++p == pe ) + goto _out17; +case 17: + switch( (*p) ) { + case 42: goto st17; + case 47: goto st9; + } + goto st16; +st18: + if ( ++p == pe ) + goto _out18; +case 18: + if ( (*p) == 10 ) + goto st9; + goto st18; +tr4: +#line 82 "parser.rl" + { goto _out27; } + goto st27; +st27: + if ( ++p == pe ) + goto _out27; +case 27: +#line 283 "parser.c" + goto st0; +st19: + if ( ++p == pe ) + goto _out19; +case 19: + switch( (*p) ) { + case 42: goto st20; + case 47: goto st22; + } + goto st0; +st20: + if ( ++p == pe ) + goto _out20; +case 20: + if ( (*p) == 42 ) + goto st21; + goto st20; +st21: + if ( ++p == pe ) + goto _out21; +case 21: + switch( (*p) ) { + case 42: goto st21; + case 47: goto st8; + } + goto st20; +st22: + if ( ++p == pe ) + goto _out22; +case 22: + if ( (*p) == 10 ) + goto st8; + goto st22; +st23: + if ( ++p == pe ) + goto _out23; +case 23: + switch( (*p) ) { + case 42: goto st24; + case 47: goto st26; + } + goto st0; +st24: + if ( ++p == pe ) + goto _out24; +case 24: + if ( (*p) == 42 ) + goto st25; + goto st24; +st25: + if ( ++p == pe ) + goto _out25; +case 25: + switch( (*p) ) { + case 42: goto st25; + case 47: goto st2; + } + goto st24; +st26: + if ( ++p == pe ) + goto _out26; +case 26: + if ( (*p) == 10 ) + goto st2; + goto st26; + } + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + _out8: cs = 8; goto _out; + _out9: cs = 9; goto _out; + _out10: cs = 10; goto _out; + _out11: cs = 11; goto _out; + _out12: cs = 12; goto _out; + _out13: cs = 13; goto _out; + _out14: cs = 14; goto _out; + _out15: cs = 15; goto _out; + _out16: cs = 16; goto _out; + _out17: cs = 17; goto _out; + _out18: cs = 18; goto _out; + _out27: cs = 27; goto _out; + _out19: cs = 19; goto _out; + _out20: cs = 20; goto _out; + _out21: cs = 21; goto _out; + _out22: cs = 22; goto _out; + _out23: cs = 23; goto _out; + _out24: cs = 24; goto _out; + _out25: cs = 25; goto _out; + _out26: cs = 26; goto _out; + + _out: {} + } +#line 106 "parser.rl" + + if (cs >= JSON_object_first_final) { + VALUE klassname = rb_hash_aref(*result, json->create_id); + if (!NIL_P(klassname)) { + VALUE klass = rb_path2class(StringValueCStr(klassname)); + if RTEST(rb_funcall(klass, i_json_creatable_p, 0)) { + *result = rb_funcall(klass, i_json_create, 1, *result); + } + } + return p + 1; + } else { + return NULL; + } +} + + +#line 397 "parser.c" +static const int JSON_value_start = 1; +static const int JSON_value_first_final = 12; +static const int JSON_value_error = 0; + +static const int JSON_value_en_main = 1; + +#line 177 "parser.rl" + + +static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + +#line 412 "parser.c" + { + cs = JSON_value_start; + } +#line 184 "parser.rl" + +#line 418 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + switch( (*p) ) { + case 34: goto tr0; + case 45: goto tr2; + case 91: goto tr3; + case 102: goto st2; + case 110: goto st6; + case 116: goto st9; + case 123: goto tr7; + } + if ( 48 <= (*p) && (*p) <= 57 ) + goto tr2; + goto st0; +st0: + goto _out0; +tr0: +#line 136 "parser.rl" + { + char *np = JSON_parse_string(json, p, pe, result); + if (np == NULL) goto _out12; else {p = (( np))-1;} + } + goto st12; +tr2: +#line 141 "parser.rl" + { + char *np; + np = JSON_parse_float(json, p, pe, result); + if (np != NULL) {p = (( np))-1;} + np = JSON_parse_integer(json, p, pe, result); + if (np != NULL) {p = (( np))-1;} + goto _out12; + } + goto st12; +tr3: +#line 150 "parser.rl" + { + char *np; + json->current_nesting += 1; + np = JSON_parse_array(json, p, pe, result); + json->current_nesting -= 1; + if (np == NULL) goto _out12; else {p = (( np))-1;} + } + goto st12; +tr7: +#line 158 "parser.rl" + { + char *np; + json->current_nesting += 1; + np = JSON_parse_object(json, p, pe, result); + json->current_nesting -= 1; + if (np == NULL) goto _out12; else {p = (( np))-1;} + } + goto st12; +tr11: +#line 130 "parser.rl" + { + *result = Qfalse; + } + goto st12; +tr14: +#line 127 "parser.rl" + { + *result = Qnil; + } + goto st12; +tr17: +#line 133 "parser.rl" + { + *result = Qtrue; + } + goto st12; +st12: + if ( ++p == pe ) + goto _out12; +case 12: +#line 166 "parser.rl" + { goto _out12; } +#line 501 "parser.c" + goto st0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + if ( (*p) == 97 ) + goto st3; + goto st0; +st3: + if ( ++p == pe ) + goto _out3; +case 3: + if ( (*p) == 108 ) + goto st4; + goto st0; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + if ( (*p) == 115 ) + goto st5; + goto st0; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + if ( (*p) == 101 ) + goto tr11; + goto st0; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + if ( (*p) == 117 ) + goto st7; + goto st0; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + if ( (*p) == 108 ) + goto st8; + goto st0; +st8: + if ( ++p == pe ) + goto _out8; +case 8: + if ( (*p) == 108 ) + goto tr14; + goto st0; +st9: + if ( ++p == pe ) + goto _out9; +case 9: + if ( (*p) == 114 ) + goto st10; + goto st0; +st10: + if ( ++p == pe ) + goto _out10; +case 10: + if ( (*p) == 117 ) + goto st11; + goto st0; +st11: + if ( ++p == pe ) + goto _out11; +case 11: + if ( (*p) == 101 ) + goto tr17; + goto st0; + } + _out0: cs = 0; goto _out; + _out12: cs = 12; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + _out8: cs = 8; goto _out; + _out9: cs = 9; goto _out; + _out10: cs = 10; goto _out; + _out11: cs = 11; goto _out; + + _out: {} + } +#line 185 "parser.rl" + + if (cs >= JSON_value_first_final) { + return p; + } else { + return NULL; + } +} + + +#line 599 "parser.c" +static const int JSON_integer_start = 1; +static const int JSON_integer_first_final = 5; +static const int JSON_integer_error = 0; + +static const int JSON_integer_en_main = 1; + +#line 201 "parser.rl" + + +static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + +#line 614 "parser.c" + { + cs = JSON_integer_start; + } +#line 208 "parser.rl" + json->memo = p; + +#line 621 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + switch( (*p) ) { + case 45: goto st2; + case 48: goto st3; + } + if ( 49 <= (*p) && (*p) <= 57 ) + goto st4; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + if ( (*p) == 48 ) + goto st3; + if ( 49 <= (*p) && (*p) <= 57 ) + goto st4; + goto st0; +st3: + if ( ++p == pe ) + goto _out3; +case 3: + if ( 48 <= (*p) && (*p) <= 57 ) + goto st0; + goto tr4; +tr4: +#line 198 "parser.rl" + { goto _out5; } + goto st5; +st5: + if ( ++p == pe ) + goto _out5; +case 5: +#line 661 "parser.c" + goto st0; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + if ( 48 <= (*p) && (*p) <= 57 ) + goto st4; + goto tr4; + } + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out5: cs = 5; goto _out; + _out4: cs = 4; goto _out; + + _out: {} + } +#line 210 "parser.rl" + + if (cs >= JSON_integer_first_final) { + long len = p - json->memo; + *result = rb_Integer(rb_str_new(json->memo, len)); + return p + 1; + } else { + return NULL; + } +} + + +#line 691 "parser.c" +static const int JSON_float_start = 1; +static const int JSON_float_first_final = 10; +static const int JSON_float_error = 0; + +static const int JSON_float_en_main = 1; + +#line 232 "parser.rl" + + +static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + +#line 706 "parser.c" + { + cs = JSON_float_start; + } +#line 239 "parser.rl" + json->memo = p; + +#line 713 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + switch( (*p) ) { + case 45: goto st2; + case 48: goto st3; + } + if ( 49 <= (*p) && (*p) <= 57 ) + goto st9; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + if ( (*p) == 48 ) + goto st3; + if ( 49 <= (*p) && (*p) <= 57 ) + goto st9; + goto st0; +st3: + if ( ++p == pe ) + goto _out3; +case 3: + switch( (*p) ) { + case 46: goto st4; + case 69: goto st6; + case 101: goto st6; + } + goto st0; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + if ( 48 <= (*p) && (*p) <= 57 ) + goto st5; + goto st0; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + switch( (*p) ) { + case 69: goto st6; + case 101: goto st6; + } + if ( (*p) > 46 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st5; + } else if ( (*p) >= 45 ) + goto st0; + goto tr7; +tr7: +#line 226 "parser.rl" + { goto _out10; } + goto st10; +st10: + if ( ++p == pe ) + goto _out10; +case 10: +#line 777 "parser.c" + goto st0; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + switch( (*p) ) { + case 43: goto st7; + case 45: goto st7; + } + if ( 48 <= (*p) && (*p) <= 57 ) + goto st8; + goto st0; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + if ( 48 <= (*p) && (*p) <= 57 ) + goto st8; + goto st0; +st8: + if ( ++p == pe ) + goto _out8; +case 8: + switch( (*p) ) { + case 69: goto st0; + case 101: goto st0; + } + if ( (*p) > 46 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st8; + } else if ( (*p) >= 45 ) + goto st0; + goto tr7; +st9: + if ( ++p == pe ) + goto _out9; +case 9: + switch( (*p) ) { + case 46: goto st4; + case 69: goto st6; + case 101: goto st6; + } + if ( 48 <= (*p) && (*p) <= 57 ) + goto st9; + goto st0; + } + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out10: cs = 10; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + _out8: cs = 8; goto _out; + _out9: cs = 9; goto _out; + + _out: {} + } +#line 241 "parser.rl" + + if (cs >= JSON_float_first_final) { + long len = p - json->memo; + *result = rb_Float(rb_str_new(json->memo, len)); + return p + 1; + } else { + return NULL; + } +} + + + +#line 850 "parser.c" +static const int JSON_array_start = 1; +static const int JSON_array_first_final = 17; +static const int JSON_array_error = 0; + +static const int JSON_array_en_main = 1; + +#line 277 "parser.rl" + + +static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + if (json->max_nesting && json->current_nesting > json->max_nesting) { + rb_raise(eNestingError, "nesting of %d is to deep", json->current_nesting); + } + *result = rb_ary_new(); + + +#line 870 "parser.c" + { + cs = JSON_array_start; + } +#line 289 "parser.rl" + +#line 876 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + if ( (*p) == 91 ) + goto st2; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + switch( (*p) ) { + case 13: goto st2; + case 32: goto st2; + case 34: goto tr2; + case 45: goto tr2; + case 47: goto st13; + case 91: goto tr2; + case 93: goto tr4; + case 102: goto tr2; + case 110: goto tr2; + case 116: goto tr2; + case 123: goto tr2; + } + if ( (*p) > 10 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto tr2; + } else if ( (*p) >= 9 ) + goto st2; + goto st0; +tr2: +#line 258 "parser.rl" + { + VALUE v = Qnil; + char *np = JSON_parse_value(json, p, pe, &v); + if (np == NULL) { + goto _out3; + } else { + rb_ary_push(*result, v); + {p = (( np))-1;} + } + } + goto st3; +st3: + if ( ++p == pe ) + goto _out3; +case 3: +#line 928 "parser.c" + switch( (*p) ) { + case 13: goto st3; + case 32: goto st3; + case 44: goto st4; + case 47: goto st9; + case 93: goto tr4; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st3; + goto st0; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + switch( (*p) ) { + case 13: goto st4; + case 32: goto st4; + case 34: goto tr2; + case 45: goto tr2; + case 47: goto st5; + case 91: goto tr2; + case 102: goto tr2; + case 110: goto tr2; + case 116: goto tr2; + case 123: goto tr2; + } + if ( (*p) > 10 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto tr2; + } else if ( (*p) >= 9 ) + goto st4; + goto st0; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + switch( (*p) ) { + case 42: goto st6; + case 47: goto st8; + } + goto st0; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + if ( (*p) == 42 ) + goto st7; + goto st6; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + switch( (*p) ) { + case 42: goto st7; + case 47: goto st4; + } + goto st6; +st8: + if ( ++p == pe ) + goto _out8; +case 8: + if ( (*p) == 10 ) + goto st4; + goto st8; +st9: + if ( ++p == pe ) + goto _out9; +case 9: + switch( (*p) ) { + case 42: goto st10; + case 47: goto st12; + } + goto st0; +st10: + if ( ++p == pe ) + goto _out10; +case 10: + if ( (*p) == 42 ) + goto st11; + goto st10; +st11: + if ( ++p == pe ) + goto _out11; +case 11: + switch( (*p) ) { + case 42: goto st11; + case 47: goto st3; + } + goto st10; +st12: + if ( ++p == pe ) + goto _out12; +case 12: + if ( (*p) == 10 ) + goto st3; + goto st12; +tr4: +#line 269 "parser.rl" + { goto _out17; } + goto st17; +st17: + if ( ++p == pe ) + goto _out17; +case 17: +#line 1033 "parser.c" + goto st0; +st13: + if ( ++p == pe ) + goto _out13; +case 13: + switch( (*p) ) { + case 42: goto st14; + case 47: goto st16; + } + goto st0; +st14: + if ( ++p == pe ) + goto _out14; +case 14: + if ( (*p) == 42 ) + goto st15; + goto st14; +st15: + if ( ++p == pe ) + goto _out15; +case 15: + switch( (*p) ) { + case 42: goto st15; + case 47: goto st2; + } + goto st14; +st16: + if ( ++p == pe ) + goto _out16; +case 16: + if ( (*p) == 10 ) + goto st2; + goto st16; + } + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + _out8: cs = 8; goto _out; + _out9: cs = 9; goto _out; + _out10: cs = 10; goto _out; + _out11: cs = 11; goto _out; + _out12: cs = 12; goto _out; + _out17: cs = 17; goto _out; + _out13: cs = 13; goto _out; + _out14: cs = 14; goto _out; + _out15: cs = 15; goto _out; + _out16: cs = 16; goto _out; + + _out: {} + } +#line 290 "parser.rl" + + if(cs >= JSON_array_first_final) { + return p + 1; + } else { + rb_raise(eParserError, "unexpected token at '%s'", p); + } +} + +static VALUE json_string_unescape(char *p, char *pe) +{ + VALUE result = rb_str_buf_new(pe - p + 1); + + while (p < pe) { + if (*p == '\\') { + p++; + if (p >= pe) return Qnil; /* raise an exception later, \ at end */ + switch (*p) { + case '"': + case '\\': + rb_str_buf_cat(result, p, 1); + p++; + break; + case 'b': + rb_str_buf_cat2(result, "\b"); + p++; + break; + case 'f': + rb_str_buf_cat2(result, "\f"); + p++; + break; + case 'n': + rb_str_buf_cat2(result, "\n"); + p++; + break; + case 'r': + rb_str_buf_cat2(result, "\r"); + p++; + break; + case 't': + rb_str_buf_cat2(result, "\t"); + p++; + break; + case 'u': + if (p > pe - 4) { + return Qnil; + } else { + p = JSON_convert_UTF16_to_UTF8(result, p, pe, strictConversion); + } + break; + default: + rb_str_buf_cat(result, p, 1); + p++; + break; + } + } else { + char *q = p; + while (*q != '\\' && q < pe) q++; + rb_str_buf_cat(result, p, q - p); + p = q; + } + } + return result; +} + + +#line 1154 "parser.c" +static const int JSON_string_start = 1; +static const int JSON_string_first_final = 8; +static const int JSON_string_error = 0; + +static const int JSON_string_en_main = 1; + +#line 368 "parser.rl" + + +static char *JSON_parse_string(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + *result = rb_str_new("", 0); + +#line 1170 "parser.c" + { + cs = JSON_string_start; + } +#line 376 "parser.rl" + json->memo = p; + +#line 1177 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +case 1: + if ( (*p) == 34 ) + goto st2; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + switch( (*p) ) { + case 34: goto tr2; + case 92: goto st3; + } + if ( 0 <= (*p) && (*p) <= 31 ) + goto st0; + goto st2; +tr2: +#line 360 "parser.rl" + { + *result = json_string_unescape(json->memo + 1, p); + if (NIL_P(*result)) goto _out8; else {p = (( p + 1))-1;} + } +#line 365 "parser.rl" + { goto _out8; } + goto st8; +st8: + if ( ++p == pe ) + goto _out8; +case 8: +#line 1213 "parser.c" + goto st0; +st3: + if ( ++p == pe ) + goto _out3; +case 3: + if ( (*p) == 117 ) + goto st4; + if ( 0 <= (*p) && (*p) <= 31 ) + goto st0; + goto st2; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + if ( (*p) < 65 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st5; + } else if ( (*p) > 70 ) { + if ( 97 <= (*p) && (*p) <= 102 ) + goto st5; + } else + goto st5; + goto st0; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + if ( (*p) < 65 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st6; + } else if ( (*p) > 70 ) { + if ( 97 <= (*p) && (*p) <= 102 ) + goto st6; + } else + goto st6; + goto st0; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + if ( (*p) < 65 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st7; + } else if ( (*p) > 70 ) { + if ( 97 <= (*p) && (*p) <= 102 ) + goto st7; + } else + goto st7; + goto st0; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + if ( (*p) < 65 ) { + if ( 48 <= (*p) && (*p) <= 57 ) + goto st2; + } else if ( (*p) > 70 ) { + if ( 97 <= (*p) && (*p) <= 102 ) + goto st2; + } else + goto st2; + goto st0; + } + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out8: cs = 8; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + + _out: {} + } +#line 378 "parser.rl" + + if (cs >= JSON_string_first_final) { + return p + 1; + } else { + return NULL; + } +} + + + +#line 1299 "parser.c" +static const int JSON_start = 1; +static const int JSON_first_final = 10; +static const int JSON_error = 0; + +static const int JSON_en_main = 1; + +#line 412 "parser.rl" + + +/* + * Document-class: JSON::Ext::Parser + * + * This is the JSON parser implemented as a C extension. It can be configured + * to be used by setting + * + * JSON.parser = JSON::Ext::Parser + * + * with the method parser= in JSON. + * + */ + +/* + * call-seq: new(source, opts => {}) + * + * Creates a new JSON::Ext::Parser instance for the string _source_. + * + * Creates a new JSON::Ext::Parser instance for the string _source_. + * + * It will be configured by the _opts_ hash. _opts_ can have the following + * keys: + * + * _opts_ can have the following keys: + * * *max_nesting*: The maximum depth of nesting allowed in the parsed data + * structures. Disable depth checking with :max_nesting => false. + */ +static VALUE cParser_initialize(int argc, VALUE *argv, VALUE self) +{ + char *ptr; + long len; + VALUE source, opts; + GET_STRUCT; + rb_scan_args(argc, argv, "11", &source, &opts); + source = StringValue(source); + ptr = RSTRING_PTR(source); + len = RSTRING_LEN(source); + if (len < 2) { + rb_raise(eParserError, "A JSON text must at least contain two octets!"); + } + json->max_nesting = 19; + if (!NIL_P(opts)) { + opts = rb_convert_type(opts, T_HASH, "Hash", "to_hash"); + if (NIL_P(opts)) { + rb_raise(rb_eArgError, "opts needs to be like a hash"); + } else { + VALUE s_max_nesting = ID2SYM(i_max_nesting); + if (st_lookup(RHASH(opts)->tbl, s_max_nesting, 0)) { + VALUE max_nesting = rb_hash_aref(opts, s_max_nesting); + if (RTEST(max_nesting)) { + Check_Type(max_nesting, T_FIXNUM); + json->max_nesting = FIX2INT(max_nesting); + } else { + json->max_nesting = 0; + } + } + } + } + json->current_nesting = 0; + /* + Convert these? + if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } + */ + json->len = len; + json->source = ptr; + json->Vsource = source; + json->create_id = rb_funcall(mJSON, i_create_id, 0); + return self; +} + +/* + * call-seq: parse() + * + * Parses the current JSON text _source_ and returns the complete data + * structure as a result. + */ +static VALUE cParser_parse(VALUE self) +{ + char *p, *pe; + int cs = EVIL; + VALUE result = Qnil; + GET_STRUCT; + + +#line 1400 "parser.c" + { + cs = JSON_start; + } +#line 505 "parser.rl" + p = json->source; + pe = p + json->len; + +#line 1408 "parser.c" + { + if ( p == pe ) + goto _out; + switch ( cs ) + { +st1: + if ( ++p == pe ) + goto _out1; +case 1: + switch( (*p) ) { + case 13: goto st1; + case 32: goto st1; + case 47: goto st2; + case 91: goto tr3; + case 123: goto tr4; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st1; + goto st0; +st0: + goto _out0; +st2: + if ( ++p == pe ) + goto _out2; +case 2: + switch( (*p) ) { + case 42: goto st3; + case 47: goto st5; + } + goto st0; +st3: + if ( ++p == pe ) + goto _out3; +case 3: + if ( (*p) == 42 ) + goto st4; + goto st3; +st4: + if ( ++p == pe ) + goto _out4; +case 4: + switch( (*p) ) { + case 42: goto st4; + case 47: goto st1; + } + goto st3; +st5: + if ( ++p == pe ) + goto _out5; +case 5: + if ( (*p) == 10 ) + goto st1; + goto st5; +tr3: +#line 401 "parser.rl" + { + char *np; + json->current_nesting = 1; + np = JSON_parse_array(json, p, pe, &result); + if (np == NULL) goto _out10; else {p = (( np))-1;} + } + goto st10; +tr4: +#line 394 "parser.rl" + { + char *np; + json->current_nesting = 1; + np = JSON_parse_object(json, p, pe, &result); + if (np == NULL) goto _out10; else {p = (( np))-1;} + } + goto st10; +st10: + if ( ++p == pe ) + goto _out10; +case 10: +#line 1484 "parser.c" + switch( (*p) ) { + case 13: goto st10; + case 32: goto st10; + case 47: goto st6; + } + if ( 9 <= (*p) && (*p) <= 10 ) + goto st10; + goto st0; +st6: + if ( ++p == pe ) + goto _out6; +case 6: + switch( (*p) ) { + case 42: goto st7; + case 47: goto st9; + } + goto st0; +st7: + if ( ++p == pe ) + goto _out7; +case 7: + if ( (*p) == 42 ) + goto st8; + goto st7; +st8: + if ( ++p == pe ) + goto _out8; +case 8: + switch( (*p) ) { + case 42: goto st8; + case 47: goto st10; + } + goto st7; +st9: + if ( ++p == pe ) + goto _out9; +case 9: + if ( (*p) == 10 ) + goto st10; + goto st9; + } + _out1: cs = 1; goto _out; + _out0: cs = 0; goto _out; + _out2: cs = 2; goto _out; + _out3: cs = 3; goto _out; + _out4: cs = 4; goto _out; + _out5: cs = 5; goto _out; + _out10: cs = 10; goto _out; + _out6: cs = 6; goto _out; + _out7: cs = 7; goto _out; + _out8: cs = 8; goto _out; + _out9: cs = 9; goto _out; + + _out: {} + } +#line 508 "parser.rl" + + if (cs >= JSON_first_final && p == pe) { + return result; + } else { + rb_raise(eParserError, "unexpected token at '%s'", p); + } +} + +static JSON_Parser *JSON_allocate() +{ + JSON_Parser *json = ALLOC(JSON_Parser); + MEMZERO(json, JSON_Parser, 1); + return json; +} + +static void JSON_mark(JSON_Parser *json) +{ + rb_gc_mark_maybe(json->Vsource); + rb_gc_mark_maybe(json->create_id); +} + +static void JSON_free(JSON_Parser *json) +{ + free(json); +} + +static VALUE cJSON_parser_s_allocate(VALUE klass) +{ + JSON_Parser *json = JSON_allocate(); + return Data_Wrap_Struct(klass, JSON_mark, JSON_free, json); +} + +/* + * call-seq: source() + * + * Returns a copy of the current _source_ string, that was used to construct + * this Parser. + */ +static VALUE cParser_source(VALUE self) +{ + GET_STRUCT; + return rb_str_dup(json->Vsource); +} + +void Init_parser() +{ + mJSON = rb_define_module("JSON"); + mExt = rb_define_module_under(mJSON, "Ext"); + cParser = rb_define_class_under(mExt, "Parser", rb_cObject); + eParserError = rb_path2class("JSON::ParserError"); + eNestingError = rb_path2class("JSON::NestingError"); + rb_define_alloc_func(cParser, cJSON_parser_s_allocate); + rb_define_method(cParser, "initialize", cParser_initialize, -1); + rb_define_method(cParser, "parse", cParser_parse, 0); + rb_define_method(cParser, "source", cParser_source, 0); + + i_json_creatable_p = rb_intern("json_creatable?"); + i_json_create = rb_intern("json_create"); + i_create_id = rb_intern("create_id"); + i_chr = rb_intern("chr"); + i_max_nesting = rb_intern("max_nesting"); +} diff --git a/ext/json/ext/parser/parser.rl b/ext/json/ext/parser/parser.rl new file mode 100644 index 0000000000..9ce8c6fc24 --- /dev/null +++ b/ext/json/ext/parser/parser.rl @@ -0,0 +1,569 @@ +/* vim: set cin et sw=4 ts=4: */ + +#include "ruby.h" +#include "re.h" +#include "st.h" +#include "unicode.h" + +#define EVIL 0x666 + +static VALUE mJSON, mExt, cParser, eParserError, eNestingError; + +static ID i_json_creatable_p, i_json_create, i_create_id, i_chr, i_max_nesting; + +typedef struct JSON_ParserStruct { + VALUE Vsource; + char *source; + long len; + char *memo; + VALUE create_id; + int max_nesting; + int current_nesting; +} JSON_Parser; + +static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_string(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result); +static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result); + +#define GET_STRUCT \ + JSON_Parser *json; \ + Data_Get_Struct(self, JSON_Parser, json); + +%%{ + machine JSON_common; + + cr = '\n'; + cr_neg = [^\n]; + ws = [ \t\r\n]; + c_comment = '/*' ( any* - (any* '*/' any* ) ) '*/'; + cpp_comment = '//' cr_neg* cr; + comment = c_comment | cpp_comment; + ignore = ws | comment; + name_separator = ':'; + value_separator = ','; + Vnull = 'null'; + Vfalse = 'false'; + Vtrue = 'true'; + begin_value = [nft"\-[{] | digit; + begin_object = '{'; + end_object = '}'; + begin_array = '['; + end_array = ']'; + begin_string = '"'; + begin_name = begin_string; + begin_number = digit | '-'; +}%% + +%%{ + machine JSON_object; + include JSON_common; + + write data; + + action parse_value { + VALUE v = Qnil; + char *np = JSON_parse_value(json, fpc, pe, &v); + if (np == NULL) { + fbreak; + } else { + rb_hash_aset(*result, last_name, v); + fexec np; + } + } + + action parse_name { + char *np = JSON_parse_string(json, fpc, pe, &last_name); + if (np == NULL) fbreak; else fexec np; + } + + action exit { fbreak; } + + a_pair = ignore* begin_name >parse_name + ignore* name_separator ignore* + begin_value >parse_value; + + main := begin_object + (a_pair (ignore* value_separator a_pair)*)? + ignore* end_object @exit; +}%% + +static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + VALUE last_name = Qnil; + + if (json->max_nesting && json->current_nesting > json->max_nesting) { + rb_raise(eNestingError, "nesting of %d is to deep", json->current_nesting); + } + + *result = rb_hash_new(); + + %% write init; + %% write exec; + + if (cs >= JSON_object_first_final) { + VALUE klassname = rb_hash_aref(*result, json->create_id); + if (!NIL_P(klassname)) { + VALUE klass = rb_path2class(StringValueCStr(klassname)); + if RTEST(rb_funcall(klass, i_json_creatable_p, 0)) { + *result = rb_funcall(klass, i_json_create, 1, *result); + } + } + return p + 1; + } else { + return NULL; + } +} + +%%{ + machine JSON_value; + include JSON_common; + + write data; + + action parse_null { + *result = Qnil; + } + action parse_false { + *result = Qfalse; + } + action parse_true { + *result = Qtrue; + } + action parse_string { + char *np = JSON_parse_string(json, fpc, pe, result); + if (np == NULL) fbreak; else fexec np; + } + + action parse_number { + char *np; + np = JSON_parse_float(json, fpc, pe, result); + if (np != NULL) fexec np; + np = JSON_parse_integer(json, fpc, pe, result); + if (np != NULL) fexec np; + fbreak; + } + + action parse_array { + char *np; + json->current_nesting += 1; + np = JSON_parse_array(json, fpc, pe, result); + json->current_nesting -= 1; + if (np == NULL) fbreak; else fexec np; + } + + action parse_object { + char *np; + json->current_nesting += 1; + np = JSON_parse_object(json, fpc, pe, result); + json->current_nesting -= 1; + if (np == NULL) fbreak; else fexec np; + } + + action exit { fbreak; } + +main := ( + Vnull @parse_null | + Vfalse @parse_false | + Vtrue @parse_true | + begin_number >parse_number | + begin_string >parse_string | + begin_array >parse_array | + begin_object >parse_object + ) %*exit; +}%% + +static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + %% write init; + %% write exec; + + if (cs >= JSON_value_first_final) { + return p; + } else { + return NULL; + } +} + +%%{ + machine JSON_integer; + + write data; + + action exit { fbreak; } + + main := '-'? ('0' | [1-9][0-9]*) (^[0-9] @exit); +}%% + +static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + %% write init; + json->memo = p; + %% write exec; + + if (cs >= JSON_integer_first_final) { + long len = p - json->memo; + *result = rb_Integer(rb_str_new(json->memo, len)); + return p + 1; + } else { + return NULL; + } +} + +%%{ + machine JSON_float; + include JSON_common; + + write data; + + action exit { fbreak; } + + main := '-'? ( + (('0' | [1-9][0-9]*) '.' [0-9]+ ([Ee] [+\-]?[0-9]+)?) + | (('0' | [1-9][0-9]*) ([Ee] [+\-]?[0-9]+)) + ) (^[0-9Ee.\-] @exit ); +}%% + +static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + %% write init; + json->memo = p; + %% write exec; + + if (cs >= JSON_float_first_final) { + long len = p - json->memo; + *result = rb_Float(rb_str_new(json->memo, len)); + return p + 1; + } else { + return NULL; + } +} + + +%%{ + machine JSON_array; + include JSON_common; + + write data; + + action parse_value { + VALUE v = Qnil; + char *np = JSON_parse_value(json, fpc, pe, &v); + if (np == NULL) { + fbreak; + } else { + rb_ary_push(*result, v); + fexec np; + } + } + + action exit { fbreak; } + + next_element = value_separator ignore* begin_value >parse_value; + + main := begin_array ignore* + ((begin_value >parse_value ignore*) + (ignore* next_element ignore*)*)? + end_array @exit; +}%% + +static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + if (json->max_nesting && json->current_nesting > json->max_nesting) { + rb_raise(eNestingError, "nesting of %d is to deep", json->current_nesting); + } + *result = rb_ary_new(); + + %% write init; + %% write exec; + + if(cs >= JSON_array_first_final) { + return p + 1; + } else { + rb_raise(eParserError, "unexpected token at '%s'", p); + } +} + +static VALUE json_string_unescape(char *p, char *pe) +{ + VALUE result = rb_str_buf_new(pe - p + 1); + + while (p < pe) { + if (*p == '\\') { + p++; + if (p >= pe) return Qnil; /* raise an exception later, \ at end */ + switch (*p) { + case '"': + case '\\': + rb_str_buf_cat(result, p, 1); + p++; + break; + case 'b': + rb_str_buf_cat2(result, "\b"); + p++; + break; + case 'f': + rb_str_buf_cat2(result, "\f"); + p++; + break; + case 'n': + rb_str_buf_cat2(result, "\n"); + p++; + break; + case 'r': + rb_str_buf_cat2(result, "\r"); + p++; + break; + case 't': + rb_str_buf_cat2(result, "\t"); + p++; + break; + case 'u': + if (p > pe - 4) { + return Qnil; + } else { + p = JSON_convert_UTF16_to_UTF8(result, p, pe, strictConversion); + } + break; + default: + rb_str_buf_cat(result, p, 1); + p++; + break; + } + } else { + char *q = p; + while (*q != '\\' && q < pe) q++; + rb_str_buf_cat(result, p, q - p); + p = q; + } + } + return result; +} + +%%{ + machine JSON_string; + include JSON_common; + + write data; + + action parse_string { + *result = json_string_unescape(json->memo + 1, p); + if (NIL_P(*result)) fbreak; else fexec p + 1; + } + + action exit { fbreak; } + + main := '"' ((^(["\\] | 0..0x1f) | '\\'["\\/bfnrt] | '\\u'[0-9a-fA-F]{4} | '\\'^(["\\/bfnrtu]|0..0x1f))* %parse_string) '"' @exit; +}%% + +static char *JSON_parse_string(JSON_Parser *json, char *p, char *pe, VALUE *result) +{ + int cs = EVIL; + + *result = rb_str_new("", 0); + %% write init; + json->memo = p; + %% write exec; + + if (cs >= JSON_string_first_final) { + return p + 1; + } else { + return NULL; + } +} + + +%%{ + machine JSON; + + write data; + + include JSON_common; + + action parse_object { + char *np; + json->current_nesting = 1; + np = JSON_parse_object(json, fpc, pe, &result); + if (np == NULL) fbreak; else fexec np; + } + + action parse_array { + char *np; + json->current_nesting = 1; + np = JSON_parse_array(json, fpc, pe, &result); + if (np == NULL) fbreak; else fexec np; + } + + main := ignore* ( + begin_object >parse_object | + begin_array >parse_array + ) ignore*; +}%% + +/* + * Document-class: JSON::Ext::Parser + * + * This is the JSON parser implemented as a C extension. It can be configured + * to be used by setting + * + * JSON.parser = JSON::Ext::Parser + * + * with the method parser= in JSON. + * + */ + +/* + * call-seq: new(source, opts => {}) + * + * Creates a new JSON::Ext::Parser instance for the string _source_. + * + * Creates a new JSON::Ext::Parser instance for the string _source_. + * + * It will be configured by the _opts_ hash. _opts_ can have the following + * keys: + * + * _opts_ can have the following keys: + * * *max_nesting*: The maximum depth of nesting allowed in the parsed data + * structures. Disable depth checking with :max_nesting => false. + */ +static VALUE cParser_initialize(int argc, VALUE *argv, VALUE self) +{ + char *ptr; + long len; + VALUE source, opts; + GET_STRUCT; + rb_scan_args(argc, argv, "11", &source, &opts); + source = StringValue(source); + ptr = RSTRING_PTR(source); + len = RSTRING_LEN(source); + if (len < 2) { + rb_raise(eParserError, "A JSON text must at least contain two octets!"); + } + json->max_nesting = 19; + if (!NIL_P(opts)) { + opts = rb_convert_type(opts, T_HASH, "Hash", "to_hash"); + if (NIL_P(opts)) { + rb_raise(rb_eArgError, "opts needs to be like a hash"); + } else { + VALUE s_max_nesting = ID2SYM(i_max_nesting); + if (st_lookup(RHASH(opts)->tbl, s_max_nesting, 0)) { + VALUE max_nesting = rb_hash_aref(opts, s_max_nesting); + if (RTEST(max_nesting)) { + Check_Type(max_nesting, T_FIXNUM); + json->max_nesting = FIX2INT(max_nesting); + } else { + json->max_nesting = 0; + } + } + } + } + json->current_nesting = 0; + /* + Convert these? + if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) { + rb_raise(eParserError, "Only UTF8 octet streams are supported atm!"); + } + */ + json->len = len; + json->source = ptr; + json->Vsource = source; + json->create_id = rb_funcall(mJSON, i_create_id, 0); + return self; +} + +/* + * call-seq: parse() + * + * Parses the current JSON text _source_ and returns the complete data + * structure as a result. + */ +static VALUE cParser_parse(VALUE self) +{ + char *p, *pe; + int cs = EVIL; + VALUE result = Qnil; + GET_STRUCT; + + %% write init; + p = json->source; + pe = p + json->len; + %% write exec; + + if (cs >= JSON_first_final && p == pe) { + return result; + } else { + rb_raise(eParserError, "unexpected token at '%s'", p); + } +} + +static JSON_Parser *JSON_allocate() +{ + JSON_Parser *json = ALLOC(JSON_Parser); + MEMZERO(json, JSON_Parser, 1); + return json; +} + +static void JSON_mark(JSON_Parser *json) +{ + rb_gc_mark_maybe(json->Vsource); + rb_gc_mark_maybe(json->create_id); +} + +static void JSON_free(JSON_Parser *json) +{ + free(json); +} + +static VALUE cJSON_parser_s_allocate(VALUE klass) +{ + JSON_Parser *json = JSON_allocate(); + return Data_Wrap_Struct(klass, JSON_mark, JSON_free, json); +} + +/* + * call-seq: source() + * + * Returns a copy of the current _source_ string, that was used to construct + * this Parser. + */ +static VALUE cParser_source(VALUE self) +{ + GET_STRUCT; + return rb_str_dup(json->Vsource); +} + +void Init_parser() +{ + mJSON = rb_define_module("JSON"); + mExt = rb_define_module_under(mJSON, "Ext"); + cParser = rb_define_class_under(mExt, "Parser", rb_cObject); + eParserError = rb_path2class("JSON::ParserError"); + eNestingError = rb_path2class("JSON::NestingError"); + rb_define_alloc_func(cParser, cJSON_parser_s_allocate); + rb_define_method(cParser, "initialize", cParser_initialize, -1); + rb_define_method(cParser, "parse", cParser_parse, 0); + rb_define_method(cParser, "source", cParser_source, 0); + + i_json_creatable_p = rb_intern("json_creatable?"); + i_json_create = rb_intern("json_create"); + i_create_id = rb_intern("create_id"); + i_chr = rb_intern("chr"); + i_max_nesting = rb_intern("max_nesting"); +} diff --git a/ext/json/ext/parser/unicode.c b/ext/json/ext/parser/unicode.c new file mode 100644 index 0000000000..609a0e83e2 --- /dev/null +++ b/ext/json/ext/parser/unicode.c @@ -0,0 +1,156 @@ +/* vim: set cin et sw=4 ts=4: */ + +#include "unicode.h" + +/* + * Copyright 2001-2004 Unicode, Inc. + * + * 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. + * + * 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. + */ + +/* + * Index into the table below with the first byte of a UTF-8 sequence to + * get the number of trailing bytes that are supposed to follow it. + * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is + * left as-is for anyone who may want to do such conversion, which was + * allowed in earlier algorithms. + */ +static const char trailingBytesForUTF8[256] = { + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 +}; + +/* + * Magic values subtracted from a buffer value during UTF8 conversion. + * This table contains as many values as there might be trailing bytes + * in a UTF-8 sequence. + */ +static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, + 0x03C82080UL, 0xFA082080UL, 0x82082080UL }; + +/* + * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed + * into the first byte, depending on how many bytes follow. There are + * as many entries in this table as there are UTF-8 sequence types. + * (I.e., one byte sequence, two byte... etc.). Remember that sequencs + * for *legal* UTF-8 will be 4 or fewer bytes total. + */ +static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC }; + +char *JSON_convert_UTF16_to_UTF8 ( + VALUE buffer, + char *source, + char *sourceEnd, + ConversionFlags flags) +{ + UTF16 *tmp, *tmpPtr, *tmpEnd; + char buf[5]; + long n = 0, i; + char *p = source - 1; + + while (p < sourceEnd && p[0] == '\\' && p[1] == 'u') { + p += 6; + n++; + } + p = source + 1; + buf[4] = 0; + tmpPtr = tmp = ALLOC_N(UTF16, n); + tmpEnd = tmp + n; + for (i = 0; i < n; i++) { + buf[0] = *p++; + buf[1] = *p++; + buf[2] = *p++; + buf[3] = *p++; + tmpPtr[i] = strtol(buf, NULL, 16); + p += 2; + } + + while (tmpPtr < tmpEnd) { + UTF32 ch; + unsigned short bytesToWrite = 0; + const UTF32 byteMask = 0xBF; + const UTF32 byteMark = 0x80; + ch = *tmpPtr++; + /* If we have a surrogate pair, convert to UTF32 first. */ + if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) { + /* If the 16 bits following the high surrogate are in the source + * buffer... */ + if (tmpPtr < tmpEnd) { + UTF32 ch2 = *tmpPtr; + /* If it's a low surrogate, convert to UTF32. */ + if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) { + ch = ((ch - UNI_SUR_HIGH_START) << halfShift) + + (ch2 - UNI_SUR_LOW_START) + halfBase; + ++tmpPtr; + } else if (flags == strictConversion) { /* it's an unpaired high surrogate */ + free(tmp); + rb_raise(rb_path2class("JSON::ParserError"), + "source sequence is illegal/malformed near %s", source); + } + } else { /* We don't have the 16 bits following the high surrogate. */ + free(tmp); + rb_raise(rb_path2class("JSON::ParserError"), + "partial character in source, but hit end near %s", source); + break; + } + } else if (flags == strictConversion) { + /* UTF-16 surrogate values are illegal in UTF-32 */ + if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) { + free(tmp); + rb_raise(rb_path2class("JSON::ParserError"), + "source sequence is illegal/malformed near %s", source); + } + } + /* Figure out how many bytes the result will require */ + if (ch < (UTF32) 0x80) { + bytesToWrite = 1; + } else if (ch < (UTF32) 0x800) { + bytesToWrite = 2; + } else if (ch < (UTF32) 0x10000) { + bytesToWrite = 3; + } else if (ch < (UTF32) 0x110000) { + bytesToWrite = 4; + } else { + bytesToWrite = 3; + ch = UNI_REPLACEMENT_CHAR; + } + + buf[0] = 0; + buf[1] = 0; + buf[2] = 0; + buf[3] = 0; + p = buf + bytesToWrite; + switch (bytesToWrite) { /* note: everything falls through. */ + case 4: *--p = (UTF8) ((ch | byteMark) & byteMask); ch >>= 6; + case 3: *--p = (UTF8) ((ch | byteMark) & byteMask); ch >>= 6; + case 2: *--p = (UTF8) ((ch | byteMark) & byteMask); ch >>= 6; + case 1: *--p = (UTF8) (ch | firstByteMark[bytesToWrite]); + } + rb_str_buf_cat(buffer, p, bytesToWrite); + } + free(tmp); + source += 5 + (n - 1) * 6; + return source; +} diff --git a/ext/json/ext/parser/unicode.h b/ext/json/ext/parser/unicode.h new file mode 100755 index 0000000000..155da0ceee --- /dev/null +++ b/ext/json/ext/parser/unicode.h @@ -0,0 +1,58 @@ + +#ifndef _PARSER_UNICODE_H_ +#define _PARSER_UNICODE_H_ + +#include "ruby.h" + +typedef unsigned long UTF32; /* at least 32 bits */ +typedef unsigned short UTF16; /* at least 16 bits */ +typedef unsigned char UTF8; /* typically 8 bits */ + +#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD +#define UNI_MAX_BMP (UTF32)0x0000FFFF +#define UNI_MAX_UTF16 (UTF32)0x0010FFFF +#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF +#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF + +#define UNI_SUR_HIGH_START (UTF32)0xD800 +#define UNI_SUR_HIGH_END (UTF32)0xDBFF +#define UNI_SUR_LOW_START (UTF32)0xDC00 +#define UNI_SUR_LOW_END (UTF32)0xDFFF + +static const int halfShift = 10; /* used for shifting by 10 bits */ + +static const UTF32 halfBase = 0x0010000UL; +static const UTF32 halfMask = 0x3FFUL; + +typedef enum { + conversionOK = 0, /* conversion successful */ + sourceExhausted, /* partial character in source, but hit end */ + targetExhausted, /* insuff. room in target for conversion */ + sourceIllegal /* source sequence is illegal/malformed */ +} ConversionResult; + +typedef enum { + strictConversion = 0, + lenientConversion +} ConversionFlags; + +char *JSON_convert_UTF16_to_UTF8 ( + VALUE buffer, + char *source, + char *sourceEnd, + ConversionFlags flags); + +#ifndef RARRAY_PTR +#define RARRAY_PTR(ARRAY) RARRAY(ARRAY)->ptr +#endif +#ifndef RARRAY_LEN +#define RARRAY_LEN(ARRAY) RARRAY(ARRAY)->len +#endif +#ifndef RSTRING_PTR +#define RSTRING_PTR(string) RSTRING(string)->ptr +#endif +#ifndef RSTRING_LEN +#define RSTRING_LEN(string) RSTRING(string)->len +#endif + +#endif diff --git a/ext/nkf/nkf-utf8/nkf.c b/ext/nkf/nkf-utf8/nkf.c index 3cd1b160da..30bb6c47b8 100644 --- a/ext/nkf/nkf-utf8/nkf.c +++ b/ext/nkf/nkf-utf8/nkf.c @@ -41,7 +41,7 @@ ***********************************************************************/ /* $Id$ */ #define NKF_VERSION "2.0.8" -#define NKF_RELEASE_DATE "2007-01-28" +#define NKF_RELEASE_DATE "2007-05-28" #include "config.h" #include "utf8tbl.h" @@ -351,10 +351,12 @@ static nkf_char e_iconv(nkf_char c2,nkf_char c1,nkf_char c0); * 0: Shift_JIS, eucJP-ascii * 1: eucJP-ms * 2: CP932, CP51932 + * 3: CP10001 */ -#define UCS_MAP_ASCII 0 -#define UCS_MAP_MS 1 -#define UCS_MAP_CP932 2 +#define UCS_MAP_ASCII 0 +#define UCS_MAP_MS 1 +#define UCS_MAP_CP932 2 +#define UCS_MAP_CP10001 3 static int ms_ucs_map_f = UCS_MAP_ASCII; #endif #ifdef UTF8_INPUT_ENABLE @@ -1232,6 +1234,14 @@ void options(unsigned char *cp) #endif #ifdef UTF8_OUTPUT_ENABLE ms_ucs_map_f = UCS_MAP_CP932; +#endif + }else if(strcmp(codeset, "CP10001") == 0){ + input_f = SJIS_INPUT; +#ifdef SHIFTJIS_CP932 + cp51932_f = TRUE; +#endif +#ifdef UTF8_OUTPUT_ENABLE + ms_ucs_map_f = UCS_MAP_CP10001; #endif }else if(strcmp(codeset, "EUCJP") == 0 || strcmp(codeset, "EUC-JP") == 0){ @@ -1370,6 +1380,11 @@ void options(unsigned char *cp) output_conv = s_oconv; #ifdef UTF8_OUTPUT_ENABLE ms_ucs_map_f = UCS_MAP_CP932; +#endif + }else if(strcmp(codeset, "CP10001") == 0){ + output_conv = s_oconv; +#ifdef UTF8_OUTPUT_ENABLE + ms_ucs_map_f = UCS_MAP_CP10001; #endif }else if(strcmp(codeset, "EUCJP") == 0 || strcmp(codeset, "EUC-JP") == 0){ @@ -2676,6 +2691,12 @@ nkf_char kanji_convert(FILE *f) } else { /* bogus code, skip SSO and one byte */ NEXT; } + } else if (ms_ucs_map_f == UCS_MAP_CP10001 && + (c1 == 0xFD || c1 == 0xFE)) { + /* CP10001 */ + c2 = X0201; + c1 &= 0x7f; + SEND; } else { /* already established */ c2 = c1; @@ -2885,35 +2906,41 @@ nkf_char kanji_convert(FILE *f) (*oconv)(0, ESC); SEND; } - } else if ((c1 == NL || c1 == CR) && broken_f&4) { - input_mode = ASCII; set_iconv(FALSE, 0); - SEND; - } else if (c1 == NL && mime_decode_f && !mime_decode_mode ) { - if ((c1=(*i_getc)(f))!=EOF && c1 == SPACE) { - i_ungetc(SPACE,f); - continue; - } else { - i_ungetc(c1,f); - } - c1 = NL; - SEND; - } else if (c1 == CR && mime_decode_f && !mime_decode_mode ) { - if ((c1=(*i_getc)(f))!=EOF) { - if (c1==SPACE) { - i_ungetc(SPACE,f); - continue; - } else if (c1 == NL && (c1=(*i_getc)(f))!=EOF && c1 == SPACE) { - i_ungetc(SPACE,f); - continue; - } else { - i_ungetc(c1,f); + } else if (c1 == NL || c1 == CR) { + if (broken_f&4) { + input_mode = ASCII; set_iconv(FALSE, 0); + SEND; + } else if (mime_decode_f && !mime_decode_mode){ + if (c1 == NL) { + if ((c1=(*i_getc)(f))!=EOF && c1 == SPACE) { + i_ungetc(SPACE,f); + continue; + } else { + i_ungetc(c1,f); + } + c1 = NL; + SEND; + } else { /* if (c1 == CR)*/ + if ((c1=(*i_getc)(f))!=EOF) { + if (c1==SPACE) { + i_ungetc(SPACE,f); + continue; + } else if (c1 == NL && (c1=(*i_getc)(f))!=EOF && c1 == SPACE) { + i_ungetc(SPACE,f); + continue; + } else { + i_ungetc(c1,f); + } + i_ungetc(NL,f); + } else { + i_ungetc(c1,f); + } + c1 = CR; + SEND; } - i_ungetc(NL,f); - } else { - i_ungetc(c1,f); } - c1 = CR; - SEND; + if (crmode_f == CR && c1 == NL) crmode_f = CRLF; + else crmode_f = c1; } else if (c1 == DEL && input_mode == X0208 ) { /* CP5022x */ c2 = c1; @@ -3125,9 +3152,6 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) static const nkf_char shift_jisx0213_s1a3_table[5][2] ={ { 1, 8}, { 3, 4}, { 5,12}, {13,14}, {15, 0} }; #ifdef SHIFTJIS_CP932 if (!cp932inv_f && is_ibmext_in_sjis(c2)){ -#if 0 - extern const unsigned short shiftjis_cp932[3][189]; -#endif val = shiftjis_cp932[c2 - CP932_TABLE_BEGIN][c1 - 0x40]; if (val){ c2 = val >> 8; @@ -3136,9 +3160,6 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) } if (cp932inv_f && CP932INV_TABLE_BEGIN <= c2 && c2 <= CP932INV_TABLE_END){ -#if 0 - extern const unsigned short cp932inv[2][189]; -#endif nkf_char c = cp932inv[c2 - CP932INV_TABLE_BEGIN][c1 - 0x40]; if (c){ c2 = c >> 8; @@ -3148,9 +3169,6 @@ nkf_char s2e_conv(nkf_char c2, nkf_char c1, nkf_char *p2, nkf_char *p1) #endif /* SHIFTJIS_CP932 */ #ifdef X0212_ENABLE if (!x0213_f && is_ibmext_in_sjis(c2)){ -#if 0 - extern const unsigned short shiftjis_x0212[3][189]; -#endif val = shiftjis_x0212[c2 - 0xfa][c1 - 0x40]; if (val){ if (val > 0x7FFF){ @@ -3481,14 +3499,6 @@ nkf_char w_iconv32(nkf_char c2, nkf_char c1, nkf_char c0) nkf_char unicode_to_jis_common(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char *p2, nkf_char *p1) { -#if 0 - extern const unsigned short *const utf8_to_euc_2bytes[]; - extern const unsigned short *const utf8_to_euc_2bytes_ms[]; - extern const unsigned short *const utf8_to_euc_2bytes_932[]; - extern const unsigned short *const *const utf8_to_euc_3bytes[]; - extern const unsigned short *const *const utf8_to_euc_3bytes_ms[]; - extern const unsigned short *const *const utf8_to_euc_3bytes_932[]; -#endif const unsigned short *const *pp; const unsigned short *const *const *ppp; static const int no_best_fit_chars_table_C2[] = @@ -3538,11 +3548,27 @@ nkf_char unicode_to_jis_common(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char * } }else if(ms_ucs_map_f == UCS_MAP_MS){ if(c2 == 0xC2 && no_best_fit_chars_table_C2_ms[c1&0x3F]) return 1; + }else if(ms_ucs_map_f == UCS_MAP_CP10001){ + switch(c2){ + case 0xC2: + switch(c1){ + case 0xA2: + case 0xA3: + case 0xA5: + case 0xA6: + case 0xAC: + case 0xAF: + case 0xB8: + return 1; + } + break; + } } } pp = ms_ucs_map_f == UCS_MAP_CP932 ? utf8_to_euc_2bytes_932 : ms_ucs_map_f == UCS_MAP_MS ? utf8_to_euc_2bytes_ms : + ms_ucs_map_f == UCS_MAP_CP10001 ? utf8_to_euc_2bytes_mac : utf8_to_euc_2bytes; ret = w_iconv_common(c2, c1, pp, sizeof_utf8_to_euc_2bytes, p2, p1); }else if(c0 < 0xF0){ @@ -3565,6 +3591,19 @@ nkf_char unicode_to_jis_common(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char * if(c1 == 0x80 || c0 == 0x9C) return 1; break; } + }else if(ms_ucs_map_f == UCS_MAP_CP10001){ + switch(c2){ + case 0xE3: + switch(c1){ + case 0x82: + if(c0 == 0x94) return 1; + break; + case 0x83: + if(c0 == 0xBB) return 1; + break; + } + break; + } }else{ switch(c2){ case 0xE2: @@ -3596,8 +3635,10 @@ nkf_char unicode_to_jis_common(nkf_char c2, nkf_char c1, nkf_char c0, nkf_char * ppp = ms_ucs_map_f == UCS_MAP_CP932 ? utf8_to_euc_3bytes_932 : ms_ucs_map_f == UCS_MAP_MS ? utf8_to_euc_3bytes_ms : + ms_ucs_map_f == UCS_MAP_CP10001 ? utf8_to_euc_3bytes_mac : utf8_to_euc_3bytes; ret = w_iconv_common(c1, c0, ppp[c2 - 0xE0], sizeof_utf8_to_euc_C2, p2, p1); +// fprintf(stderr, "wret: %X %X %X -> %X %X\n",c2,c1,c0,*p2,*p1,ret); }else return -1; #ifdef SHIFTJIS_CP932 if (!ret && !cp932inv_f && is_eucg3(*p2)) { @@ -3739,15 +3780,17 @@ void encode_fallback_subchar(nkf_char c) #ifdef UTF8_OUTPUT_ENABLE nkf_char e2w_conv(nkf_char c2, nkf_char c1) { -#if 0 - extern const unsigned short euc_to_utf8_1byte[]; - extern const unsigned short *const euc_to_utf8_2bytes[]; - extern const unsigned short *const euc_to_utf8_2bytes_ms[]; - extern const unsigned short *const x0212_to_utf8_2bytes[]; -#endif const unsigned short *p; if (c2 == X0201) { + if (ms_ucs_map_f == UCS_MAP_CP10001) { + switch (c1) { + case 0x20: + return 0xA0; + case 0x7D: + return 0xA9; + } + } p = euc_to_utf8_1byte; #ifdef X0212_ENABLE } else if (is_eucg3(c2)){ @@ -3764,7 +3807,10 @@ nkf_char e2w_conv(nkf_char c2, nkf_char c1) c2 &= 0x7f; c2 = (c2&0x7f) - 0x21; if (0<=c2 && c2> 8; @@ -4539,6 +4579,10 @@ void z_conv(nkf_char c2, nkf_char c1) /* if (c2) c1 &= 0x7f; assertion */ + if (c2 == X0201 && (c1 == 0x20 || c1 == 0x7D || c1 == 0x7E)) { + (*o_zconv)(c2,c1); + return; + } if (x0201_f && z_prev2==X0201) { /* X0201 */ if (c1==(0xde&0x7f)) { /* $BByE@(B */ z_prev2=0; @@ -4942,15 +4986,20 @@ void set_input_codename(char *codename) void print_guessed_code(char *filename) { char *codename = "BINARY"; + char *str_crmode = NULL; if (!is_inputcode_mixed) { if (strcmp(input_codename, "") == 0) { codename = "ASCII"; } else { codename = input_codename; } + if (crmode_f == CR) str_crmode = "CR"; + else if (crmode_f == NL) str_crmode = "LF"; + else if (crmode_f == CRLF) str_crmode = "CRLF"; } if (filename != NULL) printf("%s:", filename); - printf("%s\n", codename); + if (str_crmode != NULL) printf("%s (%s)\n", codename, str_crmode); + else printf("%s\n", codename); } #endif /*WIN32DLL*/ @@ -5068,9 +5117,6 @@ nkf_char nfc_getc(FILE *f) int i=0, j, k=1, lower, upper; nkf_char buf[9]; const nkf_nfchar *array; -#if 0 - extern const struct normalization_pair normalization_table[]; -#endif buf[i] = (*g)(f); while (k > 0 && ((buf[i] & 0xc0) != 0x80)){ @@ -5437,7 +5483,7 @@ void open_mime(nkf_char mode) int i; int j; p = mime_pattern[0]; - for(i=0;mime_encode[i];i++) { + for(i=0;mime_pattern[i];i++) { if (mode == mime_encode[i]) { p = mime_pattern[i]; break; @@ -5643,10 +5689,21 @@ void mime_putc(nkf_char c) if (mimeout_mode=='Q') { if (c <= DEL && (output_mode==ASCII ||output_mode == ISO8859_1 ) ) { - if (c <= SPACE) { + if (c == CR || c == NL) { + close_mime(); + (*o_mputc)(c); + base64_count = 0; + return; + } else if (c <= SPACE) { close_mime(); - (*o_mputc)(SPACE); - base64_count++; + if (base64_count > 70) { + (*o_mputc)(NL); + base64_count = 0; + } + if (!nkf_isblank(c)) { + (*o_mputc)(SPACE); + base64_count++; + } } (*o_mputc)(c); base64_count++; @@ -5678,7 +5735,8 @@ void mime_putc(nkf_char c) mimeout_buf_count = 1; }else{ if (base64_count > 1 - && base64_count + mimeout_buf_count > 76){ + && base64_count + mimeout_buf_count > 76 + && mimeout_buf[0] != CR && mimeout_buf[0] != NL){ (*o_mputc)(NL); base64_count = 0; if (!nkf_isspace(mimeout_buf[0])){ diff --git a/ext/nkf/nkf-utf8/utf8tbl.c b/ext/nkf/nkf-utf8/utf8tbl.c index e43ad553d6..fb6c3b7362 100644 --- a/ext/nkf/nkf-utf8/utf8tbl.c +++ b/ext/nkf/nkf-utf8/utf8tbl.c @@ -201,6 +201,20 @@ const unsigned short euc_to_utf8_AC[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short euc_to_utf8_AC_mac[] = { + 0x2664, 0x2667, 0x2661, 0x2662, 0x2660, 0x2663, 0x2665, + 0x2666, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0x3020, 0x260E, 0x3004, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0x261E, 0x261C, 0x261D, 0x261F, 0x21C6, 0x21C4, 0x21C5, + 0, 0x21E8, 0x21E6, 0x21E7, 0x21E9, 0x2192, 0x2190, 0x2191, + 0x2193, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short euc_to_utf8_AD[] = { 0x2460, 0x2461, 0x2462, 0x2463, 0x2464, 0x2465, 0x2466, 0x2467, 0x2468, 0x2469, 0x246A, 0x246B, 0x246C, 0x246D, 0x246E, @@ -215,6 +229,20 @@ const unsigned short euc_to_utf8_AD[] = { 0x2252, 0x2261, 0x222B, 0x222E, 0x2211, 0x221A, 0x22A5, 0x2220, 0x221F, 0x22BF, 0x2235, 0x2229, 0x222A, 0, 0x3299, }; +const unsigned short euc_to_utf8_AD_mac[] = { + 0x65E5, 0x6708, 0x706B, 0x6C34, 0x6728, 0x91D1, 0x571F, + 0x796D, 0x795D, 0x81EA, 0x81F3, 0x3239, 0x547C, 0x3231, 0x8CC7, + 0x540D, 0x3232, 0x5B66, 0x8CA1, 0x793E, 0x7279, 0x76E3, 0x4F01, + 0x5354, 0x52B4, 0x2165, 0x2166, 0x2167, 0x2168, 0x2169, 0, + 0x3349, 0x3314, 0x3322, 0x334D, 0x3318, 0x3327, 0x3303, 0x3336, + 0x3351, 0x3357, 0x330D, 0x3326, 0x3323, 0x332B, 0x334A, 0x333B, + 0x339C, 0x339D, 0x339E, 0x338E, 0x338F, 0x33C4, 0x33A1, 0, + 0, 0, 0, 0, 0, 0, 0, 0x337B, + 0x301D, 0x301F, 0x2116, 0x33CD, 0x2121, 0x32A4, 0x32A5, 0x32A6, + 0x32A7, 0x32A8, 0x3231, 0x3232, 0x3239, 0x337E, 0x337D, 0x337C, + 0x2252, 0x5927, 0x5C0F, 0x32A4, 0x32A5, 0x32A6, 0x32A7, 0x32A8, + 0x533B, 0x8CA1, 0x512A, 0x52B4, 0x5370, 0x63A7, 0x79D8, +}; const unsigned short euc_to_utf8_AE[] = { 0x3349, 0x3322, 0x334D, 0x3314, 0x3316, 0x3305, 0x3333, 0x334E, 0x3303, 0x3336, 0x3318, 0x3315, 0x3327, 0x3351, 0x334A, @@ -2346,6 +2374,33 @@ const unsigned short *const euc_to_utf8_2bytes_ms[] = { 0, euc_to_utf8_F9, euc_to_utf8_FA, euc_to_utf8_FB, euc_to_utf8_FC_ms, 0, 0, }; +/* CP10001 */ +const unsigned short *const euc_to_utf8_2bytes_mac[] = { + euc_to_utf8_A1_ms, euc_to_utf8_A2_ms, euc_to_utf8_A3, + euc_to_utf8_A4, euc_to_utf8_A5, euc_to_utf8_A6, euc_to_utf8_A7, + euc_to_utf8_A8, euc_to_utf8_A9, euc_to_utf8_AA, euc_to_utf8_AB, + euc_to_utf8_AC_mac, euc_to_utf8_AD_mac, euc_to_utf8_AE, euc_to_utf8_AF, + euc_to_utf8_B0, euc_to_utf8_B1, euc_to_utf8_B2, euc_to_utf8_B3, + euc_to_utf8_B4, euc_to_utf8_B5, euc_to_utf8_B6, euc_to_utf8_B7, + euc_to_utf8_B8, euc_to_utf8_B9, euc_to_utf8_BA, euc_to_utf8_BB, + euc_to_utf8_BC, euc_to_utf8_BD, euc_to_utf8_BE, euc_to_utf8_BF, + euc_to_utf8_C0, euc_to_utf8_C1, euc_to_utf8_C2, euc_to_utf8_C3, + euc_to_utf8_C4, euc_to_utf8_C5, euc_to_utf8_C6, euc_to_utf8_C7, + euc_to_utf8_C8, euc_to_utf8_C9, euc_to_utf8_CA, euc_to_utf8_CB, + euc_to_utf8_CC, euc_to_utf8_CD, euc_to_utf8_CE, euc_to_utf8_CF, + euc_to_utf8_D0, euc_to_utf8_D1, euc_to_utf8_D2, euc_to_utf8_D3, + euc_to_utf8_D4, euc_to_utf8_D5, euc_to_utf8_D6, euc_to_utf8_D7, + euc_to_utf8_D8, euc_to_utf8_D9, euc_to_utf8_DA, euc_to_utf8_DB, + euc_to_utf8_DC, euc_to_utf8_DD, euc_to_utf8_DE, euc_to_utf8_DF, + euc_to_utf8_E0, euc_to_utf8_E1, euc_to_utf8_E2, euc_to_utf8_E3, + euc_to_utf8_E4, euc_to_utf8_E5, euc_to_utf8_E6, euc_to_utf8_E7, + euc_to_utf8_E8, euc_to_utf8_E9, euc_to_utf8_EA, euc_to_utf8_EB, + euc_to_utf8_EC, euc_to_utf8_ED, euc_to_utf8_EE, euc_to_utf8_EF, + euc_to_utf8_F0, euc_to_utf8_F1, euc_to_utf8_F2, euc_to_utf8_F3, + euc_to_utf8_F4, euc_to_utf8_F5, 0, 0, + 0, euc_to_utf8_F9, euc_to_utf8_FA, euc_to_utf8_FB, + euc_to_utf8_FC_ms, 0, 0, +}; #ifdef X0212_ENABLE const unsigned short *const x0212_to_utf8_2bytes[] = { @@ -2397,6 +2452,16 @@ const unsigned short utf8_to_euc_C2_ms[] = { 0x216B, 0x215E, 0, 0, 0x212D, 0, 0x2279, 0, 0xA231, 0, 0xA26B, 0, 0, 0, 0, 0xA244, }; +const unsigned short utf8_to_euc_C2_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0x0220, 0xA242, 0x2171, 0x2172, 0xA270, 0x5C, 0xA243, 0x2178, + 0x212F, 0x027D, 0xA26C, 0, 0x224C, 0, 0xA26E, 0xA234, + 0x216B, 0x215E, 0, 0, 0x212D, 0, 0x2279, 0, + 0xA231, 0, 0xA26B, 0, 0, 0, 0, 0xA244, +}; const unsigned short utf8_to_euc_C2_932[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2547,6 +2612,16 @@ const unsigned short utf8_to_euc_E284[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E284_mac[] = { + 0, 0, 0, 0x216E, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0x2B7B, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0x2B7D, 0x027E, 0, 0, 0, 0, 0, + 0, 0, 0, 0x2272, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E285[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2557,6 +2632,16 @@ const unsigned short utf8_to_euc_E285[] = { 0xF373, 0xF374, 0xF375, 0xF376, 0xF377, 0xF378, 0xF379, 0xF37A, 0xF37B, 0xF37C, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E285_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0x2A21, 0x2A22, 0x2A23, 0x2A24, 0x2A25, 0x2A26, 0x2A27, 0x2A28, + 0x2A29, 0x2A2A, 0, 0, 0, 0, 0, 0, + 0x2A35, 0x2A36, 0x2A37, 0x2A38, 0x2A39, 0x2A3A, 0x2A3B, 0x2A3C, + 0x2A3D, 0x2A3E, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E286[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2597,6 +2682,16 @@ const unsigned short utf8_to_euc_E288_932[] = { 0, 0, 0, 0, 0x2168, 0x2268, 0, 0, 0, 0, 0, 0, 0, 0x2266, 0, 0, }; +const unsigned short utf8_to_euc_E288_mac[] = { + 0x224F, 0, 0x225F, 0x2250, 0, 0, 0, 0x2260, + 0x223A, 0, 0, 0x223B, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0x2265, 0, 0, 0x2267, 0x2167, 0x2F22, + 0x225C, 0, 0, 0, 0, 0x2142, 0, 0x224A, + 0x224B, 0x2241, 0x2240, 0x2269, 0x226A, 0, 0x2F21, 0, + 0, 0, 0, 0, 0x2168, 0x2268, 0, 0, + 0, 0, 0, 0, 0, 0x2266, 0, 0, +}; const unsigned short utf8_to_euc_E289[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2617,6 +2712,16 @@ const unsigned short utf8_to_euc_E28A[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D79, }; +const unsigned short utf8_to_euc_E28A_mac[] = { + 0, 0, 0x223E, 0x223F, 0, 0, 0x223C, 0x223D, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0x225D, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0x2F23, +}; const unsigned short utf8_to_euc_E28C[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2637,6 +2742,16 @@ const unsigned short utf8_to_euc_E291[] = { 0x2D31, 0x2D32, 0x2D33, 0x2D34, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E291_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0x2921, 0x2922, 0x2923, 0x2924, 0x2925, 0x2926, 0x2927, 0x2928, + 0x2929, 0x292A, 0x292B, 0x292C, 0x292D, 0x292E, 0x292F, 0x2930, + 0x2931, 0x2932, 0x2933, 0x2934, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E294[] = { 0x2821, 0x282C, 0x2822, 0x282D, 0, 0, 0, 0, 0, 0, 0, 0, 0x2823, 0, 0, 0x282E, @@ -2767,6 +2882,16 @@ const unsigned short utf8_to_euc_E388[] = { 0, 0x2D6A, 0x2D6B, 0, 0, 0, 0, 0, 0, 0x2D6C, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E388_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0x2D2E, 0x2D31, 0, 0, 0, 0, 0, + 0, 0x2D2C, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E38A[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, @@ -2777,6 +2902,16 @@ const unsigned short utf8_to_euc_E38A[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E38A_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0x2D73, 0x2D74, 0x2D75, 0x2D76, + 0x2D77, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E38C[] = { 0, 0, 0, 0x2D46, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D4A, 0, 0, @@ -2787,6 +2922,16 @@ const unsigned short utf8_to_euc_E38C[] = { 0, 0, 0, 0, 0, 0, 0x2D47, 0, 0, 0, 0, 0x2D4F, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E38C_mac[] = { + 0, 0, 0, 0x2E29, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0x2E32, 0, 0, + 0, 0, 0, 0, 0x2E24, 0, 0, 0, + 0x2E2B, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0x2E22, 0x2E34, 0, 0, 0x2E35, 0x2E2D, + 0, 0, 0, 0x2E37, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0x2E2A, 0, + 0, 0, 0, 0x2E36, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E38D[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D40, 0x2D4E, 0, 0, 0x2D43, 0, 0, @@ -2797,6 +2942,16 @@ const unsigned short utf8_to_euc_E38D[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D5F, 0x2D6F, 0x2D6E, 0x2D6D, 0, }; +const unsigned short utf8_to_euc_E38D_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0x2E21, 0x2E2F, 0, 0, 0x2E23, 0, 0, + 0, 0x2E2E, 0, 0, 0, 0, 0, 0x2E31, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0x2E6A, 0x2E69, 0x2E68, 0x2E67, 0, +}; const unsigned short utf8_to_euc_E38E[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D53, 0x2D54, @@ -2807,6 +2962,16 @@ const unsigned short utf8_to_euc_E38E[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E38E_mac[] = { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0x2B2B, 0x2B2D, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0x2B21, 0x2B23, 0x2B29, 0, + 0, 0x2B27, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E38F[] = { 0, 0, 0, 0, 0x2D55, 0, 0, 0, 0, 0, 0, 0, 0, 0x2D63, 0, 0, @@ -2817,6 +2982,16 @@ const unsigned short utf8_to_euc_E38F[] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short utf8_to_euc_E38F_mac[] = { + 0, 0, 0, 0, 0x2B2E, 0, 0, 0, + 0, 0, 0, 0, 0, 0x2B7C, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, +}; const unsigned short utf8_to_euc_E4B8[] = { 0x306C, 0x437A, 0xB021, 0x3C37, 0xB022, 0xB023, 0, 0x4B7C, 0x3E66, 0x3B30, 0x3E65, 0x323C, 0xB024, 0x4954, 0x4D3F, 0, @@ -6171,6 +6346,24 @@ const unsigned short *const utf8_to_euc_E2_932[] = { 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short *const utf8_to_euc_E2_mac[] = { + utf8_to_euc_E280_932, 0, 0, 0, + utf8_to_euc_E284_mac, utf8_to_euc_E285_mac, utf8_to_euc_E286, utf8_to_euc_E287, + utf8_to_euc_E288_mac, utf8_to_euc_E289, utf8_to_euc_E28A_mac, 0, + utf8_to_euc_E28C, 0, 0, 0, + 0, utf8_to_euc_E291_mac, 0, 0, + utf8_to_euc_E294, utf8_to_euc_E295, utf8_to_euc_E296, utf8_to_euc_E297, + utf8_to_euc_E298, utf8_to_euc_E299, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, +}; const unsigned short *const utf8_to_euc_E3[] = { utf8_to_euc_E380, utf8_to_euc_E381, utf8_to_euc_E382, utf8_to_euc_E383, 0, 0, 0, 0, @@ -6207,6 +6400,24 @@ const unsigned short *const utf8_to_euc_E3_932[] = { 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short *const utf8_to_euc_E3_mac[] = { + utf8_to_euc_E380_932, utf8_to_euc_E381, utf8_to_euc_E382_932, utf8_to_euc_E383, + 0, 0, 0, 0, + utf8_to_euc_E388_mac, 0, utf8_to_euc_E38A_mac, 0, + utf8_to_euc_E38C_mac, utf8_to_euc_E38D_mac, utf8_to_euc_E38E_mac, utf8_to_euc_E38F_mac, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, +}; const unsigned short *const utf8_to_euc_E4[] = { 0, 0, 0, 0, 0, 0, 0, 0, @@ -6441,6 +6652,36 @@ const unsigned short *const utf8_to_euc_2bytes_932[] = { 0, 0, 0, 0, 0, 0, 0, 0, }; +const unsigned short *const utf8_to_euc_2bytes_mac[] = { + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, utf8_to_euc_C2_mac, utf8_to_euc_C3, + utf8_to_euc_C4, utf8_to_euc_C5, 0, utf8_to_euc_C7, + 0, 0, 0, utf8_to_euc_CB, + 0, 0, utf8_to_euc_CE, utf8_to_euc_CF, + utf8_to_euc_D0, utf8_to_euc_D1, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, + 0, 0, 0, 0, +}; const unsigned short *const *const utf8_to_euc_3bytes[] = { 0, 0, utf8_to_euc_E2, utf8_to_euc_E3, utf8_to_euc_E4, utf8_to_euc_E5, utf8_to_euc_E6, utf8_to_euc_E7, @@ -6459,6 +6700,12 @@ const unsigned short *const *const utf8_to_euc_3bytes_932[] = { utf8_to_euc_E8, utf8_to_euc_E9, 0, 0, 0, 0, 0, utf8_to_euc_EF_ms, }; +const unsigned short *const *const utf8_to_euc_3bytes_mac[] = { + 0, 0, utf8_to_euc_E2_mac, utf8_to_euc_E3_mac, + utf8_to_euc_E4, utf8_to_euc_E5, utf8_to_euc_E6, utf8_to_euc_E7, + utf8_to_euc_E8, utf8_to_euc_E9, 0, 0, + 0, 0, 0, utf8_to_euc_EF_ms, +}; #ifdef UNICODE_NORMALIZATION diff --git a/ext/nkf/nkf-utf8/utf8tbl.h b/ext/nkf/nkf-utf8/utf8tbl.h index 1f40f0b363..29413d4fac 100644 --- a/ext/nkf/nkf-utf8/utf8tbl.h +++ b/ext/nkf/nkf-utf8/utf8tbl.h @@ -5,6 +5,7 @@ extern const unsigned short euc_to_utf8_1byte[]; extern const unsigned short *const euc_to_utf8_2bytes[]; extern const unsigned short *const euc_to_utf8_2bytes_ms[]; +extern const unsigned short *const euc_to_utf8_2bytes_mac[]; extern const unsigned short *const x0212_to_utf8_2bytes[]; #endif /* UTF8_OUTPUT_ENABLE */ @@ -12,9 +13,11 @@ extern const unsigned short *const x0212_to_utf8_2bytes[]; extern const unsigned short *const utf8_to_euc_2bytes[]; extern const unsigned short *const utf8_to_euc_2bytes_ms[]; extern const unsigned short *const utf8_to_euc_2bytes_932[]; +extern const unsigned short *const utf8_to_euc_2bytes_mac[]; extern const unsigned short *const *const utf8_to_euc_3bytes[]; extern const unsigned short *const *const utf8_to_euc_3bytes_ms[]; extern const unsigned short *const *const utf8_to_euc_3bytes_932[]; +extern const unsigned short *const *const utf8_to_euc_3bytes_mac[]; #endif /* UTF8_INPUT_ENABLE */ #ifdef UNICODE_NORMALIZATION diff --git a/lib/json.rb b/lib/json.rb new file mode 100644 index 0000000000..1700e0654b --- /dev/null +++ b/lib/json.rb @@ -0,0 +1,210 @@ +require 'json/common' +# = json - JSON for Ruby +# +# == Description +# +# This is a implementation of the JSON specification according to RFC 4627 +# (http://www.ietf.org/rfc/rfc4627.txt). Starting from version 1.0.0 on there +# will be two variants available: +# +# * A pure ruby variant, that relies on the iconv and the stringscan +# extensions, which are both part of the ruby standard library. +# * The quite a bit faster C extension variant, which is in parts implemented +# in C and comes with its own unicode conversion functions and a parser +# generated by the ragel state machine compiler +# (http://www.cs.queensu.ca/~thurston/ragel). +# +# Both variants of the JSON generator escape all non-ASCII an control +# characters with \uXXXX escape sequences, and support UTF-16 surrogate pairs +# in order to be able to generate the whole range of unicode code points. This +# means that generated JSON text is encoded as UTF-8 (because ASCII is a subset +# of UTF-8) and at the same time avoids decoding problems for receiving +# endpoints, that don't expect UTF-8 encoded texts. On the negative side this +# may lead to a bit longer strings than necessarry. +# +# All strings, that are to be encoded as JSON strings, should be UTF-8 byte +# sequences on the Ruby side. To encode raw binary strings, that aren't UTF-8 +# encoded, please use the to_json_raw_object method of String (which produces +# an object, that contains a byte array) and decode the result on the receiving +# endpoint. +# +# == Author +# +# Florian Frank +# +# == License +# +# This software is distributed under the same license as Ruby itself, see +# http://www.ruby-lang.org/en/LICENSE.txt. +# +# == Download +# +# The latest version of this library can be downloaded at +# +# * http://rubyforge.org/frs?group_id=953 +# +# Online Documentation should be located at +# +# * http://json.rubyforge.org +# +# == Speed Comparisons +# +# I have created some benchmark results (see the benchmarks subdir of the +# package) for the JSON-Parser to estimate the speed up in the C extension: +# +# JSON::Pure::Parser:: 28.90 calls/second +# JSON::Ext::Parser:: 505.50 calls/second +# +# This is ca. 17.5 times the speed of the pure Ruby implementation. +# +# I have benchmarked the JSON-Generator as well. This generates a few more +# values, because there are different modes, that also influence the achieved +# speed: +# +# * JSON::Pure::Generator: +# generate:: 35.06 calls/second +# pretty_generate:: 34.00 calls/second +# fast_generate:: 41.06 calls/second +# +# * JSON::Ext::Generator: +# generate:: 492.11 calls/second +# pretty_generate:: 348.85 calls/second +# fast_generate:: 541.60 calls/second +# +# * Speedup Ext/Pure: +# generate safe:: 14.0 times +# generate pretty:: 10.3 times +# generate fast:: 13.2 times +# +# The rails framework includes a generator as well, also it seems to be rather +# slow: I measured only 23.87 calls/second which is slower than any of my pure +# generator results. Here a comparison of the different speedups with the Rails +# measurement as the divisor: +# +# * Speedup Pure/Rails: +# generate safe:: 1.5 times +# generate pretty:: 1.4 times +# generate fast:: 1.7 times +# +# * Speedup Ext/Rails: +# generate safe:: 20.6 times +# generate pretty:: 14.6 times +# generate fast:: 22.7 times +# +# To achieve the fastest JSON text output, you can use the +# fast_generate/fast_unparse methods. Beware, that this will disable the +# checking for circular Ruby data structures, which may cause JSON to go into +# an infinite loop. +# +# == Examples +# +# To create a JSON text from a ruby data structure, you +# can call JSON.generate (or JSON.unparse) like that: +# +# json = JSON.generate [1, 2, {"a"=>3.141}, false, true, nil, 4..10] +# # => "[1,2,{\"a\":3.141},false,true,null,\"4..10\"]" +# +# It's also possible to call the #to_json method directly. +# +# json = [1, 2, {"a"=>3.141}, false, true, nil, 4..10].to_json +# # => "[1,2,{\"a\":3.141},false,true,null,\"4..10\"]" +# +# To create a valid JSON text you have to make sure, that the output is +# embedded in either a JSON array [] or a JSON object {}. The easiest way to do +# this, is by putting your values in a Ruby Array or Hash instance. +# +# To get back a ruby data structure from a JSON text, you have to call +# JSON.parse on it: +# +# JSON.parse json +# # => [1, 2, {"a"=>3.141}, false, true, nil, "4..10"] +# +# Note, that the range from the original data structure is a simple +# string now. The reason for this is, that JSON doesn't support ranges +# or arbitrary classes. In this case the json library falls back to call +# Object#to_json, which is the same as #to_s.to_json. +# +# It's possible to extend JSON to support serialization of arbitrary classes by +# simply implementing a more specialized version of the #to_json method, that +# should return a JSON object (a hash converted to JSON with #to_json) +# like this (don't forget the *a for all the arguments): +# +# class Range +# def to_json(*a) +# { +# 'json_class' => self.class.name, # = 'Range' +# 'data' => [ first, last, exclude_end? ] +# }.to_json(*a) +# end +# end +# +# The hash key 'json_class' is the class, that will be asked to deserialize the +# JSON representation later. In this case it's 'Range', but any namespace of +# the form 'A::B' or '::A::B' will do. All other keys are arbitrary and can be +# used to store the necessary data to configure the object to be deserialized. +# +# If a the key 'json_class' is found in a JSON object, the JSON parser checks +# if the given class responds to the json_create class method. If so, it is +# called with the JSON object converted to a Ruby hash. So a range can +# be deserialized by implementing Range.json_create like this: +# +# class Range +# def self.json_create(o) +# new(*o['data']) +# end +# end +# +# Now it possible to serialize/deserialize ranges as well: +# +# json = JSON.generate [1, 2, {"a"=>3.141}, false, true, nil, 4..10] +# # => "[1,2,{\"a\":3.141},false,true,null,{\"json_class\":\"Range\",\"data\":[4,10,false]}]" +# JSON.parse json +# # => [1, 2, {"a"=>3.141}, false, true, nil, 4..10] +# +# JSON.generate always creates the shortest possible string representation of a +# ruby data structure in one line. This good for data storage or network +# protocols, but not so good for humans to read. Fortunately there's also +# JSON.pretty_generate (or JSON.pretty_generate) that creates a more +# readable output: +# +# puts JSON.pretty_generate([1, 2, {"a"=>3.141}, false, true, nil, 4..10]) +# [ +# 1, +# 2, +# { +# "a": 3.141 +# }, +# false, +# true, +# null, +# { +# "json_class": "Range", +# "data": [ +# 4, +# 10, +# false +# ] +# } +# ] +# +# There are also the methods Kernel#j for unparse, and Kernel#jj for +# pretty_unparse output to the console, that work analogous to Core Ruby's p +# and the pp library's pp methods. +# +# The script tools/server.rb contains a small example if you want to test, how +# receiving a JSON object from a webrick server in your browser with the +# javasript prototype library (http://www.prototypejs.org) works. +# +module JSON + require 'json/version' + + if VARIANT_BINARY + require 'json/ext' + else + begin + require 'json/ext' + rescue LoadError + require 'json/pure' + end + end +end diff --git a/lib/json/Array.xpm b/lib/json/Array.xpm new file mode 100644 index 0000000000..27c48011f9 --- /dev/null +++ b/lib/json/Array.xpm @@ -0,0 +1,21 @@ +/* XPM */ +static char * Array_xpm[] = { +"16 16 2 1", +" c None", +". c #000000", +" ", +" ", +" ", +" .......... ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" .......... ", +" ", +" ", +" "}; diff --git a/lib/json/FalseClass.xpm b/lib/json/FalseClass.xpm new file mode 100644 index 0000000000..25ce60832d --- /dev/null +++ b/lib/json/FalseClass.xpm @@ -0,0 +1,21 @@ +/* XPM */ +static char * False_xpm[] = { +"16 16 2 1", +" c None", +". c #FF0000", +" ", +" ", +" ", +" ...... ", +" . ", +" . ", +" . ", +" ...... ", +" . ", +" . ", +" . ", +" . ", +" . ", +" ", +" ", +" "}; diff --git a/lib/json/Hash.xpm b/lib/json/Hash.xpm new file mode 100644 index 0000000000..cd8f6f7b53 --- /dev/null +++ b/lib/json/Hash.xpm @@ -0,0 +1,21 @@ +/* XPM */ +static char * Hash_xpm[] = { +"16 16 2 1", +" c None", +". c #000000", +" ", +" ", +" ", +" . . ", +" . . ", +" . . ", +" ......... ", +" . . ", +" . . ", +" ......... ", +" . . ", +" . . ", +" . . ", +" ", +" ", +" "}; diff --git a/lib/json/Key.xpm b/lib/json/Key.xpm new file mode 100644 index 0000000000..9fd7281388 --- /dev/null +++ b/lib/json/Key.xpm @@ -0,0 +1,73 @@ +/* XPM */ +static char * Key_xpm[] = { +"16 16 54 1", +" c None", +". c #110007", +"+ c #0E0900", +"@ c #000013", +"# c #070600", +"$ c #F6F006", +"% c #ECE711", +"& c #E5EE00", +"* c #16021E", +"= c #120900", +"- c #EDF12B", +"; c #000033", +"> c #0F0000", +", c #FFFE03", +"' c #E6E500", +") c #16021B", +"! c #F7F502", +"~ c #000E00", +"{ c #130000", +"] c #FFF000", +"^ c #FFE711", +"/ c #140005", +"( c #190025", +"_ c #E9DD27", +": c #E7DC04", +"< c #FFEC09", +"[ c #FFE707", +"} c #FFDE10", +"| c #150021", +"1 c #160700", +"2 c #FAF60E", +"3 c #EFE301", +"4 c #FEF300", +"5 c #E7E000", +"6 c #FFFF08", +"7 c #0E0206", +"8 c #040000", +"9 c #03052E", +"0 c #041212", +"a c #070300", +"b c #F2E713", +"c c #F9DE13", +"d c #36091E", +"e c #00001C", +"f c #1F0010", +"g c #FFF500", +"h c #DEDE00", +"i c #050A00", +"j c #FAF14A", +"k c #F5F200", +"l c #040404", +"m c #1A0D00", +"n c #EDE43D", +"o c #ECE007", +" ", +" ", +" .+@ ", +" #$%&* ", +" =-;>,') ", +" >!~{]^/ ", +" (_:<[}| ", +" 1234567 ", +" 890abcd ", +" efghi ", +" >jkl ", +" mnol ", +" >kl ", +" ll ", +" ", +" "}; diff --git a/lib/json/NilClass.xpm b/lib/json/NilClass.xpm new file mode 100644 index 0000000000..3509f06c99 --- /dev/null +++ b/lib/json/NilClass.xpm @@ -0,0 +1,21 @@ +/* XPM */ +static char * False_xpm[] = { +"16 16 2 1", +" c None", +". c #000000", +" ", +" ", +" ", +" ... ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" . . ", +" ... ", +" ", +" ", +" "}; diff --git a/lib/json/Numeric.xpm b/lib/json/Numeric.xpm new file mode 100644 index 0000000000..e071e2ee9c --- /dev/null +++ b/lib/json/Numeric.xpm @@ -0,0 +1,28 @@ +/* XPM */ +static char * Numeric_xpm[] = { +"16 16 9 1", +" c None", +". c #FF0000", +"+ c #0000FF", +"@ c #0023DB", +"# c #00EA14", +"$ c #00FF00", +"% c #004FAF", +"& c #0028D6", +"* c #00F20C", +" ", +" ", +" ", +" ... +++@#$$$$ ", +" .+ %& $$ ", +" . + $ ", +" . + $$ ", +" . ++$$$$ ", +" . + $$ ", +" . + $ ", +" . + $ ", +" . + $ $$ ", +" .....++++*$$ ", +" ", +" ", +" "}; diff --git a/lib/json/String.xpm b/lib/json/String.xpm new file mode 100644 index 0000000000..f79a89cdc1 --- /dev/null +++ b/lib/json/String.xpm @@ -0,0 +1,96 @@ +/* XPM */ +static char * String_xpm[] = { +"16 16 77 1", +" c None", +". c #000000", +"+ c #040404", +"@ c #080806", +"# c #090606", +"$ c #EEEAE1", +"% c #E7E3DA", +"& c #E0DBD1", +"* c #D4B46F", +"= c #0C0906", +"- c #E3C072", +"; c #E4C072", +"> c #060505", +", c #0B0A08", +"' c #D5B264", +") c #D3AF5A", +"! c #080602", +"~ c #E1B863", +"{ c #DDB151", +"] c #DBAE4A", +"^ c #DDB152", +"/ c #DDB252", +"( c #070705", +"_ c #0C0A07", +": c #D3A33B", +"< c #020201", +"[ c #DAAA41", +"} c #040302", +"| c #E4D9BF", +"1 c #0B0907", +"2 c #030201", +"3 c #020200", +"4 c #C99115", +"5 c #080704", +"6 c #DBC8A2", +"7 c #E7D7B4", +"8 c #E0CD9E", +"9 c #080601", +"0 c #040400", +"a c #010100", +"b c #0B0B08", +"c c #DCBF83", +"d c #DCBC75", +"e c #DEB559", +"f c #040301", +"g c #BC8815", +"h c #120E07", +"i c #060402", +"j c #0A0804", +"k c #D4A747", +"l c #D6A12F", +"m c #0E0C05", +"n c #C8C1B0", +"o c #1D1B15", +"p c #D7AD51", +"q c #070502", +"r c #080804", +"s c #BC953B", +"t c #C4BDAD", +"u c #0B0807", +"v c #DBAC47", +"w c #1B150A", +"x c #B78A2C", +"y c #D8A83C", +"z c #D4A338", +"A c #0F0B03", +"B c #181105", +"C c #C59325", +"D c #C18E1F", +"E c #060600", +"F c #CC992D", +"G c #B98B25", +"H c #B3831F", +"I c #C08C1C", +"J c #060500", +"K c #0E0C03", +"L c #0D0A00", +" ", +" .+@# ", +" .$%&*= ", +" .-;>,')! ", +" .~. .{]. ", +" .^/. (_:< ", +" .[.}|$12 ", +" 345678}90 ", +" a2bcdefgh ", +" ijkl.mno ", +" false. + def parse(source, opts = {}) + JSON.parser.new(source, opts).parse + end + + # Unparse the Ruby data structure _obj_ into a single line JSON string and + # return it. _state_ is a JSON::State object, that can be used to configure + # the output further. + # + # It defaults to a state object, that creates the shortest possible JSON text + # in one line and only checks for circular data structures. If you are sure, + # that the objects don't contain any circles, you can set _state_ to nil, to + # disable these checks in order to create the JSON text faster. See also + # fast_generate. + def generate(obj, state = JSON.state.new) + obj.to_json(state) + end + + alias unparse generate + module_function :unparse + + # Unparse the Ruby data structure _obj_ into a single line JSON string and + # return it. This method disables the checks for circles in Ruby objects. + # + # *WARNING*: Be careful not to pass any Ruby data structures with circles as + # _obj_ argument, because this will cause JSON to go into an infinite loop. + def fast_generate(obj) + obj.to_json(nil) + end + + alias fast_unparse fast_generate + module_function :fast_unparse + + # Unparse the Ruby data structure _obj_ into a JSON string and return it. The + # returned string is a prettier form of the string returned by #unparse. + def pretty_generate(obj) + state = JSON.state.new( + :indent => ' ', + :space => ' ', + :object_nl => "\n", + :array_nl => "\n", + :check_circular => true + ) + obj.to_json(state) + end + + alias pretty_unparse pretty_generate + module_function :pretty_unparse +end + +module ::Kernel + # Outputs _objs_ to STDOUT as JSON strings in the shortest form, that is in + # one line. + def j(*objs) + objs.each do |obj| + puts JSON::generate(obj) + end + nil + end + + # Ouputs _objs_ to STDOUT as JSON strings in a pretty format, with + # indentation and over many lines. + def jj(*objs) + objs.each do |obj| + puts JSON::pretty_generate(obj) + end + nil + end + + # If object is string like parse the string and return the parsed result as a + # Ruby data structure. Otherwise generate a JSON text from the Ruby data + # structure object and return it. + def JSON(object) + if object.respond_to? :to_str + JSON.parse(object.to_str) + else + JSON.generate(object) + end + end +end + +class ::Class + # Returns true, if this class can be used to create an instance + # from a serialised JSON string. The class has to implement a class + # method _json_create_ that expects a hash as first parameter, which includes + # the required data. + def json_creatable? + respond_to?(:json_create) + end +end + # vim: set et sw=2 ts=2: diff --git a/lib/json/editor.rb b/lib/json/editor.rb new file mode 100644 index 0000000000..063df69533 --- /dev/null +++ b/lib/json/editor.rb @@ -0,0 +1,1295 @@ +# To use the GUI JSON editor, start the edit_json.rb executable script. It +# requires ruby-gtk to be installed. + +require 'gtk2' +require 'iconv' +require 'json' +require 'rbconfig' +require 'open-uri' + +module JSON + module Editor + include Gtk + + # Beginning of the editor window title + TITLE = 'JSON Editor'.freeze + + # Columns constants + ICON_COL, TYPE_COL, CONTENT_COL = 0, 1, 2 + + # JSON primitive types (Containers) + CONTAINER_TYPES = %w[Array Hash].sort + # All JSON primitive types + ALL_TYPES = (%w[TrueClass FalseClass Numeric String NilClass] + + CONTAINER_TYPES).sort + + # The Nodes necessary for the tree representation of a JSON document + ALL_NODES = (ALL_TYPES + %w[Key]).sort + + DEFAULT_DIALOG_KEY_PRESS_HANDLER = lambda do |dialog, event| + case event.keyval + when Gdk::Keyval::GDK_Return + dialog.response Dialog::RESPONSE_ACCEPT + when Gdk::Keyval::GDK_Escape + dialog.response Dialog::RESPONSE_REJECT + end + end + + # Returns the Gdk::Pixbuf of the icon named _name_ from the icon cache. + def Editor.fetch_icon(name) + @icon_cache ||= {} + unless @icon_cache.key?(name) + path = File.dirname(__FILE__) + @icon_cache[name] = Gdk::Pixbuf.new(File.join(path, name + '.xpm')) + end + @icon_cache[name] + end + + # Opens an error dialog on top of _window_ showing the error message + # _text_. + def Editor.error_dialog(window, text) + dialog = MessageDialog.new(window, Dialog::MODAL, + MessageDialog::ERROR, + MessageDialog::BUTTONS_CLOSE, text) + dialog.show_all + window.focus = dialog + dialog.run + rescue TypeError + dialog = MessageDialog.new(Editor.window, Dialog::MODAL, + MessageDialog::ERROR, + MessageDialog::BUTTONS_CLOSE, text) + dialog.show_all + window.focus = dialog + dialog.run + ensure + dialog.destroy if dialog + end + + # Opens a yes/no question dialog on top of _window_ showing the error + # message _text_. If yes was answered _true_ is returned, otherwise + # _false_. + def Editor.question_dialog(window, text) + dialog = MessageDialog.new(window, Dialog::MODAL, + MessageDialog::QUESTION, + MessageDialog::BUTTONS_YES_NO, text) + dialog.show_all + window.focus = dialog + dialog.run do |response| + return Gtk::Dialog::RESPONSE_YES === response + end + ensure + dialog.destroy if dialog + end + + # Convert the tree model starting from Gtk::TreeIter _iter_ into a Ruby + # data structure and return it. + def Editor.model2data(iter) + return nil if iter.nil? + case iter.type + when 'Hash' + hash = {} + iter.each { |c| hash[c.content] = Editor.model2data(c.first_child) } + hash + when 'Array' + array = Array.new(iter.n_children) + iter.each_with_index { |c, i| array[i] = Editor.model2data(c) } + array + when 'Key' + iter.content + when 'String' + iter.content + when 'Numeric' + content = iter.content + if /\./.match(content) + content.to_f + else + content.to_i + end + when 'TrueClass' + true + when 'FalseClass' + false + when 'NilClass' + nil + else + fail "Unknown type found in model: #{iter.type}" + end + end + + # Convert the Ruby data structure _data_ into tree model data for Gtk and + # returns the whole model. If the parameter _model_ wasn't given a new + # Gtk::TreeStore is created as the model. The _parent_ parameter specifies + # the parent node (iter, Gtk:TreeIter instance) to which the data is + # appended, alternativeley the result of the yielded block is used as iter. + def Editor.data2model(data, model = nil, parent = nil) + model ||= TreeStore.new(Gdk::Pixbuf, String, String) + iter = if block_given? + yield model + else + model.append(parent) + end + case data + when Hash + iter.type = 'Hash' + data.sort.each do |key, value| + pair_iter = model.append(iter) + pair_iter.type = 'Key' + pair_iter.content = key.to_s + Editor.data2model(value, model, pair_iter) + end + when Array + iter.type = 'Array' + data.each do |value| + Editor.data2model(value, model, iter) + end + when Numeric + iter.type = 'Numeric' + iter.content = data.to_s + when String, true, false, nil + iter.type = data.class.name + iter.content = data.nil? ? 'null' : data.to_s + else + iter.type = 'String' + iter.content = data.to_s + end + model + end + + # The Gtk::TreeIter class is reopened and some auxiliary methods are added. + class Gtk::TreeIter + include Enumerable + + # Traverse each of this Gtk::TreeIter instance's children + # and yield to them. + def each + n_children.times { |i| yield nth_child(i) } + end + + # Recursively traverse all nodes of this Gtk::TreeIter's subtree + # (including self) and yield to them. + def recursive_each(&block) + yield self + each do |i| + i.recursive_each(&block) + end + end + + # Remove the subtree of this Gtk::TreeIter instance from the + # model _model_. + def remove_subtree(model) + while current = first_child + model.remove(current) + end + end + + # Returns the type of this node. + def type + self[TYPE_COL] + end + + # Sets the type of this node to _value_. This implies setting + # the respective icon accordingly. + def type=(value) + self[TYPE_COL] = value + self[ICON_COL] = Editor.fetch_icon(value) + end + + # Returns the content of this node. + def content + self[CONTENT_COL] + end + + # Sets the content of this node to _value_. + def content=(value) + self[CONTENT_COL] = value + end + end + + # This module bundles some method, that can be used to create a menu. It + # should be included into the class in question. + module MenuExtension + include Gtk + + # Creates a Menu, that includes MenuExtension. _treeview_ is the + # Gtk::TreeView, on which it operates. + def initialize(treeview) + @treeview = treeview + @menu = Menu.new + end + + # Returns the Gtk::TreeView of this menu. + attr_reader :treeview + + # Returns the menu. + attr_reader :menu + + # Adds a Gtk::SeparatorMenuItem to this instance's #menu. + def add_separator + menu.append SeparatorMenuItem.new + end + + # Adds a Gtk::MenuItem to this instance's #menu. _label_ is the label + # string, _klass_ is the item type, and _callback_ is the procedure, that + # is called if the _item_ is activated. + def add_item(label, keyval = nil, klass = MenuItem, &callback) + label = "#{label} (C-#{keyval.chr})" if keyval + item = klass.new(label) + item.signal_connect(:activate, &callback) + if keyval + self.signal_connect(:'key-press-event') do |item, event| + if event.state & Gdk::Window::ModifierType::CONTROL_MASK != 0 and + event.keyval == keyval + callback.call item + end + end + end + menu.append item + item + end + + # This method should be implemented in subclasses to create the #menu of + # this instance. It has to be called after an instance of this class is + # created, to build the menu. + def create + raise NotImplementedError + end + + def method_missing(*a, &b) + treeview.__send__(*a, &b) + end + end + + # This class creates the popup menu, that opens when clicking onto the + # treeview. + class PopUpMenu + include MenuExtension + + # Change the type or content of the selected node. + def change_node(item) + if current = selection.selected + parent = current.parent + old_type, old_content = current.type, current.content + if ALL_TYPES.include?(old_type) + @clipboard_data = Editor.model2data(current) + type, content = ask_for_element(parent, current.type, + current.content) + if type + current.type, current.content = type, content + current.remove_subtree(model) + toplevel.display_status("Changed a node in tree.") + window.change + end + else + toplevel.display_status( + "Cannot change node of type #{old_type} in tree!") + end + end + end + + # Cut the selected node and its subtree, and save it into the + # clipboard. + def cut_node(item) + if current = selection.selected + if current and current.type == 'Key' + @clipboard_data = { + current.content => Editor.model2data(current.first_child) + } + else + @clipboard_data = Editor.model2data(current) + end + model.remove(current) + window.change + toplevel.display_status("Cut a node from tree.") + end + end + + # Copy the selected node and its subtree, and save it into the + # clipboard. + def copy_node(item) + if current = selection.selected + if current and current.type == 'Key' + @clipboard_data = { + current.content => Editor.model2data(current.first_child) + } + else + @clipboard_data = Editor.model2data(current) + end + window.change + toplevel.display_status("Copied a node from tree.") + end + end + + # Paste the data in the clipboard into the selected Array or Hash by + # appending it. + def paste_node_appending(item) + if current = selection.selected + if @clipboard_data + case current.type + when 'Array' + Editor.data2model(@clipboard_data, model, current) + expand_collapse(current) + when 'Hash' + if @clipboard_data.is_a? Hash + parent = current.parent + hash = Editor.model2data(current) + model.remove(current) + hash.update(@clipboard_data) + Editor.data2model(hash, model, parent) + if parent + expand_collapse(parent) + elsif @expanded + expand_all + end + window.change + else + toplevel.display_status( + "Cannot paste non-#{current.type} data into '#{current.type}'!") + end + else + toplevel.display_status( + "Cannot paste node below '#{current.type}'!") + end + else + toplevel.display_status("Nothing to paste in clipboard!") + end + else + toplevel.display_status("Append a node into the root first!") + end + end + + # Paste the data in the clipboard into the selected Array inserting it + # before the selected element. + def paste_node_inserting_before(item) + if current = selection.selected + if @clipboard_data + parent = current.parent or return + parent_type = parent.type + if parent_type == 'Array' + selected_index = parent.each_with_index do |c, i| + break i if c == current + end + Editor.data2model(@clipboard_data, model, parent) do |m| + m.insert_before(parent, current) + end + expand_collapse(current) + toplevel.display_status("Inserted an element to " + + "'#{parent_type}' before index #{selected_index}.") + window.change + else + toplevel.display_status( + "Cannot insert node below '#{parent_type}'!") + end + else + toplevel.display_status("Nothing to paste in clipboard!") + end + else + toplevel.display_status("Append a node into the root first!") + end + end + + # Append a new node to the selected Hash or Array. + def append_new_node(item) + if parent = selection.selected + parent_type = parent.type + case parent_type + when 'Hash' + key, type, content = ask_for_hash_pair(parent) + key or return + iter = create_node(parent, 'Key', key) + iter = create_node(iter, type, content) + toplevel.display_status( + "Added a (key, value)-pair to '#{parent_type}'.") + window.change + when 'Array' + type, content = ask_for_element(parent) + type or return + iter = create_node(parent, type, content) + window.change + toplevel.display_status("Appendend an element to '#{parent_type}'.") + else + toplevel.display_status("Cannot append to '#{parent_type}'!") + end + else + type, content = ask_for_element + type or return + iter = create_node(nil, type, content) + window.change + end + end + + # Insert a new node into an Array before the selected element. + def insert_new_node(item) + if current = selection.selected + parent = current.parent or return + parent_parent = parent.parent + parent_type = parent.type + if parent_type == 'Array' + selected_index = parent.each_with_index do |c, i| + break i if c == current + end + type, content = ask_for_element(parent) + type or return + iter = model.insert_before(parent, current) + iter.type, iter.content = type, content + toplevel.display_status("Inserted an element to " + + "'#{parent_type}' before index #{selected_index}.") + window.change + else + toplevel.display_status( + "Cannot insert node below '#{parent_type}'!") + end + else + toplevel.display_status("Append a node into the root first!") + end + end + + # Recursively collapse/expand a subtree starting from the selected node. + def collapse_expand(item) + if current = selection.selected + if row_expanded?(current.path) + collapse_row(current.path) + else + expand_row(current.path, true) + end + else + toplevel.display_status("Append a node into the root first!") + end + end + + # Create the menu. + def create + add_item("Change node", ?n, &method(:change_node)) + add_separator + add_item("Cut node", ?x, &method(:cut_node)) + add_item("Copy node", ?c, &method(:copy_node)) + add_item("Paste node (appending)", ?v, &method(:paste_node_appending)) + add_item("Paste node (inserting before)", ?V, + &method(:paste_node_inserting_before)) + add_separator + add_item("Append new node", ?a, &method(:append_new_node)) + add_item("Insert new node before", ?i, &method(:insert_new_node)) + add_separator + add_item("Collapse/Expand node (recursively)", ?C, + &method(:collapse_expand)) + + menu.show_all + signal_connect(:button_press_event) do |widget, event| + if event.kind_of? Gdk::EventButton and event.button == 3 + menu.popup(nil, nil, event.button, event.time) + end + end + signal_connect(:popup_menu) do + menu.popup(nil, nil, 0, Gdk::Event::CURRENT_TIME) + end + end + end + + # This class creates the File pulldown menu. + class FileMenu + include MenuExtension + + # Clear the model and filename, but ask to save the JSON document, if + # unsaved changes have occured. + def new(item) + window.clear + end + + # Open a file and load it into the editor. Ask to save the JSON document + # first, if unsaved changes have occured. + def open(item) + window.file_open + end + + def open_location(item) + window.location_open + end + + # Revert the current JSON document in the editor to the saved version. + def revert(item) + window.instance_eval do + @filename and file_open(@filename) + end + end + + # Save the current JSON document. + def save(item) + window.file_save + end + + # Save the current JSON document under the given filename. + def save_as(item) + window.file_save_as + end + + # Quit the editor, after asking to save any unsaved changes first. + def quit(item) + window.quit + end + + # Create the menu. + def create + title = MenuItem.new('File') + title.submenu = menu + add_item('New', &method(:new)) + add_item('Open', ?o, &method(:open)) + add_item('Open location', ?l, &method(:open_location)) + add_item('Revert', &method(:revert)) + add_separator + add_item('Save', ?s, &method(:save)) + add_item('Save As', ?S, &method(:save_as)) + add_separator + add_item('Quit', ?q, &method(:quit)) + title + end + end + + # This class creates the Edit pulldown menu. + class EditMenu + include MenuExtension + + # Find a string in all nodes' contents and select the found node in the + # treeview. + def find(item) + search = ask_for_find_term or return + begin + @search = Regexp.new(search) + rescue => e + Editor.error_dialog(self, "Evaluation of regex /#{search}/ failed: #{e}!") + return + end + iter = model.get_iter('0') + iter.recursive_each do |i| + if @iter + if @iter != i + next + else + @iter = nil + next + end + elsif @search.match(i[CONTENT_COL]) + set_cursor(i.path, nil, false) + @iter = i + break + end + end + end + + # Repeat the last search given by #find. + def find_again(item) + @search or return + iter = model.get_iter('0') + iter.recursive_each do |i| + if @iter + if @iter != i + next + else + @iter = nil + next + end + elsif @search.match(i[CONTENT_COL]) + set_cursor(i.path, nil, false) + @iter = i + break + end + end + end + + # Sort (Reverse sort) all elements of the selected array by the given + # expression. _x_ is the element in question. + def sort(item) + if current = selection.selected + if current.type == 'Array' + parent = current.parent + ary = Editor.model2data(current) + order, reverse = ask_for_order + order or return + begin + block = eval "lambda { |x| #{order} }" + if reverse + ary.sort! { |a,b| block[b] <=> block[a] } + else + ary.sort! { |a,b| block[a] <=> block[b] } + end + rescue => e + Editor.error_dialog(self, "Failed to sort Array with #{order}: #{e}!") + else + Editor.data2model(ary, model, parent) do |m| + m.insert_before(parent, current) + end + model.remove(current) + expand_collapse(parent) + window.change + toplevel.display_status("Array has been sorted.") + end + else + toplevel.display_status("Only Array nodes can be sorted!") + end + else + toplevel.display_status("Select an Array to sort first!") + end + end + + # Create the menu. + def create + title = MenuItem.new('Edit') + title.submenu = menu + add_item('Find', ?f, &method(:find)) + add_item('Find Again', ?g, &method(:find_again)) + add_separator + add_item('Sort', ?S, &method(:sort)) + title + end + end + + class OptionsMenu + include MenuExtension + + # Collapse/Expand all nodes by default. + def collapsed_nodes(item) + if expanded + self.expanded = false + collapse_all + else + self.expanded = true + expand_all + end + end + + # Toggle pretty saving mode on/off. + def pretty_saving(item) + @pretty_item.toggled + window.change + end + + attr_reader :pretty_item + + # Create the menu. + def create + title = MenuItem.new('Options') + title.submenu = menu + add_item('Collapsed nodes', nil, CheckMenuItem, &method(:collapsed_nodes)) + @pretty_item = add_item('Pretty saving', nil, CheckMenuItem, + &method(:pretty_saving)) + @pretty_item.active = true + window.unchange + title + end + end + + # This class inherits from Gtk::TreeView, to configure it and to add a lot + # of behaviour to it. + class JSONTreeView < Gtk::TreeView + include Gtk + + # Creates a JSONTreeView instance, the parameter _window_ is + # a MainWindow instance and used for self delegation. + def initialize(window) + @window = window + super(TreeStore.new(Gdk::Pixbuf, String, String)) + self.selection.mode = SELECTION_BROWSE + + @expanded = false + self.headers_visible = false + add_columns + add_popup_menu + end + + # Returns the MainWindow instance of this JSONTreeView. + attr_reader :window + + # Returns true, if nodes are autoexpanding, false otherwise. + attr_accessor :expanded + + private + + def add_columns + cell = CellRendererPixbuf.new + column = TreeViewColumn.new('Icon', cell, + 'pixbuf' => ICON_COL + ) + append_column(column) + + cell = CellRendererText.new + column = TreeViewColumn.new('Type', cell, + 'text' => TYPE_COL + ) + append_column(column) + + cell = CellRendererText.new + cell.editable = true + column = TreeViewColumn.new('Content', cell, + 'text' => CONTENT_COL + ) + cell.signal_connect(:edited, &method(:cell_edited)) + append_column(column) + end + + def unify_key(iter, key) + return unless iter.type == 'Key' + parent = iter.parent + if parent.any? { |c| c != iter and c.content == key } + old_key = key + i = 0 + begin + key = sprintf("%s.%d", old_key, i += 1) + end while parent.any? { |c| c != iter and c.content == key } + end + iter.content = key + end + + def cell_edited(cell, path, value) + iter = model.get_iter(path) + case iter.type + when 'Key' + unify_key(iter, value) + toplevel.display_status('Key has been changed.') + when 'FalseClass' + value.downcase! + if value == 'true' + iter.type, iter.content = 'TrueClass', 'true' + end + when 'TrueClass' + value.downcase! + if value == 'false' + iter.type, iter.content = 'FalseClass', 'false' + end + when 'Numeric' + iter.content = (Integer(value) rescue Float(value) rescue 0).to_s + when 'String' + iter.content = value + when 'Hash', 'Array' + return + else + fail "Unknown type found in model: #{iter.type}" + end + window.change + end + + def configure_value(value, type) + value.editable = false + case type + when 'Array', 'Hash' + value.text = '' + when 'TrueClass' + value.text = 'true' + when 'FalseClass' + value.text = 'false' + when 'NilClass' + value.text = 'null' + when 'Numeric', 'String' + value.text ||= '' + value.editable = true + else + raise ArgumentError, "unknown type '#{type}' encountered" + end + end + + def add_popup_menu + menu = PopUpMenu.new(self) + menu.create + end + + public + + # Create a _type_ node with content _content_, and add it to _parent_ + # in the model. If _parent_ is nil, create a new model and put it into + # the editor treeview. + def create_node(parent, type, content) + iter = if parent + model.append(parent) + else + new_model = Editor.data2model(nil) + toplevel.view_new_model(new_model) + new_model.iter_first + end + iter.type, iter.content = type, content + expand_collapse(parent) if parent + iter + end + + # Ask for a hash key, value pair to be added to the Hash node _parent_. + def ask_for_hash_pair(parent) + key_input = type_input = value_input = nil + + dialog = Dialog.new("New (key, value) pair for Hash", nil, nil, + [ Stock::OK, Dialog::RESPONSE_ACCEPT ], + [ Stock::CANCEL, Dialog::RESPONSE_REJECT ] + ) + + hbox = HBox.new(false, 5) + hbox.pack_start(Label.new("Key:")) + hbox.pack_start(key_input = Entry.new) + key_input.text = @key || '' + dialog.vbox.add(hbox) + key_input.signal_connect(:activate) do + if parent.any? { |c| c.content == key_input.text } + toplevel.display_status('Key already exists in Hash!') + key_input.text = '' + else + toplevel.display_status('Key has been changed.') + end + end + + hbox = HBox.new(false, 5) + hbox.add(Label.new("Type:")) + hbox.pack_start(type_input = ComboBox.new(true)) + ALL_TYPES.each { |t| type_input.append_text(t) } + type_input.active = @type || 0 + dialog.vbox.add(hbox) + + type_input.signal_connect(:changed) do + value_input.editable = false + case ALL_TYPES[type_input.active] + when 'Array', 'Hash' + value_input.text = '' + when 'TrueClass' + value_input.text = 'true' + when 'FalseClass' + value_input.text = 'false' + when 'NilClass' + value_input.text = 'null' + else + value_input.text = '' + value_input.editable = true + end + end + + hbox = HBox.new(false, 5) + hbox.add(Label.new("Value:")) + hbox.pack_start(value_input = Entry.new) + value_input.text = @value || '' + dialog.vbox.add(hbox) + + dialog.signal_connect(:'key-press-event', &DEFAULT_DIALOG_KEY_PRESS_HANDLER) + dialog.show_all + self.focus = dialog + dialog.run do |response| + if response == Dialog::RESPONSE_ACCEPT + @key = key_input.text + type = ALL_TYPES[@type = type_input.active] + content = value_input.text + return @key, type, content + end + end + return + ensure + dialog.destroy + end + + # Ask for an element to be appended _parent_. + def ask_for_element(parent = nil, default_type = nil, value_text = @content) + type_input = value_input = nil + + dialog = Dialog.new( + "New element into #{parent ? parent.type : 'root'}", + nil, nil, + [ Stock::OK, Dialog::RESPONSE_ACCEPT ], + [ Stock::CANCEL, Dialog::RESPONSE_REJECT ] + ) + hbox = HBox.new(false, 5) + hbox.add(Label.new("Type:")) + hbox.pack_start(type_input = ComboBox.new(true)) + default_active = 0 + types = parent ? ALL_TYPES : CONTAINER_TYPES + types.each_with_index do |t, i| + type_input.append_text(t) + if t == default_type + default_active = i + end + end + type_input.active = default_active + dialog.vbox.add(hbox) + type_input.signal_connect(:changed) do + configure_value(value_input, types[type_input.active]) + end + + hbox = HBox.new(false, 5) + hbox.add(Label.new("Value:")) + hbox.pack_start(value_input = Entry.new) + value_input.text = value_text if value_text + configure_value(value_input, types[type_input.active]) + + dialog.vbox.add(hbox) + + dialog.signal_connect(:'key-press-event', &DEFAULT_DIALOG_KEY_PRESS_HANDLER) + dialog.show_all + self.focus = dialog + dialog.run do |response| + if response == Dialog::RESPONSE_ACCEPT + type = types[type_input.active] + @content = case type + when 'Numeric' + Integer(value_input.text) rescue Float(value_input.text) rescue 0 + else + value_input.text + end.to_s + return type, @content + end + end + return + ensure + dialog.destroy if dialog + end + + # Ask for an order criteria for sorting, using _x_ for the element in + # question. Returns the order criterium, and true/false for reverse + # sorting. + def ask_for_order + dialog = Dialog.new( + "Give an order criterium for 'x'.", + nil, nil, + [ Stock::OK, Dialog::RESPONSE_ACCEPT ], + [ Stock::CANCEL, Dialog::RESPONSE_REJECT ] + ) + hbox = HBox.new(false, 5) + + hbox.add(Label.new("Order:")) + hbox.pack_start(order_input = Entry.new) + order_input.text = @order || 'x' + + hbox.pack_start(reverse_checkbox = CheckButton.new('Reverse')) + + dialog.vbox.add(hbox) + + dialog.signal_connect(:'key-press-event', &DEFAULT_DIALOG_KEY_PRESS_HANDLER) + dialog.show_all + self.focus = dialog + dialog.run do |response| + if response == Dialog::RESPONSE_ACCEPT + return @order = order_input.text, reverse_checkbox.active? + end + end + return + ensure + dialog.destroy if dialog + end + + # Ask for a find term to search for in the tree. Returns the term as a + # string. + def ask_for_find_term + dialog = Dialog.new( + "Find a node matching regex in tree.", + nil, nil, + [ Stock::OK, Dialog::RESPONSE_ACCEPT ], + [ Stock::CANCEL, Dialog::RESPONSE_REJECT ] + ) + hbox = HBox.new(false, 5) + + hbox.add(Label.new("Regex:")) + hbox.pack_start(regex_input = Entry.new) + regex_input.text = @regex || '' + + dialog.vbox.add(hbox) + + dialog.signal_connect(:'key-press-event', &DEFAULT_DIALOG_KEY_PRESS_HANDLER) + dialog.show_all + self.focus = dialog + dialog.run do |response| + if response == Dialog::RESPONSE_ACCEPT + return @regex = regex_input.text + end + end + return + ensure + dialog.destroy if dialog + end + + # Expand or collapse row pointed to by _iter_ according + # to the #expanded attribute. + def expand_collapse(iter) + if expanded + expand_row(iter.path, true) + else + collapse_row(iter.path) + end + end + end + + # The editor main window + class MainWindow < Gtk::Window + include Gtk + + def initialize(encoding) + @changed = false + @encoding = encoding + super(TOPLEVEL) + display_title + set_default_size(800, 600) + signal_connect(:delete_event) { quit } + + vbox = VBox.new(false, 0) + add(vbox) + #vbox.border_width = 0 + + @treeview = JSONTreeView.new(self) + @treeview.signal_connect(:'cursor-changed') do + display_status('') + end + + menu_bar = create_menu_bar + vbox.pack_start(menu_bar, false, false, 0) + + sw = ScrolledWindow.new(nil, nil) + sw.shadow_type = SHADOW_ETCHED_IN + sw.set_policy(POLICY_AUTOMATIC, POLICY_AUTOMATIC) + vbox.pack_start(sw, true, true, 0) + sw.add(@treeview) + + @status_bar = Statusbar.new + vbox.pack_start(@status_bar, false, false, 0) + + @filename ||= nil + if @filename + data = read_data(@filename) + view_new_model Editor.data2model(data) + end + end + + # Creates the menu bar with the pulldown menus and returns it. + def create_menu_bar + menu_bar = MenuBar.new + @file_menu = FileMenu.new(@treeview) + menu_bar.append @file_menu.create + @edit_menu = EditMenu.new(@treeview) + menu_bar.append @edit_menu.create + @options_menu = OptionsMenu.new(@treeview) + menu_bar.append @options_menu.create + menu_bar + end + + # Sets editor status to changed, to indicate that the edited data + # containts unsaved changes. + def change + @changed = true + display_title + end + + # Sets editor status to unchanged, to indicate that the edited data + # doesn't containt unsaved changes. + def unchange + @changed = false + display_title + end + + # Puts a new model _model_ into the Gtk::TreeView to be edited. + def view_new_model(model) + @treeview.model = model + @treeview.expanded = true + @treeview.expand_all + unchange + end + + # Displays _text_ in the status bar. + def display_status(text) + @cid ||= nil + @status_bar.pop(@cid) if @cid + @cid = @status_bar.get_context_id('dummy') + @status_bar.push(@cid, text) + end + + # Opens a dialog, asking, if changes should be saved to a file. + def ask_save + if Editor.question_dialog(self, + "Unsaved changes to JSON model. Save?") + if @filename + file_save + else + file_save_as + end + end + end + + # Quit this editor, that is, leave this editor's main loop. + def quit + ask_save if @changed + destroy + Gtk.main_quit + true + end + + # Display the new title according to the editor's current state. + def display_title + title = TITLE.dup + title << ": #@filename" if @filename + title << " *" if @changed + self.title = title + end + + # Clear the current model, after asking to save all unsaved changes. + def clear + ask_save if @changed + @filename = nil + self.view_new_model nil + end + + def check_pretty_printed(json) + pretty = !!((nl_index = json.index("\n")) && nl_index != json.size - 1) + @options_menu.pretty_item.active = pretty + end + private :check_pretty_printed + + # Open the data at the location _uri_, if given. Otherwise open a dialog + # to ask for the _uri_. + def location_open(uri = nil) + uri = ask_for_location unless uri + uri or return + data = load_location(uri) or return + view_new_model Editor.data2model(data) + end + + # Open the file _filename_ or call the #select_file method to ask for a + # filename. + def file_open(filename = nil) + filename = select_file('Open as a JSON file') unless filename + data = load_file(filename) or return + view_new_model Editor.data2model(data) + end + + # Save the current file. + def file_save + if @filename + store_file(@filename) + else + file_save_as + end + end + + # Save the current file as the filename + def file_save_as + filename = select_file('Save as a JSON file') + store_file(filename) + end + + # Store the current JSON document to _path_. + def store_file(path) + if path + data = Editor.model2data(@treeview.model.iter_first) + File.open(path + '.tmp', 'wb') do |output| + if @options_menu.pretty_item.active? + output.puts JSON.pretty_generate(data) + else + output.write JSON.unparse(data) + end + end + File.rename path + '.tmp', path + @filename = path + toplevel.display_status("Saved data to '#@filename'.") + unchange + end + rescue SystemCallError => e + Editor.error_dialog(self, "Failed to store JSON file: #{e}!") + end + + # Load the file named _filename_ into the editor as a JSON document. + def load_file(filename) + if filename + if File.directory?(filename) + Editor.error_dialog(self, "Try to select a JSON file!") + return + else + data = read_data(filename) + @filename = filename + toplevel.display_status("Loaded data from '#@filename'.") + display_title + return data + end + end + end + + # Load the data at location _uri_ into the editor as a JSON document. + def load_location(uri) + data = read_data(uri) + @filename = nil + toplevel.display_status("Loaded data from '#{uri}'.") + display_title + data + end + + # Read a JSON document from the file named _filename_, parse it into a + # ruby data structure, and return the data. + def read_data(filename) + open(filename) do |f| + json = f.read + check_pretty_printed(json) + if @encoding && !/^utf8$/i.match(@encoding) + iconverter = Iconv.new('utf8', @encoding) + json = iconverter.iconv(json) + end + return JSON::parse(json, :max_nesting => false) + end + rescue JSON::JSONError => e + Editor.error_dialog(self, "Failed to parse JSON file: #{e}!") + return + rescue SystemCallError => e + quit + end + + # Open a file selecton dialog, displaying _message_, and return the + # selected filename or nil, if no file was selected. + def select_file(message) + filename = nil + fs = FileSelection.new(message).set_modal(true). + set_filename(Dir.pwd + "/").set_transient_for(self) + fs.signal_connect(:destroy) { Gtk.main_quit } + fs.ok_button.signal_connect(:clicked) do + filename = fs.filename + fs.destroy + Gtk.main_quit + end + fs.cancel_button.signal_connect(:clicked) do + fs.destroy + Gtk.main_quit + end + fs.show_all + Gtk.main + filename + end + + # Ask for location URI a to load data from. Returns the URI as a string. + def ask_for_location + dialog = Dialog.new( + "Load data from location...", + nil, nil, + [ Stock::OK, Dialog::RESPONSE_ACCEPT ], + [ Stock::CANCEL, Dialog::RESPONSE_REJECT ] + ) + hbox = HBox.new(false, 5) + + hbox.add(Label.new("Location:")) + hbox.pack_start(location_input = Entry.new) + location_input.width_chars = 60 + location_input.text = @location || '' + + dialog.vbox.add(hbox) + + dialog.signal_connect(:'key-press-event', &DEFAULT_DIALOG_KEY_PRESS_HANDLER) + dialog.show_all + dialog.run do |response| + if response == Dialog::RESPONSE_ACCEPT + return @location = location_input.text + end + end + return + ensure + dialog.destroy if dialog + end + end + + class << self + # Starts a JSON Editor. If a block was given, it yields + # to the JSON::Editor::MainWindow instance. + def start(encoding = nil) # :yield: window + encoding ||= 'utf8' + Gtk.init + @window = Editor::MainWindow.new(encoding) + @window.icon_list = [ Editor.fetch_icon('json') ] + yield @window if block_given? + @window.show_all + Gtk.main + end + + attr_reader :window + end + end +end + # vim: set et sw=2 ts=2: diff --git a/lib/json/ext.rb b/lib/json/ext.rb new file mode 100644 index 0000000000..ff4fa42329 --- /dev/null +++ b/lib/json/ext.rb @@ -0,0 +1,13 @@ +require 'json/common' + +module JSON + # This module holds all the modules/classes that implement JSON's + # functionality as C extensions. + module Ext + require 'json/ext/parser' + require 'json/ext/generator' + $DEBUG and warn "Using c extension for JSON." + JSON.parser = Parser + JSON.generator = Generator + end +end diff --git a/lib/json/json.xpm b/lib/json/json.xpm new file mode 100644 index 0000000000..2cb626bb05 --- /dev/null +++ b/lib/json/json.xpm @@ -0,0 +1,1499 @@ +/* XPM */ +static char * json_xpm[] = { +"64 64 1432 2", +" c None", +". c #641839", +"+ c #CF163C", +"@ c #D31C3B", +"# c #E11A38", +"$ c #5F242D", +"% c #320C22", +"& c #9B532D", +"* c #F32E34", +"= c #820F33", +"- c #4B0F34", +"; c #8E1237", +"> c #944029", +", c #961325", +"' c #A00C24", +") c #872C23", +"! c #694021", +"~ c #590D1F", +"{ c #420528", +"] c #D85A2D", +"^ c #7E092B", +"/ c #0E0925", +"( c #0D081F", +"_ c #0F081E", +": c #12071F", +"< c #360620", +"[ c #682A21", +"} c #673F21", +"| c #780E21", +"1 c #A82320", +"2 c #8D1D1F", +"3 c #970127", +"4 c #0D0123", +"5 c #0D0324", +"6 c #3B1E28", +"7 c #C28429", +"8 c #0C0523", +"9 c #0C041E", +"0 c #0E031A", +"a c #11031A", +"b c #13031B", +"c c #13031C", +"d c #11031D", +"e c #19051E", +"f c #390E20", +"g c #9C0C20", +"h c #C00721", +"i c #980320", +"j c #14031E", +"k c #CD9F32", +"l c #C29F2E", +"m c #0F0325", +"n c #0D0321", +"o c #0E0324", +"p c #D08329", +"q c #9D1B27", +"r c #1C0320", +"s c #0D011A", +"t c #120117", +"u c #130017", +"v c #150018", +"w c #160119", +"x c #17021A", +"y c #15021B", +"z c #11021E", +"A c #0F021F", +"B c #8C1821", +"C c #CF4522", +"D c #831821", +"E c #BA7033", +"F c #EDB339", +"G c #C89733", +"H c #280727", +"I c #0F051F", +"J c #0E0420", +"K c #591F27", +"L c #E47129", +"M c #612224", +"N c #0C021D", +"O c #120018", +"P c #140017", +"Q c #170017", +"R c #190018", +"S c #1B0019", +"T c #1B011A", +"U c #18011B", +"V c #15011C", +"W c #12031E", +"X c #460A21", +"Y c #A13823", +"Z c #784323", +"` c #5A0C21", +" . c #BC4530", +".. c #EB5B38", +"+. c #CE4E3B", +"@. c #DD9334", +"#. c #751A27", +"$. c #11071E", +"%. c #0F041C", +"&. c #1E0824", +"*. c #955A28", +"=. c #9A5027", +"-. c #1E0321", +";. c #11011A", +">. c #140018", +",. c #180018", +"'. c #1F001A", +"). c #20001B", +"!. c #1E001A", +"~. c #1B001A", +"{. c #16021B", +"]. c #16041E", +"^. c #220622", +"/. c #5F3525", +"(. c #DE5724", +"_. c #611021", +":. c #0F0925", +"<. c #D1892E", +"[. c #F27036", +"}. c #EC633B", +"|. c #DA293C", +"1. c #E64833", +"2. c #912226", +"3. c #11081C", +"4. c #110419", +"5. c #0F041E", +"6. c #451425", +"7. c #BF6F28", +"8. c #332225", +"9. c #0E021E", +"0. c #13001B", +"a. c #17001A", +"b. c #1C001B", +"c. c #21001C", +"d. c #23001C", +"e. c #21001B", +"f. c #19021A", +"g. c #17041E", +"h. c #150721", +"i. c #602424", +"j. c #D51223", +"k. c #540820", +"l. c #D04D2D", +"m. c #EA8933", +"n. c #875637", +"o. c #88543A", +"p. c #E5923A", +"q. c #891931", +"r. c #130B25", +"s. c #10051B", +"t. c #110217", +"u. c #12021A", +"v. c #761826", +"w. c #E2A728", +"x. c #300224", +"y. c #10011E", +"z. c #16001B", +"A. c #1B001B", +"B. c #21001A", +"C. c #1E0019", +"D. c #1D0019", +"E. c #1A011A", +"F. c #17031C", +"G. c #120720", +"H. c #4E0822", +"I. c #670721", +"J. c #C07630", +"K. c #F59734", +"L. c #BE1B35", +"M. c #0E1435", +"N. c #522037", +"O. c #DB8039", +"P. c #D45933", +"Q. c #420927", +"R. c #0F041D", +"S. c #140118", +"T. c #13021D", +"U. c #100423", +"V. c #7B6227", +"W. c #C04326", +"X. c #0E0020", +"Y. c #13001D", +"Z. c #18001B", +"`. c #1E001B", +" + c #22001C", +".+ c #22001B", +"++ c #1B011B", +"@+ c #16041D", +"#+ c #130520", +"$+ c #860521", +"%+ c #710520", +"&+ c #670A2A", +"*+ c #A66431", +"=+ c #E97536", +"-+ c #F8833A", +";+ c #F77A3A", +">+ c #C45337", +",+ c #0A1C35", +"'+ c #993638", +")+ c #F7863B", +"!+ c #F49736", +"~+ c #94462B", +"{+ c #0E031F", +"]+ c #130119", +"^+ c #160018", +"/+ c #16011B", +"(+ c #15021F", +"_+ c #120123", +":+ c #A65C28", +"<+ c #5C4D23", +"[+ c #0F001F", +"}+ c #14001D", +"|+ c #1A001B", +"1+ c #1F001B", +"2+ c #24001D", +"3+ c #25001D", +"4+ c #24001C", +"5+ c #1F001C", +"6+ c #1A011C", +"7+ c #16021E", +"8+ c #3F0421", +"9+ c #BC0522", +"0+ c #1C041E", +"a+ c #7F5531", +"b+ c #E68A38", +"c+ c #F8933E", +"d+ c #FA7942", +"e+ c #FB7543", +"f+ c #FA6F41", +"g+ c #F1793D", +"h+ c #7D3B3A", +"i+ c #28263B", +"j+ c #D45441", +"k+ c #F8A238", +"l+ c #996B2D", +"m+ c #0E0421", +"n+ c #12011A", +"o+ c #180019", +"p+ c #17001C", +"q+ c #12001F", +"r+ c #4C2B2A", +"s+ c #DB8130", +"t+ c #540023", +"u+ c #0F0120", +"v+ c #16011C", +"w+ c #22001D", +"x+ c #25001F", +"y+ c #26001F", +"z+ c #25001E", +"A+ c #24001E", +"B+ c #1D001C", +"C+ c #18011D", +"D+ c #16031F", +"E+ c #3C0522", +"F+ c #9B0821", +"G+ c #13041E", +"H+ c #F6462E", +"I+ c #E6AB37", +"J+ c #E7A03E", +"K+ c #FA9F44", +"L+ c #FB8A48", +"M+ c #FD7A4A", +"N+ c #FD794A", +"O+ c #FD7748", +"P+ c #FD7E45", +"Q+ c #FD8343", +"R+ c #FB5D42", +"S+ c #6E3A40", +"T+ c #EE8A37", +"U+ c #7E252B", +"V+ c #100520", +"W+ c #13011A", +"X+ c #170019", +"Y+ c #15001C", +"Z+ c #0F0020", +"`+ c #564427", +" @ c #E0BA29", +".@ c #5E2B25", +"+@ c #10011F", +"@@ c #17011C", +"#@ c #1E001D", +"$@ c #23001F", +"%@ c #250020", +"&@ c #24001F", +"*@ c #23001E", +"=@ c #21001E", +"-@ c #1B001C", +";@ c #17021D", +">@ c #14041E", +",@ c #AC0B25", +"'@ c #5E1420", +")@ c #F28635", +"!@ c #C2733E", +"~@ c #984C44", +"{@ c #EA9148", +"]@ c #FB844B", +"^@ c #FD7E4C", +"/@ c #FE7E4C", +"(@ c #FE7E4B", +"_@ c #FE7749", +":@ c #FD7148", +"<@ c #FB7D46", +"[@ c #F89641", +"}@ c #B95634", +"|@ c #0D0927", +"1@ c #11041D", +"2@ c #150119", +"3@ c #180017", +"4@ c #16001A", +"5@ c #13001E", +"6@ c #110023", +"7@ c #944C29", +"8@ c #EE6229", +"9@ c #3D0324", +"0@ c #12021F", +"a@ c #19011D", +"b@ c #21001F", +"c@ c #22001F", +"d@ c #20001E", +"e@ c #1F001D", +"f@ c #1C001C", +"g@ c #19011C", +"h@ c #3D1621", +"i@ c #B53622", +"j@ c #31061F", +"k@ c #841D34", +"l@ c #F2703F", +"m@ c #C14445", +"n@ c #E67349", +"o@ c #FB8E4B", +"p@ c #FD834C", +"q@ c #FE834D", +"r@ c #FE834C", +"s@ c #FE804C", +"t@ c #FD814B", +"u@ c #FB7D49", +"v@ c #F79B43", +"w@ c #AF1234", +"x@ c #0D0625", +"y@ c #13021C", +"z@ c #1A0019", +"A@ c #190019", +"B@ c #410225", +"C@ c #D39729", +"D@ c #AA5927", +"E@ c #0E0422", +"F@ c #15021E", +"G@ c #1A011D", +"H@ c #1D001D", +"I@ c #15031D", +"J@ c #240820", +"K@ c #A01023", +"L@ c #670B21", +"M@ c #3D0D33", +"N@ c #E63C3E", +"O@ c #EF7C45", +"P@ c #F59048", +"Q@ c #FB944A", +"R@ c #FD904A", +"S@ c #FE8E4B", +"T@ c #FE854A", +"U@ c #FE854B", +"V@ c #FE884C", +"W@ c #FC954B", +"X@ c #F8AB45", +"Y@ c #C37A35", +"Z@ c #0D0425", +"`@ c #13011B", +" # c #170018", +".# c #1A0018", +"+# c #1C0019", +"@# c #15001B", +"## c #100120", +"$# c #311F25", +"%# c #E68E28", +"&# c #7A1425", +"*# c #130321", +"=# c #17011E", +"-# c #1A001D", +";# c #19001B", +"># c #16021C", +",# c #130521", +"'# c #6F3123", +")# c #6D3022", +"!# c #C89433", +"~# c #EA7E3E", +"{# c #DB2943", +"]# c #EF7745", +"^# c #FB8544", +"/# c #FD9A43", +"(# c #FE9941", +"_# c #FE9D43", +":# c #FEA548", +"<# c #FEAE49", +"[# c #FCB944", +"}# c #CA9F35", +"|# c #0E0225", +"1# c #11001B", +"2# c #160019", +"3# c #12011B", +"4# c #0F0220", +"5# c #351D26", +"6# c #D85B28", +"7# c #6C0F26", +"8# c #190121", +"9# c #1B001E", +"0# c #1A001C", +"a# c #1D001B", +"b# c #130220", +"c# c #703A23", +"d# c #713A23", +"e# c #140327", +"f# c #411B36", +"g# c #C8713E", +"h# c #7A3A3F", +"i# c #CE2C3C", +"j# c #E77338", +"k# c #9C6535", +"l# c #9C6233", +"m# c #9C6332", +"n# c #9C6A35", +"o# c #C37D3C", +"p# c #FEAC41", +"q# c #FEC23E", +"r# c #826330", +"s# c #100122", +"t# c #120019", +"u# c #150017", +"v# c #190017", +"w# c #1B0018", +"x# c #12001A", +"y# c #10021F", +"z# c #1A0326", +"A# c #5F292A", +"B# c #7B4E29", +"C# c #3C0E25", +"D# c #1A0020", +"E# c #14021F", +"F# c #723B23", +"G# c #14001A", +"H# c #58042A", +"I# c #A28337", +"J# c #C8813B", +"K# c #B14B38", +"L# c #761231", +"M# c #5A132A", +"N# c #0D0726", +"O# c #0C0623", +"P# c #0B0723", +"Q# c #0B0A26", +"R# c #321C2D", +"S# c #C45B33", +"T# c #FEBB33", +"U# c #13052A", +"V# c #13011F", +"W# c #160017", +"X# c #15001A", +"Y# c #12001D", +"Z# c #94062A", +"`# c #630D2C", +" $ c #85292B", +".$ c #AA5E29", +"+$ c #1F0123", +"@$ c #19011F", +"#$ c #1E001C", +"$$ c #15031F", +"%$ c #712122", +"&$ c #712223", +"*$ c #14011B", +"=$ c #110321", +"-$ c #AF0C2B", +";$ c #E7D534", +">$ c #EAC934", +",$ c #84582D", +"'$ c #1B0824", +")$ c #11041E", +"!$ c #10021B", +"~$ c #100119", +"{$ c #100218", +"]$ c #0F041A", +"^$ c #0E0720", +"/$ c #2C1026", +"($ c #D8A328", +"_$ c #140322", +":$ c #160016", +"<$ c #14001F", +"[$ c #120024", +"}$ c #100128", +"|$ c #3C032F", +"1$ c #2C062E", +"2$ c #29022B", +"3$ c #A31D29", +"4$ c #976A25", +"5$ c #1A0321", +"6$ c #17031E", +"7$ c #1B021D", +"8$ c #20001C", +"9$ c #14041F", +"0$ c #703422", +"a$ c #6F3522", +"b$ c #8D0328", +"c$ c #920329", +"d$ c #0F0326", +"e$ c #100321", +"f$ c #11021B", +"g$ c #130117", +"h$ c #140016", +"i$ c #150015", +"j$ c #140015", +"k$ c #130116", +"l$ c #120219", +"m$ c #11031C", +"n$ c #12031D", +"o$ c #170016", +"p$ c #160020", +"q$ c #250029", +"r$ c #670033", +"s$ c #DCA238", +"t$ c #F5C736", +"u$ c #9A732E", +"v$ c #110227", +"w$ c #110324", +"x$ c #811924", +"y$ c #A04323", +"z$ c #250721", +"A$ c #1A041F", +"B$ c #1E011D", +"C$ c #1C011C", +"D$ c #18031D", +"E$ c #130721", +"F$ c #6F3623", +"G$ c #6B3622", +"H$ c #1A001A", +"I$ c #14011F", +"J$ c #12011E", +"K$ c #11011C", +"L$ c #140117", +"M$ c #170015", +"N$ c #150016", +"O$ c #120119", +"P$ c #11011B", +"Q$ c #11001A", +"R$ c #130018", +"S$ c #170118", +"T$ c #170119", +"U$ c #18021E", +"V$ c #1A0126", +"W$ c #6F2332", +"X$ c #E5563B", +"Y$ c #F1B83F", +"Z$ c #F6CC38", +"`$ c #9D7A2D", +" % c #130123", +".% c #130320", +"+% c #2A0721", +"@% c #B00E24", +"#% c #7D0B23", +"$% c #1F0522", +"%% c #1E0220", +"&% c #1D011E", +"*% c #1A031E", +"=% c #15051F", +"-% c #241322", +";% c #A32F23", +">% c #670E21", +",% c #1C001A", +"'% c #19001A", +")% c #180016", +"!% c #160118", +"~% c #140219", +"{% c #11021C", +"]% c #10021E", +"^% c #0F011D", +"/% c #170117", +"(% c #160219", +"_% c #17041D", +":% c #190523", +"<% c #8C042E", +"[% c #B65838", +"}% c #E9D73F", +"|% c #EED43E", +"1% c #D85538", +"2% c #493129", +"3% c #130120", +"4% c #15021D", +"5% c #330822", +"6% c #8A0825", +"7% c #3C0424", +"8% c #1E0322", +"9% c #1C0321", +"0% c #180421", +"a% c #130822", +"b% c #AF2D24", +"c% c #BC5623", +"d% c #2F071F", +"e% c #1A041C", +"f% c #1C031C", +"g% c #1D011C", +"h% c #160117", +"i% c #150419", +"j% c #12081D", +"k% c #0F0923", +"l% c #A77027", +"m% c #A60525", +"n% c #11021A", +"o% c #130218", +"p% c #150319", +"q% c #16061D", +"r% c #180923", +"s% c #9C1D2B", +"t% c #A32636", +"u% c #A66E3B", +"v% c #4B2E3C", +"w% c #412C36", +"x% c #36012D", +"y% c #140123", +"z% c #17001E", +"A% c #19011B", +"B% c #1A0421", +"C% c #340425", +"D% c #9E0326", +"E% c #1F0424", +"F% c #1C0524", +"G% c #180724", +"H% c #A91024", +"I% c #D55D24", +"J% c #90071E", +"K% c #3C051D", +"L% c #1C021C", +"M% c #1C011A", +"N% c #1D001A", +"O% c #160116", +"P% c #150216", +"Q% c #140217", +"R% c #140618", +"S% c #120D1D", +"T% c #231925", +"U% c #B16A2E", +"V% c #FDAC34", +"W% c #D58631", +"X% c #280E2A", +"Y% c #0D0A23", +"Z% c #0F0920", +"`% c #120C21", +" & c #1F1026", +".& c #A3352E", +"+& c #EE9F36", +"@& c #5D2A3C", +"#& c #960D3C", +"$& c #970638", +"%& c #A00330", +"&& c #4D0126", +"*& c #1C001F", +"=& c #280120", +"-& c #290223", +";& c #1F0425", +">& c #260726", +",& c #340A26", +"'& c #850925", +")& c #3A0823", +"!& c #82071D", +"~& c #5E071D", +"{& c #18051C", +"]& c #18021A", +"^& c #190118", +"/& c #160217", +"(& c #150418", +"_& c #130618", +":& c #110718", +"<& c #10081A", +"[& c #110D1D", +"}& c #291C24", +"|& c #A73B2D", +"1& c #FD6B36", +"2& c #FD853C", +"3& c #FD863B", +"4& c #C24A35", +"5& c #6B442F", +"6& c #6D302D", +"7& c #6E252E", +"8& c #8E3B32", +"9& c #DE7739", +"0& c #F48E3F", +"a& c #DD8D41", +"b& c #854F3D", +"c& c #7E2D35", +"d& c #33082B", +"e& c #1C0222", +"f& c #20001F", +"g& c #1F0222", +"h& c #1A0524", +"i& c #440C27", +"j& c #BC1427", +"k& c #20041B", +"l& c #53061C", +"m& c #25071B", +"n& c #11061A", +"o& c #130418", +"p& c #140317", +"q& c #150217", +"r& c #160318", +"s& c #12051B", +"t& c #100C1D", +"u& c #0E101E", +"v& c #0C121F", +"w& c #0C1321", +"x& c #781725", +"y& c #B25D2C", +"z& c #FA6335", +"A& c #FD633C", +"B& c #FE6D42", +"C& c #FE7C42", +"D& c #FE813F", +"E& c #FE873C", +"F& c #FD743B", +"G& c #FB683B", +"H& c #FA7A3E", +"I& c #F98242", +"J& c #F97844", +"K& c #F98943", +"L& c #F79C3D", +"M& c #A25133", +"N& c #280B28", +"O& c #1D021F", +"P& c #1F011C", +"Q& c #280321", +"R& c #1C0724", +"S& c #3F1C27", +"T& c #D33C27", +"U& c #0E061B", +"V& c #0C091C", +"W& c #0C0A1B", +"X& c #0E091A", +"Y& c #11081B", +"Z& c #100A20", +"`& c #0E0D23", +" * c #551227", +".* c #B21829", +"+* c #C42329", +"@* c #C62C29", +"#* c #C55429", +"$* c #E76F2B", +"%* c #F14232", +"&* c #F95E3A", +"** c #FC6740", +"=* c #FE6E45", +"-* c #FE7246", +";* c #FE7545", +">* c #FE7744", +",* c #FD7745", +"'* c #FD7845", +")* c #FD7847", +"!* c #FD7948", +"~* c #FD7B44", +"{* c #FC7C3B", +"]* c #6F3130", +"^* c #140B24", +"/* c #19031D", +"(* c #1C011B", +"_* c #5A011F", +":* c #B70421", +"<* c #380824", +"[* c #3E2626", +"}* c #9F5626", +"|* c #13051E", +"1* c #360A21", +"2* c #361223", +"3* c #371724", +"4* c #381824", +"5* c #3B1524", +"6* c #3E1E26", +"7* c #471A29", +"8* c #DB252E", +"9* c #ED2733", +"0* c #EE5436", +"a* c #F04237", +"b* c #F33934", +"c* c #F53D2F", +"d* c #D7312B", +"e* c #AF212B", +"f* c #3A2C31", +"g* c #F65F39", +"h* c #FB6F41", +"i* c #FD6D45", +"j* c #FE7047", +"k* c #FE7647", +"l* c #FE7847", +"m* c #FE7848", +"n* c #FE7748", +"o* c #FE7948", +"p* c #FE7C48", +"q* c #FE7C47", +"r* c #FE7642", +"s* c #FE7439", +"t* c #6D332C", +"u* c #100B21", +"v* c #16031B", +"w* c #2B001B", +"x* c #22011F", +"y* c #220521", +"z* c #1B0A23", +"A* c #421425", +"B* c #951924", +"C* c #381023", +"D* c #E94028", +"E* c #E7302B", +"F* c #EF432D", +"G* c #F4302E", +"H* c #F32C30", +"I* c #CB4432", +"J* c #DD3235", +"K* c #EF4B3A", +"L* c #F0333E", +"M* c #CC3D3F", +"N* c #E4313C", +"O* c #F34834", +"P* c #D13E2C", +"Q* c #431825", +"R* c #0E1424", +"S* c #3C202C", +"T* c #F15537", +"U* c #F97140", +"V* c #FC6E45", +"W* c #FE7547", +"X* c #FE7947", +"Y* c #FE7B48", +"Z* c #FE7D48", +"`* c #FE8047", +" = c #FE7A42", +".= c #FE7A38", +"+= c #6D442B", +"@= c #0F0B21", +"#= c #15031A", +"$= c #49001B", +"%= c #2F001C", +"&= c #21021E", +"*= c #220620", +"== c #1B0D23", +"-= c #641625", +";= c #951823", +">= c #390F25", +",= c #AC3A2A", +"'= c #B6492E", +")= c #ED7531", +"!= c #F45A34", +"~= c #F54C36", +"{= c #C72D39", +"]= c #DE283C", +"^= c #F33B40", +"/= c #F34142", +"(= c #D0393F", +"_= c #E72E39", +":= c #DB3C2E", +"<= c #461724", +"[= c #0F0D1E", +"}= c #140B1E", +"|= c #341427", +"1= c #CB4834", +"2= c #F7743F", +"3= c #FB7145", +"4= c #FE7747", +"5= c #FE7A47", +"6= c #FF7B48", +"7= c #FF7C48", +"8= c #FE7F47", +"9= c #FE8247", +"0= c #FE8642", +"a= c #FE8439", +"b= c #6D442D", +"c= c #0F0A21", +"d= c #14031A", +"e= c #20031D", +"f= c #210821", +"g= c #191024", +"h= c #CC1C25", +"i= c #961423", +"j= c #2C162C", +"k= c #BD242E", +"l= c #EF2C31", +"m= c #F54C34", +"n= c #F34037", +"o= c #F5353A", +"p= c #F7413D", +"q= c #F8423D", +"r= c #F93A39", +"s= c #F95731", +"t= c #341425", +"u= c #110A1D", +"v= c #140619", +"w= c #18051B", +"x= c #200F26", +"y= c #864833", +"z= c #F8773F", +"A= c #FC7445", +"B= c #FF7E48", +"C= c #FF7E49", +"D= c #FF7D49", +"E= c #FF7D48", +"F= c #FE8347", +"G= c #FE8743", +"H= c #FE893B", +"I= c #6E452F", +"J= c #100E23", +"K= c #14041A", +"L= c #55041D", +"M= c #540921", +"N= c #161124", +"O= c #CE6A25", +"P= c #3F1129", +"Q= c #170A29", +"R= c #0F0F29", +"S= c #15132B", +"T= c #1E182D", +"U= c #A82B3D", +"V= c #CB6633", +"W= c #CC6932", +"X= c #CC3D2D", +"Y= c #331225", +"Z= c #0F091C", +"`= c #120417", +" - c #160216", +".- c #190419", +"+- c #210F26", +"@- c #8C4934", +"#- c #F97A40", +"$- c #FC7545", +"%- c #FF7B49", +"&- c #FE7D46", +"*- c #FE7E43", +"=- c #FD7B3E", +"-- c #FA6934", +";- c #532328", +">- c #130B1D", +",- c #150519", +"'- c #14041C", +")- c #120920", +"!- c #C43624", +"~- c #A21E23", +"{- c #F87C30", +"]- c #C9302D", +"^- c #300F2A", +"/- c #591129", +"(- c #171328", +"_- c #171628", +":- c #141829", +"<- c #101A2B", +"[- c #0F172B", +"}- c #0F1226", +"|- c #0E0C20", +"1- c #100619", +"2- c #140316", +"3- c #19051B", +"4- c #3C1428", +"5- c #E04B36", +"6- c #FA7B41", +"7- c #FD7346", +"8- c #FE7548", +"9- c #FF7849", +"0- c #FF7749", +"a- c #FE7B47", +"b- c #FE7945", +"c- c #FC7740", +"d- c #FA7E39", +"e- c #C1432F", +"f- c #131523", +"g- c #130A1C", +"h- c #420621", +"i- c #D08423", +"j- c #F87739", +"k- c #C03D37", +"l- c #962B34", +"m- c #A14332", +"n- c #E54B30", +"o- c #9E3E2F", +"p- c #7F262E", +"q- c #922D2E", +"r- c #9C4B2E", +"s- c #65212C", +"t- c #101628", +"u- c #101022", +"v- c #11091C", +"w- c #130619", +"x- c #160A1E", +"y- c #43252C", +"z- c #F66439", +"A- c #FA6942", +"B- c #FD6C47", +"C- c #FE6E48", +"D- c #FE6F48", +"E- c #FE7049", +"F- c #FE714A", +"G- c #FE744A", +"H- c #FE7846", +"I- c #FD7243", +"J- c #FC703E", +"K- c #FA6C37", +"L- c #81312B", +"M- c #121123", +"N- c #15071D", +"O- c #16031A", +"P- c #17021B", +"Q- c #8F3D22", +"R- c #F8393E", +"S- c #E42A3D", +"T- c #E7473B", +"U- c #FB503B", +"V- c #FB4F3A", +"W- c #F95439", +"X- c #ED4C38", +"Y- c #F45938", +"Z- c #FB6537", +"`- c #EA5236", +" ; c #CE6232", +".; c #CD392C", +"+; c #181425", +"@; c #120F21", +"#; c #130D20", +"$; c #151225", +"%; c #903431", +"&; c #F8703D", +"*; c #FB6344", +"=; c #FD6748", +"-; c #FE6849", +";; c #FE6949", +">; c #FE6A49", +",; c #FE6C4A", +"'; c #FE704A", +"); c #FE734A", +"!; c #FE7449", +"~; c #FE7347", +"{; c #FE7145", +"]; c #FD6C42", +"^; c #FD753D", +"/; c #F36E35", +"(; c #CB452C", +"_; c #600D24", +":; c #1C061F", +"<; c #1E031F", +"[; c #5B3821", +"}; c #CE9822", +"|; c #FA4341", +"1; c #FB4341", +"2; c #FC4541", +"3; c #FC4542", +"4; c #FC4143", +"5; c #FC4D42", +"6; c #FB5042", +"7; c #FB5342", +"8; c #FC5242", +"9; c #FD4F40", +"0; c #FD503E", +"a; c #FB6339", +"b; c #F45E33", +"c; c #A12A2E", +"d; c #401E2C", +"e; c #452D2F", +"f; c #F74F38", +"g; c #FA5940", +"h; c #FC6245", +"i; c #FE6447", +"j; c #FE6449", +"k; c #FE6549", +"l; c #FE6749", +"m; c #FE6B49", +"n; c #FE6D49", +"o; c #FE6D48", +"p; c #FE6D47", +"q; c #FE6D45", +"r; c #FE6C44", +"s; c #FE6A42", +"t; c #FE663C", +"u; c #FC6233", +"v; c #752129", +"w; c #1F0922", +"x; c #750520", +"y; c #81061F", +"z; c #FA3D42", +"A; c #FB4142", +"B; c #FD4543", +"C; c #FD4844", +"D; c #FD4A45", +"E; c #FD4D45", +"F; c #FD5045", +"G; c #FD5345", +"H; c #FE5346", +"I; c #FE5445", +"J; c #FD5444", +"K; c #FC4F41", +"L; c #FA513D", +"M; c #F95339", +"N; c #F63736", +"O; c #F75737", +"P; c #F95F3B", +"Q; c #FB5840", +"R; c #FD5F43", +"S; c #FE6345", +"T; c #FE6547", +"U; c #FE6548", +"V; c #FE6448", +"W; c #FE6248", +"X; c #FE6348", +"Y; c #FE6748", +"Z; c #FE6848", +"`; c #FE6846", +" > c #FE6A45", +".> c #FE6D43", +"+> c #FE703F", +"@> c #FC6F36", +"#> c #6F302B", +"$> c #140A22", +"%> c #FA3B42", +"&> c #FC4243", +"*> c #FD4744", +"=> c #FE4A45", +"-> c #FE4C47", +";> c #FE4D47", +">> c #FE5047", +",> c #FE5347", +"'> c #FE5447", +")> c #FD5246", +"!> c #FB503F", +"~> c #FA543D", +"{> c #9B3D3B", +"]> c #A3433B", +"^> c #F9683D", +"/> c #FC6940", +"(> c #FE6342", +"_> c #FE6645", +":> c #FE6646", +"<> c #FE6147", +"[> c #FE6048", +"}> c #FE6148", +"|> c #FE6746", +"1> c #FE6A46", +"2> c #FE6F45", +"3> c #FE7441", +"4> c #FC7D39", +"5> c #6C422E", +"6> c #0F0F23", +"7> c #FA4142", +"8> c #FC4643", +"9> c #FE4D46", +"0> c #FE4E47", +"a> c #FE4F48", +"b> c #FE5148", +"c> c #FE5348", +"d> c #FE5548", +"e> c #FE5247", +"f> c #FD5445", +"g> c #FC5544", +"h> c #F96041", +"i> c #D33F3D", +"j> c #392D39", +"k> c #973C38", +"l> c #F94E3A", +"m> c #FD693E", +"n> c #FE6C43", +"o> c #FE6047", +"p> c #FE5D47", +"q> c #FE5E48", +"r> c #FE6948", +"s> c #FE6947", +"t> c #FE6B47", +"u> c #FE6E46", +"v> c #FD6D43", +"w> c #FB723D", +"x> c #D54A33", +"y> c #301C29", +"z> c #FB4A42", +"A> c #FD4B44", +"B> c #FE4F47", +"C> c #FE5048", +"D> c #FE5648", +"E> c #FE5848", +"F> c #FE5747", +"G> c #FE5547", +"H> c #FC5945", +"I> c #F95742", +"J> c #F3543D", +"K> c #A33336", +"L> c #302032", +"M> c #152433", +"N> c #CD3E38", +"O> c #FD5A3F", +"P> c #FE6343", +"Q> c #FE6446", +"R> c #FE6247", +"S> c #FE6A47", +"T> c #FC6542", +"U> c #FB6A3B", +"V> c #FA6D34", +"W> c #D73C2D", +"X> c #442428", +"Y> c #281323", +"Z> c #FD4E42", +"`> c #FD4D43", +" , c #FE4D45", +"., c #FE5248", +"+, c #FE5947", +"@, c #FE5C47", +"#, c #FE5B47", +"$, c #FE5A47", +"%, c #FE5847", +"&, c #FC5C45", +"*, c #F95B43", +"=, c #F3613F", +"-, c #E74F37", +";, c #8C2431", +">, c #161E2F", +",, c #CD4E33", +"', c #FD503A", +"), c #FE5D40", +"!, c #FE6445", +"~, c #FE6946", +"{, c #FE6847", +"], c #FE6747", +"^, c #FD6644", +"/, c #FD6241", +"(, c #FD5B3D", +"_, c #FE6739", +":, c #FE6135", +"<, c #AB4830", +"[, c #733E2A", +"}, c #161224", +"|, c #FC4E42", +"1, c #FE4D44", +"2, c #FE4E46", +"3, c #FE5147", +"4, c #FE5E47", +"5, c #FD5C46", +"6, c #FA5B44", +"7, c #F45441", +"8, c #EB393A", +"9, c #CC3433", +"0, c #47212F", +"a, c #59242F", +"b, c #FC6734", +"c, c #FC6F3A", +"d, c #FC723E", +"e, c #FD6540", +"f, c #FE6442", +"g, c #FE6643", +"h, c #FE6944", +"i, c #FE6546", +"j, c #FE6444", +"k, c #FE6143", +"l, c #FE5E41", +"m, c #FE613F", +"n, c #FE683C", +"o, c #FE7937", +"p, c #A25030", +"q, c #692629", +"r, c #151122", +"s, c #FA573F", +"t, c #FB4D40", +"u, c #FC4F43", +"v, c #FE5246", +"w, c #FF6347", +"x, c #FE5F48", +"y, c #F65942", +"z, c #F0493D", +"A, c #ED3736", +"B, c #73262F", +"C, c #10152C", +"D, c #3B292F", +"E, c #363034", +"F, c #AC3938", +"G, c #FC6B3B", +"H, c #FD763C", +"I, c #FE6D3F", +"J, c #FE6341", +"K, c #FE6642", +"L, c #FE6745", +"M, c #FE6245", +"N, c #FE6244", +"O, c #FE6841", +"P, c #FF683B", +"Q, c #EC7035", +"R, c #D0412D", +"S, c #3A1627", +"T, c #CF3938", +"U, c #F6543C", +"V, c #FB5040", +"W, c #FD5544", +"X, c #FE5A48", +"Y, c #FE5D48", +"Z, c #FE5F47", +"`, c #FF6147", +" ' c #FD5C45", +".' c #FB5B43", +"+' c #FA5A42", +"@' c #F76040", +"#' c #F4623D", +"$' c #F26D38", +"%' c #EC4130", +"&' c #380E2B", +"*' c #13122C", +"=' c #362D31", +"-' c #353435", +";' c #352E37", +">' c #2D3337", +",' c #CC5838", +"'' c #CD6F3A", +")' c #CE6E3D", +"!' c #FE793F", +"~' c #FD7541", +"{' c #FD6243", +"]' c #FE6545", +"^' c #FF6543", +"/' c #FF6240", +"(' c #FE723B", +"_' c #FE8034", +":' c #442D2C", +"<' c #311725", +"[' c #222830", +"}' c #B73B36", +"|' c #F94C3D", +"1' c #FD5543", +"2' c #FE5B48", +"3' c #FF5E47", +"4' c #FE5C48", +"5' c #FC5B44", +"6' c #F95640", +"7' c #C34E3D", +"8' c #A45A3A", +"9' c #F37438", +"0' c #F28935", +"a' c #AF422F", +"b' c #240D2B", +"c' c #88292F", +"d' c #FA8E34", +"e' c #FC7E38", +"f' c #FC5939", +"g' c #694A37", +"h' c #693437", +"i' c #382638", +"j' c #142439", +"k' c #9F483A", +"l' c #C45E3C", +"m' c #FD7240", +"n' c #FF6645", +"o' c #FF6245", +"p' c #FF6045", +"q' c #FF6146", +"r' c #FF6246", +"s' c #FF6446", +"t' c #FF6545", +"u' c #FE763F", +"v' c #FE7237", +"w' c #C65331", +"x' c #3D272A", +"y' c #0D1E2B", +"z' c #683032", +"A' c #F9453A", +"B' c #FD5341", +"C' c #FE5A46", +"D' c #FF5A48", +"E' c #FE5948", +"F' c #FD5A47", +"G' c #FC5D43", +"H' c #F95B3D", +"I' c #713F37", +"J' c #1E2D32", +"K' c #C44531", +"L' c #EF7A2F", +"M' c #6B2E2C", +"N' c #0F0E2C", +"O' c #F56633", +"P' c #FA803A", +"Q' c #FC673E", +"R' c #FD673E", +"S' c #FC6F3C", +"T' c #FA6E3B", +"U' c #C6633A", +"V' c #A06739", +"W' c #835638", +"X' c #381F38", +"Y' c #713B38", +"Z' c #7B503C", +"`' c #FE7741", +" ) c #FE7344", +".) c #FE6D46", +"+) c #FF6946", +"@) c #FF5E46", +"#) c #FF5D46", +"$) c #FF5D47", +"%) c #FF5F48", +"&) c #FF6248", +"*) c #FE6941", +"=) c #FC783C", +"-) c #C46B35", +";) c #892730", +">) c #111629", +",) c #1F2630", +"') c #AD3939", +")) c #FC5D41", +"!) c #FE5946", +"~) c #FF5848", +"{) c #FE5549", +"]) c #FC5E42", +"^) c #FA673B", +"/) c #DB7033", +"() c #392E2B", +"_) c #311A28", +":) c #3C2127", +"<) c #1D1027", +"[) c #92102C", +"}) c #F58336", +"|) c #FA673E", +"1) c #FD6642", +"2) c #FD5A41", +"3) c #FC6D41", +"4) c #FC6D3F", +"5) c #FD683E", +"6) c #F38C39", +"7) c #CE6535", +"8) c #612E34", +"9) c #1D2637", +"0) c #71513E", +"a) c #FF6847", +"b) c #FF5F47", +"c) c #FF5A46", +"d) c #FF5847", +"e) c #FF5748", +"f) c #FF594A", +"g) c #FF5E4B", +"h) c #FE654C", +"i) c #FE694B", +"j) c #FE6B48", +"k) c #FC6A43", +"l) c #F7683E", +"m) c #EC6E39", +" ", +" ", +" ", +" ", +" ", +" ", +" ", +" ", +" . + @ # $ % ", +" & * = - ; > , ' ) ! ~ ", +" { ] ^ / ( _ : < [ } | 1 2 ", +" 3 4 5 6 7 8 9 0 a b c d e f g h i j ", +" k l m n o p q r s t u v w x y z A B C D ", +" E F G H I J K L M N O P Q R S T U V W X Y Z ` ", +" ...+.@.#.$.%.&.*.=.-.;.>.,.S '.).!.~.{.].^./.(._. ", +" :.<.[.}.|.1.2.3.4.5.6.7.8.9.0.a.b.c.d.e.!.S f.g.h.i.j.k. ", +" l.m.n.o.p.q.r.s.t.u.J v.w.x.y.z.A.c.d.d.B.C.D.E.F.G.H.I. ", +" J.K.L.M.N.O.P.Q.R.t S.T.U.V.W.X.Y.Z.`. +d.d..+B.'.++@+#+$+%+ ", +" &+*+=+-+;+>+,+'+)+!+~+{+]+^+/+(+_+:+<+[+}+|+1+d.2+3+4+d.5+6+7+8+9+0+ ", +" a+b+c+d+e+f+g+h+i+j+k+l+m+n+^+o+p+q+r+s+t+u+v+b.w+x+y+z+A+w+B+C+D+E+F+G+ ", +" H+I+J+K+L+M+N+O+P+Q+R+S+T+U+V+W+Q ,.X+Y+Z+`+ @.@+@@@#@$@%@&@*@=@#@-@;@>@,@'@ ", +" )@!@~@{@]@^@/@(@_@:@<@[@}@|@1@2@3@R ,.4@5@6@7@8@9@0@a@#@b@c@=@d@e@f@g@>@h@i@j@ ", +" k@l@m@n@o@p@q@r@s@t@u@v@w@x@y@^+R S z@A@z.q+B@C@D@E@F@G@H@#@e@#@#@f@g@I@J@K@L@ ", +" M@N@O@P@Q@R@S@T@U@V@W@X@Y@Z@`@ #.#+#+#S A@@###$#%#&#*#=#-#f@B+B+B+f@;#>#,#'#)# ", +" !#~#{#]#^#/#(#(#_#:#<#[#}#|#1#^+.#S +#+#z@2#3#4#5#6#7#8#9#0#A.B+B+a#A.@@b#c#d# ", +" e#f#g#h#i#j#k#l#m#n#o#p#q#r#s#t#u#v#.#w#S R ^+x#y#z#A#B#C#D#-#A.a#`.`.b.g@E#d#F# ", +" G#0@H#I#J#K#L#M#N#O#P#Q#R#S#T#U#V#>.W#3@v#R R X+X#Y#s#Z#`# $.$+$@$g@f@5+5+#$6+$$%$&$ ", +" *$=$-$;$>$,$'$)$!$~${$]$^$/$($_$*$u#:$Q 3@,.X+z.<$[$}$|$1$2$3$4$5$6$7$e@8$#$G@9$0$a$ ", +" ,.4@E#b$c$d$e$f$g$h$i$j$k$l$m$n$`@>.:$o$3@,. #a.p$q$r$s$t$u$v$w$x$y$z$A$B$#@C$D$E$F$G$ ", +" R S H$v+I$J$K$n+L$:$o$o$M$N$L$O$P$Q$R$N$o$3@S$T$U$V$W$X$Y$Z$`$ %.%+%@%#%$%%%&%*%=%-%;%>% ", +" E.,%~.'%Z.4@v W#o$)%)%)%Q !%~%{%]%^%Q$u u#/%(%_%:%<%[%}%|%1%2%3%4%=%5%6%7%8%9%0%a%b%c%d% ", +" e%f%g%a#,%,%z@R 3@3@3@)%Q h%i%j%k%l%m%{+n%o%p%q%r%s%t%u%v%w%x%y%z%A%*%B%C%D%E%F%G%H%I% ", +" J%K%L%M%N%D.S v#)%)%O%P%Q%R%S%T%U%V%W%X%Y%Z%`% &.&+&@&#&$&%&&&*&f@a##@=&-&;&>&,&'&)& ", +" !&~&{&]&^&.#w#^&/%/&(&_&:&<&[&}&|&1&2&3&4&5&6&7&8&9&0&a&b&c&d&e&e@1+5+e@f&g&h&i&j& ", +" k&l&m&n&o&p&q&r&i%s&3.t&u&v&w&x&y&z&A&B&C&D&E&F&G&H&I&J&K&L&M&N&O&P&1+`.e@f&Q&R&S&T& ", +" 0 U&V&W&X&<&Y&j%Z&`& *.*+*@*#*$*%*&***=*-*;*>*>*,*'*)*!*~*{*]*^*/*(*a#B+#@_*:*<*[*}* ", +" |*1*2*3*4*5*6*7*8*9*0*a*b*c*d*e*f*g*h*i*j*k*l*m*n*o*p*q*r*s*t*u*v*E.w*d.e@x*y*z*A*B* ", +" C*D*E*F*G*H*I*J*K*L*M*N*O*P*Q*R*S*T*U*V*W*l*X*o*o*Y*Z*`* =.=+=@=#='%$=%=e@&=*===-=;= ", +" >=,='=)=!=~={=]=^=/=(=_=:=<=[=}=|=1=2=3=4=5=p*6=6=7=8=9=0=a=b=c=d=A@~.b.B+e=f=g=h=i= ", +" j=k=l=m=n=o=p=q=r=s=t=u=v=w=x=y=z=A=5=Z*B=C=D=E=8=F=G=H=I=J=K=S$R z@'%L=M=N=O= ", +" P=Q=R=S=T=U=V=W=X=Y=Z=`= -.-+-@-#-$-5=p*E=D=%-%-q*&-*-=---;->-,-/%3@^+'-)-!-~- ", +" {-]-^-/-(-_-:-<-[-}-|-1-2- -3-4-5-6-7-8-n*m*9-0-9-o*a-b-c-d-e-f-g-(&h%w c h-i- ", +" j-k-l-m-n-o-p-q-r-s-t-u-v-w-,-x-y-z-A-B-C-D-E-E-F-G-_@m*H-I-J-K-L-M-N-O-P-(+Q- ", +" R-S-T-U-V-W-X-Y-Z-`- ;.;+;@;#;$;%;&;*;=;-;-;;;>;,;';);!;~;{;];^;/;(;_;:;<;[;}; ", +" |;1;2;3;4;5;6;7;8;9;0;a;b;c;d;e;f;g;h;i;j;j;k;k;l;m;n;o;p;q;r;s;t;u;v;w;x;y; ", +" z;A;B;C;D;E;F;G;H;I;J;K;L;M;N;O;P;Q;R;S;T;U;V;W;X;k;Y;Z;`; >r;.>+>@>#>$> ", +" %>&>*>=>->;>>>,>'>,>)>F;8;!>~>{>]>^>/>(>_>:>i;<>[>X;}>i;|>1>q;2>3>4>5>6> ", +" 7>8>=>9>0>a>b>c>d>,>e>e>f>g>h>i>j>k>l>m>n>:>i;o>p>q>W;r>s>t>p;u>v>w>x>y> ", +" z>A>9>0>B>C>c>D>E>F>G>G>F>H>I>J>K>L>M>N>O>P>Q>R>o>R>T;s>S>S>S>t>1>T>U>V>W>X>Y> ", +" Z>`> ,9>B>.,D>+,@,#,$,%,$,&,*,=,-,;,>,,,',),P>!,!,_>~,t>s>{,],{,],^,/,(,_,:,<,[,}, ", +" |,`>1,2,3,G>+,4,o>o>4,@,@,5,6,7,8,9,0,a,b,c,d,e,f,g,h, >~,|>T;T;T;i,j,k,l,m,n,o,p,q,r, ", +" s,t,u,v,G>%,@,o>w,R>x,p>@,5,6,y,z,A,B,C,D,E,F,G,H,I,J,K,L,L,i,i;i;i;Q>S;M,N,P>O,P,Q,R,S, ", +" T,U,V,W,%,X,Y,Z,`,[>q>@, '.'+'@'#'$'%'&'*'='-';'>',''')'!'~'{'N,i,:>_>]'M,M,Q>_>^'/'('_':'<' ", +" ['}'|'1'$,X,2'p>3'4'2'@,5'6'7'8'9'0'a'b'c'd'e'f'g'h'i'j'k'l'd,m'g, > >n'o'p'q'r's't'.>u'v'w'x' ", +" y'z'A'B'C'X,X,2'D'E'E'F'G'H'I'J'K'L'M'N'O'P'Q'R'S'T'U'V'W'X'Y'Z'`' ).)+)r'@)#)$)%)&)l;1>*)=)-);) ", +" >),)')))!)X,E'X,~){)d>!)])^)/)()_):)<)[)})|)1)f,2)3)4)5)6)7)8)9)0)*--*a)b)c)d)e)f)g)h)i)j)k)l)m) ", +" ", +" ", +" ", +" ", +" ", +" ", +" ", +" "}; diff --git a/lib/json/pure.rb b/lib/json/pure.rb new file mode 100644 index 0000000000..b86d905523 --- /dev/null +++ b/lib/json/pure.rb @@ -0,0 +1,75 @@ +require 'json/common' +require 'json/pure/parser' +require 'json/pure/generator' + +module JSON + begin + require 'iconv' + # An iconv instance to convert from UTF8 to UTF16 Big Endian. + UTF16toUTF8 = Iconv.new('utf-8', 'utf-16be') # :nodoc: + # An iconv instance to convert from UTF16 Big Endian to UTF8. + UTF8toUTF16 = Iconv.new('utf-16be', 'utf-8') # :nodoc: + UTF8toUTF16.iconv('no bom') + rescue Errno::EINVAL, Iconv::InvalidEncoding + # Iconv doesn't support big endian utf-16. Let's try to hack this manually + # into the converters. + begin + old_verbose, $VERBSOSE = $VERBOSE, nil + # An iconv instance to convert from UTF8 to UTF16 Big Endian. + UTF16toUTF8 = Iconv.new('utf-8', 'utf-16') # :nodoc: + # An iconv instance to convert from UTF16 Big Endian to UTF8. + UTF8toUTF16 = Iconv.new('utf-16', 'utf-8') # :nodoc: + UTF8toUTF16.iconv('no bom') + if UTF8toUTF16.iconv("\xe2\x82\xac") == "\xac\x20" + swapper = Class.new do + def initialize(iconv) # :nodoc: + @iconv = iconv + end + + def iconv(string) # :nodoc: + result = @iconv.iconv(string) + JSON.swap!(result) + end + end + UTF8toUTF16 = swapper.new(UTF8toUTF16) # :nodoc: + end + if UTF16toUTF8.iconv("\xac\x20") == "\xe2\x82\xac" + swapper = Class.new do + def initialize(iconv) # :nodoc: + @iconv = iconv + end + + def iconv(string) # :nodoc: + string = JSON.swap!(string.dup) + @iconv.iconv(string) + end + end + UTF16toUTF8 = swapper.new(UTF16toUTF8) # :nodoc: + end + rescue Errno::EINVAL, Iconv::InvalidEncoding + raise MissingUnicodeSupport, "iconv doesn't seem to support UTF-8/UTF-16 conversions" + ensure + $VERBOSE = old_verbose + end + rescue LoadError + raise MissingUnicodeSupport, + "iconv couldn't be loaded, which is required for UTF-8/UTF-16 conversions" + end + + # Swap consecutive bytes of _string_ in place. + def self.swap!(string) # :nodoc: + 0.upto(string.size / 2) do |i| + break unless string[2 * i + 1] + string[2 * i], string[2 * i + 1] = string[2 * i + 1], string[2 * i] + end + string + end + + # This module holds all the modules/classes that implement JSON's + # functionality in pure ruby. + module Pure + $DEBUG and warn "Using pure library for JSON." + JSON.parser = Parser + JSON.generator = Generator + end +end diff --git a/lib/json/pure/generator.rb b/lib/json/pure/generator.rb new file mode 100644 index 0000000000..498d78a660 --- /dev/null +++ b/lib/json/pure/generator.rb @@ -0,0 +1,321 @@ +module JSON + MAP = { + "\x0" => '\u0000', + "\x1" => '\u0001', + "\x2" => '\u0002', + "\x3" => '\u0003', + "\x4" => '\u0004', + "\x5" => '\u0005', + "\x6" => '\u0006', + "\x7" => '\u0007', + "\b" => '\b', + "\t" => '\t', + "\n" => '\n', + "\xb" => '\u000b', + "\f" => '\f', + "\r" => '\r', + "\xe" => '\u000e', + "\xf" => '\u000f', + "\x10" => '\u0010', + "\x11" => '\u0011', + "\x12" => '\u0012', + "\x13" => '\u0013', + "\x14" => '\u0014', + "\x15" => '\u0015', + "\x16" => '\u0016', + "\x17" => '\u0017', + "\x18" => '\u0018', + "\x19" => '\u0019', + "\x1a" => '\u001a', + "\x1b" => '\u001b', + "\x1c" => '\u001c', + "\x1d" => '\u001d', + "\x1e" => '\u001e', + "\x1f" => '\u001f', + '"' => '\"', + '\\' => '\\\\', + '/' => '\/', + } # :nodoc: + + # Convert a UTF8 encoded Ruby string _string_ to a JSON string, encoded with + # UTF16 big endian characters as \u????, and return it. + def utf8_to_json(string) # :nodoc: + string = string.gsub(/["\\\/\x0-\x1f]/) { |c| MAP[c] } + string.gsub!(/( + (?: + [\xc2-\xdf][\x80-\xbf] | + [\xe0-\xef][\x80-\xbf]{2} | + [\xf0-\xf4][\x80-\xbf]{3} + )+ | + [\x80-\xc1\xf5-\xff] # invalid + )/nx) { |c| + c.size == 1 and raise GeneratorError, "invalid utf8 byte: '#{c}'" + s = JSON::UTF8toUTF16.iconv(c).unpack('H*')[0] + s.gsub!(/.{4}/n, '\\\\u\&') + } + string + rescue Iconv::Failure => e + raise GeneratorError, "Caught #{e.class}: #{e}" + end + module_function :utf8_to_json + + module Pure + module Generator + # This class is used to create State instances, that are use to hold data + # while generating a JSON text from a a Ruby data structure. + class State + # Creates a State object from _opts_, which ought to be Hash to create + # a new State instance configured by _opts_, something else to create + # an unconfigured instance. If _opts_ is a State object, it is just + # returned. + def self.from_state(opts) + case opts + when self + opts + when Hash + new(opts) + else + new + end + end + + # Instantiates a new State object, configured by _opts_. + # + # _opts_ can have the following keys: + # + # * *indent*: a string used to indent levels (default: ''), + # * *space*: a string that is put after, a : or , delimiter (default: ''), + # * *space_before*: a string that is put before a : pair delimiter (default: ''), + # * *object_nl*: a string that is put at the end of a JSON object (default: ''), + # * *array_nl*: a string that is put at the end of a JSON array (default: ''), + # * *check_circular*: true if checking for circular data structures + # should be done, false (the default) otherwise. + def initialize(opts = {}) + @indent = opts[:indent] || '' + @space = opts[:space] || '' + @space_before = opts[:space_before] || '' + @object_nl = opts[:object_nl] || '' + @array_nl = opts[:array_nl] || '' + @check_circular = !!(opts[:check_circular] || false) + @seen = {} + end + + # This string is used to indent levels in the JSON text. + attr_accessor :indent + + # This string is used to insert a space between the tokens in a JSON + # string. + attr_accessor :space + + # This string is used to insert a space before the ':' in JSON objects. + attr_accessor :space_before + + # This string is put at the end of a line that holds a JSON object (or + # Hash). + attr_accessor :object_nl + + # This string is put at the end of a line that holds a JSON array. + attr_accessor :array_nl + + # Returns true, if circular data structures should be checked, + # otherwise returns false. + def check_circular? + @check_circular + end + + # Returns _true_, if _object_ was already seen during this generating + # run. + def seen?(object) + @seen.key?(object.__id__) + end + + # Remember _object_, to find out if it was already encountered (if a + # cyclic data structure is if a cyclic data structure is rendered). + def remember(object) + @seen[object.__id__] = true + end + + # Forget _object_ for this generating run. + def forget(object) + @seen.delete object.__id__ + end + end + + module GeneratorMethods + module Object + # Converts this object to a string (calling #to_s), converts + # it to a JSON string, and returns the result. This is a fallback, if no + # special method #to_json was defined for some object. + def to_json(*) to_s.to_json end + end + + module Hash + # Returns a JSON string containing a JSON object, that is unparsed from + # this Hash instance. + # _state_ is a JSON::State object, that can also be used to configure the + # produced JSON string output further. + # _depth_ is used to find out nesting depth, to indent accordingly. + def to_json(state = nil, depth = 0, *) + if state + state = JSON.state.from_state(state) + json_check_circular(state) { json_transform(state, depth) } + else + json_transform(state, depth) + end + end + + private + + def json_check_circular(state) + if state + state.seen?(self) and raise JSON::CircularDatastructure, + "circular data structures not supported!" + state.remember self + end + yield + ensure + state and state.forget self + end + + def json_shift(state, depth) + state and not state.object_nl.empty? or return '' + state.indent * depth + end + + def json_transform(state, depth) + delim = ',' + delim << state.object_nl if state + result = '{' + result << state.object_nl if state + result << map { |key,value| + s = json_shift(state, depth + 1) + s << key.to_s.to_json(state, depth + 1) + s << state.space_before if state + s << ':' + s << state.space if state + s << value.to_json(state, depth + 1) + }.join(delim) + result << state.object_nl if state + result << json_shift(state, depth) + result << '}' + result + end + end + + module Array + # Returns a JSON string containing a JSON array, that is unparsed from + # this Array instance. + # _state_ is a JSON::State object, that can also be used to configure the + # produced JSON string output further. + # _depth_ is used to find out nesting depth, to indent accordingly. + def to_json(state = nil, depth = 0, *) + if state + state = JSON.state.from_state(state) + json_check_circular(state) { json_transform(state, depth) } + else + json_transform(state, depth) + end + end + + private + + def json_check_circular(state) + if state + state.seen?(self) and raise JSON::CircularDatastructure, + "circular data structures not supported!" + state.remember self + end + yield + ensure + state and state.forget self + end + + def json_shift(state, depth) + state and not state.array_nl.empty? or return '' + state.indent * depth + end + + def json_transform(state, depth) + delim = ',' + delim << state.array_nl if state + result = '[' + result << state.array_nl if state + result << map { |value| + json_shift(state, depth + 1) << value.to_json(state, depth + 1) + }.join(delim) + result << state.array_nl if state + result << json_shift(state, depth) + result << ']' + result + end + end + + module Integer + # Returns a JSON string representation for this Integer number. + def to_json(*) to_s end + end + + module Float + # Returns a JSON string representation for this Float number. + def to_json(*) to_s end + end + + module String + # This string should be encoded with UTF-8 A call to this method + # returns a JSON string encoded with UTF16 big endian characters as + # \u????. + def to_json(*) + '"' << JSON.utf8_to_json(self) << '"' + end + + # Module that holds the extinding methods if, the String module is + # included. + module Extend + # Raw Strings are JSON Objects (the raw bytes are stored in an array for the + # key "raw"). The Ruby String can be created by this module method. + def json_create(o) + o['raw'].pack('C*') + end + end + + # Extends _modul_ with the String::Extend module. + def self.included(modul) + modul.extend Extend + end + + # This method creates a raw object hash, that can be nested into + # other data structures and will be unparsed as a raw string. This + # method should be used, if you want to convert raw strings to JSON + # instead of UTF-8 strings, e. g. binary data. + def to_json_raw_object + { + JSON.create_id => self.class.name, + 'raw' => self.unpack('C*'), + } + end + + # This method creates a JSON text from the result of + # a call to to_json_raw_object of this String. + def to_json_raw(*args) + to_json_raw_object.to_json(*args) + end + end + + module TrueClass + # Returns a JSON string for true: 'true'. + def to_json(*) 'true' end + end + + module FalseClass + # Returns a JSON string for false: 'false'. + def to_json(*) 'false' end + end + + module NilClass + # Returns a JSON string for nil: 'null'. + def to_json(*) 'null' end + end + end + end + end +end diff --git a/lib/json/pure/parser.rb b/lib/json/pure/parser.rb new file mode 100644 index 0000000000..6118fe489e --- /dev/null +++ b/lib/json/pure/parser.rb @@ -0,0 +1,241 @@ +require 'strscan' + +module JSON + module Pure + # This class implements the JSON parser that is used to parse a JSON string + # into a Ruby data structure. + class Parser < StringScanner + STRING = /" ((?:[^\x0-\x1f"\\] | + \\["\\\/bfnrt] | + \\u[0-9a-fA-F]{4} | + \\[\x20-\xff])*) + "/nx + INTEGER = /(-?0|-?[1-9]\d*)/ + FLOAT = /(-? + (?:0|[1-9]\d*) + (?: + \.\d+(?i:e[+-]?\d+) | + \.\d+ | + (?i:e[+-]?\d+) + ) + )/x + OBJECT_OPEN = /\{/ + OBJECT_CLOSE = /\}/ + ARRAY_OPEN = /\[/ + ARRAY_CLOSE = /\]/ + PAIR_DELIMITER = /:/ + COLLECTION_DELIMITER = /,/ + TRUE = /true/ + FALSE = /false/ + NULL = /null/ + IGNORE = %r( + (?: + //[^\n\r]*[\n\r]| # line comments + /\* # c-style comments + (?: + [^*/]| # normal chars + /[^*]| # slashes that do not start a nested comment + \*[^/]| # asterisks that do not end this comment + /(?=\*/) # single slash before this comment's end + )* + \*/ # the End of this comment + |[ \t\r\n]+ # whitespaces: space, horicontal tab, lf, cr + )+ + )mx + + UNPARSED = Object.new + + # Creates a new JSON::Pure::Parser instance for the string _source_. + # + # It will be configured by the _opts_ hash. _opts_ can have the following + # keys: + # * *max_nesting*: The maximum depth of nesting allowed in the parsed data + # structures. Disable depth checking with :max_nesting => false. + def initialize(source, opts = {}) + super + if !opts.key?(:max_nesting) # defaults to 19 + @max_nesting = 19 + elsif opts[:max_nesting] + @max_nesting = opts[:max_nesting] + else + @max_nesting = 0 + end + @create_id = JSON.create_id + end + + alias source string + + # Parses the current JSON string _source_ and returns the complete data + # structure as a result. + def parse + reset + obj = nil + until eos? + case + when scan(OBJECT_OPEN) + obj and raise ParserError, "source '#{peek(20)}' not in JSON!" + @current_nesting = 1 + obj = parse_object + when scan(ARRAY_OPEN) + obj and raise ParserError, "source '#{peek(20)}' not in JSON!" + @current_nesting = 1 + obj = parse_array + when skip(IGNORE) + ; + else + raise ParserError, "source '#{peek(20)}' not in JSON!" + end + end + obj or raise ParserError, "source did not contain any JSON!" + obj + end + + private + + # Unescape characters in strings. + UNESCAPE_MAP = Hash.new { |h, k| h[k] = k.chr } + UNESCAPE_MAP.update({ + ?" => '"', + ?\\ => '\\', + ?/ => '/', + ?b => "\b", + ?f => "\f", + ?n => "\n", + ?r => "\r", + ?t => "\t", + ?u => nil, + }) + + def parse_string + if scan(STRING) + return '' if self[1].empty? + self[1].gsub(%r((?:\\[\\bfnrt"/]|(?:\\u(?:[A-Fa-f\d]{4}))+|\\[\x20-\xff]))n) do |c| + if u = UNESCAPE_MAP[c[1]] + u + else # \uXXXX + bytes = '' + i = 0 + while c[6 * i] == ?\\ && c[6 * i + 1] == ?u + bytes << c[6 * i + 2, 2].to_i(16) << c[6 * i + 4, 2].to_i(16) + i += 1 + end + JSON::UTF16toUTF8.iconv(bytes) + end + end + else + UNPARSED + end + rescue Iconv::Failure => e + raise GeneratorError, "Caught #{e.class}: #{e}" + end + + def parse_value + case + when scan(FLOAT) + Float(self[1]) + when scan(INTEGER) + Integer(self[1]) + when scan(TRUE) + true + when scan(FALSE) + false + when scan(NULL) + nil + when (string = parse_string) != UNPARSED + string + when scan(ARRAY_OPEN) + @current_nesting += 1 + ary = parse_array + @current_nesting -= 1 + ary + when scan(OBJECT_OPEN) + @current_nesting += 1 + obj = parse_object + @current_nesting -= 1 + obj + else + UNPARSED + end + end + + def parse_array + raise NestingError, "nesting of #@current_nesting is to deep" if + @max_nesting.nonzero? && @current_nesting > @max_nesting + result = [] + delim = false + until eos? + case + when (value = parse_value) != UNPARSED + delim = false + result << value + skip(IGNORE) + if scan(COLLECTION_DELIMITER) + delim = true + elsif match?(ARRAY_CLOSE) + ; + else + raise ParserError, "expected ',' or ']' in array at '#{peek(20)}'!" + end + when scan(ARRAY_CLOSE) + if delim + raise ParserError, "expected next element in array at '#{peek(20)}'!" + end + break + when skip(IGNORE) + ; + else + raise ParserError, "unexpected token in array at '#{peek(20)}'!" + end + end + result + end + + def parse_object + raise NestingError, "nesting of #@current_nesting is to deep" if + @max_nesting.nonzero? && @current_nesting > @max_nesting + result = {} + delim = false + until eos? + case + when (string = parse_string) != UNPARSED + skip(IGNORE) + unless scan(PAIR_DELIMITER) + raise ParserError, "expected ':' in object at '#{peek(20)}'!" + end + skip(IGNORE) + unless (value = parse_value).equal? UNPARSED + result[string] = value + delim = false + skip(IGNORE) + if scan(COLLECTION_DELIMITER) + delim = true + elsif match?(OBJECT_CLOSE) + ; + else + raise ParserError, "expected ',' or '}' in object at '#{peek(20)}'!" + end + else + raise ParserError, "expected value in object at '#{peek(20)}'!" + end + when scan(OBJECT_CLOSE) + if delim + raise ParserError, "expected next name, value pair in object at '#{peek(20)}'!" + end + if klassname = result[@create_id] + klass = JSON.deep_const_get klassname + break unless klass and klass.json_creatable? + result = klass.json_create(result) + result + end + break + when skip(IGNORE) + ; + else + raise ParserError, "unexpected token in object at '#{peek(20)}'!" + end + end + result + end + end + end +end diff --git a/lib/json/version.rb b/lib/json/version.rb new file mode 100755 index 0000000000..464bb60ee3 --- /dev/null +++ b/lib/json/version.rb @@ -0,0 +1,9 @@ +module JSON + # JSON version + VERSION = '1.1.0' + VERSION_ARRAY = VERSION.split(/\./).map { |x| x.to_i } # :nodoc: + VERSION_MAJOR = VERSION_ARRAY[0] # :nodoc: + VERSION_MINOR = VERSION_ARRAY[1] # :nodoc: + VERSION_BUILD = VERSION_ARRAY[2] # :nodoc: + VARIANT_BINARY = false +end diff --git a/test/json/fixtures/fail1.json b/test/json/fixtures/fail1.json new file mode 100644 index 0000000000..6216b865f1 --- /dev/null +++ b/test/json/fixtures/fail1.json @@ -0,0 +1 @@ +"A JSON payload should be an object or array, not a string." \ No newline at end of file diff --git a/test/json/fixtures/fail10.json b/test/json/fixtures/fail10.json new file mode 100644 index 0000000000..5d8c0047bd --- /dev/null +++ b/test/json/fixtures/fail10.json @@ -0,0 +1 @@ +{"Extra value after close": true} "misplaced quoted value" \ No newline at end of file diff --git a/test/json/fixtures/fail11.json b/test/json/fixtures/fail11.json new file mode 100644 index 0000000000..76eb95b458 --- /dev/null +++ b/test/json/fixtures/fail11.json @@ -0,0 +1 @@ +{"Illegal expression": 1 + 2} \ No newline at end of file diff --git a/test/json/fixtures/fail12.json b/test/json/fixtures/fail12.json new file mode 100644 index 0000000000..77580a4522 --- /dev/null +++ b/test/json/fixtures/fail12.json @@ -0,0 +1 @@ +{"Illegal invocation": alert()} \ No newline at end of file diff --git a/test/json/fixtures/fail13.json b/test/json/fixtures/fail13.json new file mode 100644 index 0000000000..379406b59b --- /dev/null +++ b/test/json/fixtures/fail13.json @@ -0,0 +1 @@ +{"Numbers cannot have leading zeroes": 013} \ No newline at end of file diff --git a/test/json/fixtures/fail14.json b/test/json/fixtures/fail14.json new file mode 100644 index 0000000000..0ed366b38a --- /dev/null +++ b/test/json/fixtures/fail14.json @@ -0,0 +1 @@ +{"Numbers cannot be hex": 0x14} \ No newline at end of file diff --git a/test/json/fixtures/fail18.json b/test/json/fixtures/fail18.json new file mode 100644 index 0000000000..d61a345465 --- /dev/null +++ b/test/json/fixtures/fail18.json @@ -0,0 +1 @@ +[[[[[[[[[[[[[[[[[[[["Too deep"]]]]]]]]]]]]]]]]]]]] // No, we don't limit our depth: Moved to pass... diff --git a/test/json/fixtures/fail19.json b/test/json/fixtures/fail19.json new file mode 100644 index 0000000000..3b9c46fa9a --- /dev/null +++ b/test/json/fixtures/fail19.json @@ -0,0 +1 @@ +{"Missing colon" null} \ No newline at end of file diff --git a/test/json/fixtures/fail2.json b/test/json/fixtures/fail2.json new file mode 100644 index 0000000000..6b7c11e5a5 --- /dev/null +++ b/test/json/fixtures/fail2.json @@ -0,0 +1 @@ +["Unclosed array" \ No newline at end of file diff --git a/test/json/fixtures/fail20.json b/test/json/fixtures/fail20.json new file mode 100644 index 0000000000..27c1af3e72 --- /dev/null +++ b/test/json/fixtures/fail20.json @@ -0,0 +1 @@ +{"Double colon":: null} \ No newline at end of file diff --git a/test/json/fixtures/fail21.json b/test/json/fixtures/fail21.json new file mode 100644 index 0000000000..62474573b2 --- /dev/null +++ b/test/json/fixtures/fail21.json @@ -0,0 +1 @@ +{"Comma instead of colon", null} \ No newline at end of file diff --git a/test/json/fixtures/fail22.json b/test/json/fixtures/fail22.json new file mode 100644 index 0000000000..a7752581bc --- /dev/null +++ b/test/json/fixtures/fail22.json @@ -0,0 +1 @@ +["Colon instead of comma": false] \ No newline at end of file diff --git a/test/json/fixtures/fail23.json b/test/json/fixtures/fail23.json new file mode 100644 index 0000000000..494add1ca1 --- /dev/null +++ b/test/json/fixtures/fail23.json @@ -0,0 +1 @@ +["Bad value", truth] \ No newline at end of file diff --git a/test/json/fixtures/fail24.json b/test/json/fixtures/fail24.json new file mode 100644 index 0000000000..caff239bfc --- /dev/null +++ b/test/json/fixtures/fail24.json @@ -0,0 +1 @@ +['single quote'] \ No newline at end of file diff --git a/test/json/fixtures/fail25.json b/test/json/fixtures/fail25.json new file mode 100644 index 0000000000..2dfbd259ee --- /dev/null +++ b/test/json/fixtures/fail25.json @@ -0,0 +1 @@ +["tab character in string "] diff --git a/test/json/fixtures/fail27.json b/test/json/fixtures/fail27.json new file mode 100644 index 0000000000..6b01a2ca4a --- /dev/null +++ b/test/json/fixtures/fail27.json @@ -0,0 +1,2 @@ +["line +break"] \ No newline at end of file diff --git a/test/json/fixtures/fail28.json b/test/json/fixtures/fail28.json new file mode 100644 index 0000000000..621a0101c6 --- /dev/null +++ b/test/json/fixtures/fail28.json @@ -0,0 +1,2 @@ +["line\ +break"] \ No newline at end of file diff --git a/test/json/fixtures/fail3.json b/test/json/fixtures/fail3.json new file mode 100644 index 0000000000..168c81eb78 --- /dev/null +++ b/test/json/fixtures/fail3.json @@ -0,0 +1 @@ +{unquoted_key: "keys must be quoted"} \ No newline at end of file diff --git a/test/json/fixtures/fail4.json b/test/json/fixtures/fail4.json new file mode 100644 index 0000000000..9de168bf34 --- /dev/null +++ b/test/json/fixtures/fail4.json @@ -0,0 +1 @@ +["extra comma",] \ No newline at end of file diff --git a/test/json/fixtures/fail5.json b/test/json/fixtures/fail5.json new file mode 100644 index 0000000000..ddf3ce3d24 --- /dev/null +++ b/test/json/fixtures/fail5.json @@ -0,0 +1 @@ +["double extra comma",,] \ No newline at end of file diff --git a/test/json/fixtures/fail6.json b/test/json/fixtures/fail6.json new file mode 100644 index 0000000000..ed91580e1b --- /dev/null +++ b/test/json/fixtures/fail6.json @@ -0,0 +1 @@ +[ , "<-- missing value"] \ No newline at end of file diff --git a/test/json/fixtures/fail7.json b/test/json/fixtures/fail7.json new file mode 100644 index 0000000000..8a96af3e4e --- /dev/null +++ b/test/json/fixtures/fail7.json @@ -0,0 +1 @@ +["Comma after the close"], \ No newline at end of file diff --git a/test/json/fixtures/fail8.json b/test/json/fixtures/fail8.json new file mode 100644 index 0000000000..b28479c6ec --- /dev/null +++ b/test/json/fixtures/fail8.json @@ -0,0 +1 @@ +["Extra close"]] \ No newline at end of file diff --git a/test/json/fixtures/fail9.json b/test/json/fixtures/fail9.json new file mode 100644 index 0000000000..5815574f36 --- /dev/null +++ b/test/json/fixtures/fail9.json @@ -0,0 +1 @@ +{"Extra comma": true,} \ No newline at end of file diff --git a/test/json/fixtures/pass1.json b/test/json/fixtures/pass1.json new file mode 100644 index 0000000000..7828fcc137 --- /dev/null +++ b/test/json/fixtures/pass1.json @@ -0,0 +1,56 @@ +[ + "JSON Test Pattern pass1", + {"object with 1 member":["array with 1 element"]}, + {}, + [], + -42, + true, + false, + null, + { + "integer": 1234567890, + "real": -9876.543210, + "e": 0.123456789e-12, + "E": 1.234567890E+34, + "": 23456789012E666, + "zero": 0, + "one": 1, + "space": " ", + "quote": "\"", + "backslash": "\\", + "controls": "\b\f\n\r\t", + "slash": "/ & \/", + "alpha": "abcdefghijklmnopqrstuvwyz", + "ALPHA": "ABCDEFGHIJKLMNOPQRSTUVWYZ", + "digit": "0123456789", + "special": "`1~!@#$%^&*()_+-={':[,]}|;.?", + "hex": "\u0123\u4567\u89AB\uCDEF\uabcd\uef4A", + "true": true, + "false": false, + "null": null, + "array":[ ], + "object":{ }, + "address": "50 St. James Street", + "url": "http://www.JSON.org/", + "comment": "// /* */": " ", + " s p a c e d " :[1,2 , 3 + +, + +4 , 5 , 6 ,7 ], + "compact": [1,2,3,4,5,6,7], + "jsontext": "{\"object with 1 member\":[\"array with 1 element\"]}", + "quotes": "" \u0022 %22 0x22 034 "", + "\/\\\"\uCAFE\uBABE\uAB98\uFCDE\ubcda\uef4A\b\f\n\r\t`1~!@#$%^&*()_+-=[]{}|;:',./<>?" +: "A key can be any string" + }, + 0.5 ,98.6 +, +99.44 +, + +1066 + + +,"rosebud"] \ No newline at end of file diff --git a/test/json/fixtures/pass15.json b/test/json/fixtures/pass15.json new file mode 100644 index 0000000000..fc8376b605 --- /dev/null +++ b/test/json/fixtures/pass15.json @@ -0,0 +1 @@ +["Illegal backslash escape: \x15"] \ No newline at end of file diff --git a/test/json/fixtures/pass16.json b/test/json/fixtures/pass16.json new file mode 100644 index 0000000000..c43ae3c286 --- /dev/null +++ b/test/json/fixtures/pass16.json @@ -0,0 +1 @@ +["Illegal backslash escape: \'"] \ No newline at end of file diff --git a/test/json/fixtures/pass17.json b/test/json/fixtures/pass17.json new file mode 100644 index 0000000000..62b9214aed --- /dev/null +++ b/test/json/fixtures/pass17.json @@ -0,0 +1 @@ +["Illegal backslash escape: \017"] \ No newline at end of file diff --git a/test/json/fixtures/pass2.json b/test/json/fixtures/pass2.json new file mode 100644 index 0000000000..d3c63c7ad8 --- /dev/null +++ b/test/json/fixtures/pass2.json @@ -0,0 +1 @@ +[[[[[[[[[[[[[[[[[[["Not too deep"]]]]]]]]]]]]]]]]]]] \ No newline at end of file diff --git a/test/json/fixtures/pass26.json b/test/json/fixtures/pass26.json new file mode 100644 index 0000000000..845d26a6a5 --- /dev/null +++ b/test/json/fixtures/pass26.json @@ -0,0 +1 @@ +["tab\ character\ in\ string\ "] \ No newline at end of file diff --git a/test/json/fixtures/pass3.json b/test/json/fixtures/pass3.json new file mode 100644 index 0000000000..4528d51f1a --- /dev/null +++ b/test/json/fixtures/pass3.json @@ -0,0 +1,6 @@ +{ + "JSON Test Pattern pass3": { + "The outermost value": "must be an object or array.", + "In this test": "It is an object." + } +} diff --git a/test/json/runner.rb b/test/json/runner.rb new file mode 100644 index 0000000000..930fed7579 --- /dev/null +++ b/test/json/runner.rb @@ -0,0 +1,24 @@ +#!/usr/bin/env ruby + +require 'test/unit/ui/console/testrunner' +require 'test/unit/testsuite' +$:.unshift File.expand_path(File.dirname($0)) +$:.unshift 'tests' +require 'test_json' +require 'test_json_generate' +require 'test_json_unicode' +require 'test_json_addition' +require 'test_json_fixtures' + +class TS_AllTests + def self.suite + suite = Test::Unit::TestSuite.new name + suite << TC_JSONGenerate.suite + suite << TC_JSON.suite + suite << TC_JSONUnicode.suite + suite << TC_JSONAddition.suite + suite << TC_JSONFixtures.suite + end +end +Test::Unit::UI::Console::TestRunner.run(TS_AllTests) + # vim: set et sw=2 ts=2: diff --git a/test/json/test_json.rb b/test/json/test_json.rb new file mode 100755 index 0000000000..2be7b4a9a3 --- /dev/null +++ b/test/json/test_json.rb @@ -0,0 +1,255 @@ +#!/usr/bin/env ruby + +require 'test/unit' +require 'json' + +class TC_JSON < Test::Unit::TestCase + include JSON + + def setup + $KCODE = 'UTF8' + @ary = [1, "foo", 3.14, 4711.0, 2.718, nil, [1,-2,3], false, true].map do + |x| [x] + end + @ary_to_parse = ["1", '"foo"', "3.14", "4711.0", "2.718", "null", + "[1,-2,3]", "false", "true"].map do + |x| "[#{x}]" + end + @hash = { + 'a' => 2, + 'b' => 3.141, + 'c' => 'c', + 'd' => [ 1, "b", 3.14 ], + 'e' => { 'foo' => 'bar' }, + 'g' => "\"\0\037", + 'h' => 1000.0, + 'i' => 0.001 + } + @json = '{"a":2,"b":3.141,"c":"c","d":[1,"b",3.14],"e":{"foo":"bar"},' + + '"g":"\\"\\u0000\\u001f","h":1.0E3,"i":1.0E-3}' + end + suite << TC_JSON.suite + + def test_construction + parser = JSON::Parser.new('test') + assert_equal 'test', parser.source + end + + def assert_equal_float(expected, is) + assert_in_delta(expected.first, is.first, 1e-2) + end + + def test_parse_simple_arrays + assert_equal([], parse('[]')) + assert_equal([], parse(' [ ] ')) + assert_equal([nil], parse('[null]')) + assert_equal([false], parse('[false]')) + assert_equal([true], parse('[true]')) + assert_equal([-23], parse('[-23]')) + assert_equal([23], parse('[23]')) + assert_equal([0.23], parse('[0.23]')) + assert_equal([0.0], parse('[0e0]')) + assert_raises(JSON::ParserError) { parse('[+23.2]') } + assert_raises(JSON::ParserError) { parse('[+23]') } + assert_raises(JSON::ParserError) { parse('[.23]') } + assert_raises(JSON::ParserError) { parse('[023]') } + assert_equal_float [3.141], parse('[3.141]') + assert_equal_float [-3.141], parse('[-3.141]') + assert_equal_float [3.141], parse('[3141e-3]') + assert_equal_float [3.141], parse('[3141.1e-3]') + assert_equal_float [3.141], parse('[3141E-3]') + assert_equal_float [3.141], parse('[3141.0E-3]') + assert_equal_float [-3.141], parse('[-3141.0e-3]') + assert_equal_float [-3.141], parse('[-3141e-3]') + assert_equal([""], parse('[""]')) + assert_equal(["foobar"], parse('["foobar"]')) + assert_equal([{}], parse('[{}]')) + end + + def test_parse_simple_objects + assert_equal({}, parse('{}')) + assert_equal({}, parse(' { } ')) + assert_equal({ "a" => nil }, parse('{ "a" : null}')) + assert_equal({ "a" => nil }, parse('{"a":null}')) + assert_equal({ "a" => false }, parse('{ "a" : false } ')) + assert_equal({ "a" => false }, parse('{"a":false}')) + assert_raises(JSON::ParserError) { parse('{false}') } + assert_equal({ "a" => true }, parse('{"a":true}')) + assert_equal({ "a" => true }, parse(' { "a" : true } ')) + assert_equal({ "a" => -23 }, parse(' { "a" : -23 } ')) + assert_equal({ "a" => -23 }, parse(' { "a" : -23 } ')) + assert_equal({ "a" => 23 }, parse('{"a":23 } ')) + assert_equal({ "a" => 23 }, parse(' { "a" : 23 } ')) + assert_equal({ "a" => 0.23 }, parse(' { "a" : 0.23 } ')) + assert_equal({ "a" => 0.23 }, parse(' { "a" : 0.23 } ')) + end + + begin + require 'permutation' + def test_parse_more_complex_arrays + a = [ nil, false, true, "foßbar", [ "n€st€d", true ], { "nested" => true, "n€ßt€ð2" => {} }] + perms = Permutation.for a + perms.each do |perm| + orig_ary = perm.project + json = pretty_generate(orig_ary) + assert_equal orig_ary, parse(json) + end + end + + def test_parse_complex_objects + a = [ nil, false, true, "foßbar", [ "n€st€d", true ], { "nested" => true, "n€ßt€ð2" => {} }] + perms = Permutation.for a + perms.each do |perm| + s = "a" + orig_obj = perm.project.inject({}) { |h, x| h[s.dup] = x; s = s.succ; h } + json = pretty_generate(orig_obj) + assert_equal orig_obj, parse(json) + end + end + rescue LoadError + warn "Skipping permutation tests." + end + + def test_parse_arrays + assert_equal([1,2,3], parse('[1,2,3]')) + assert_equal([1.2,2,3], parse('[1.2,2,3]')) + assert_equal([[],[[],[]]], parse('[[],[[],[]]]')) + end + + def test_parse_values + assert_equal([""], parse('[""]')) + assert_equal(["\\"], parse('["\\\\"]')) + assert_equal(['"'], parse('["\""]')) + assert_equal(['\\"\\'], parse('["\\\\\\"\\\\"]')) + assert_equal(["\"\b\n\r\t\0\037"], + parse('["\"\b\n\r\t\u0000\u001f"]')) + for i in 0 ... @ary.size + assert_equal(@ary[i], parse(@ary_to_parse[i])) + end + end + + def test_parse_array + assert_equal([], parse('[]')) + assert_equal([], parse(' [ ] ')) + assert_equal([1], parse('[1]')) + assert_equal([1], parse(' [ 1 ] ')) + assert_equal(@ary, + parse('[[1],["foo"],[3.14],[47.11e+2],[2718.0E-3],[null],[[1,-2,3]]'\ + ',[false],[true]]')) + assert_equal(@ary, parse(%Q{ [ [1] , ["foo"] , [3.14] \t , [47.11e+2] + , [2718.0E-3 ],\r[ null] , [[1, -2, 3 ]], [false ],[ true]\n ] })) + end + + def test_parse_object + assert_equal({}, parse('{}')) + assert_equal({}, parse(' { } ')) + assert_equal({'foo'=>'bar'}, parse('{"foo":"bar"}')) + assert_equal({'foo'=>'bar'}, parse(' { "foo" : "bar" } ')) + end + + def test_parser_reset + parser = Parser.new(@json) + assert_equal(@hash, parser.parse) + assert_equal(@hash, parser.parse) + end + + def test_comments + json = < "value1", "key2" => "value2", "key3" => "value3" }, + parse(json)) + json = < "value1" }, parse(json)) + end + + def test_backslash + data = [ '\\.(?i:gif|jpe?g|png)$' ] + json = '["\\\\.(?i:gif|jpe?g|png)$"]' + assert_equal json, JSON.unparse(data) + assert_equal data, JSON.parse(json) + # + data = [ '\\"' ] + json = '["\\\\\""]' + assert_equal json, JSON.unparse(data) + assert_equal data, JSON.parse(json) + # + json = '["\/"]' + data = JSON.parse(json) + assert_equal ['/'], data + assert_equal json, JSON.unparse(data) + # + json = '["\""]' + data = JSON.parse(json) + assert_equal ['"'], data + assert_equal json, JSON.unparse(data) + json = '["\\\'"]' + data = JSON.parse(json) + assert_equal ["'"], data + assert_equal '["\'"]', JSON.unparse(data) + end + + def test_wrong_inputs + assert_raises(ParserError) { JSON.parse('"foo"') } + assert_raises(ParserError) { JSON.parse('123') } + assert_raises(ParserError) { JSON.parse('[] bla') } + assert_raises(ParserError) { JSON.parse('[] 1') } + assert_raises(ParserError) { JSON.parse('[] []') } + assert_raises(ParserError) { JSON.parse('[] {}') } + assert_raises(ParserError) { JSON.parse('{} []') } + assert_raises(ParserError) { JSON.parse('{} {}') } + assert_raises(ParserError) { JSON.parse('[NULL]') } + assert_raises(ParserError) { JSON.parse('[FALSE]') } + assert_raises(ParserError) { JSON.parse('[TRUE]') } + assert_raises(ParserError) { JSON.parse('[07] ') } + assert_raises(ParserError) { JSON.parse('[0a]') } + assert_raises(ParserError) { JSON.parse('[1.]') } + assert_raises(ParserError) { JSON.parse(' ') } + end + + def test_nesting + to_deep = '[[[[[[[[[[[[[[[[[[[["Too deep"]]]]]]]]]]]]]]]]]]]]' + assert_raises(JSON::NestingError) { JSON.parse to_deep } + assert_raises(JSON::NestingError) { JSON.parser.new(to_deep).parse } + assert_raises(JSON::NestingError) { JSON.parse to_deep, :max_nesting => 19 } + ok = JSON.parse to_deep, :max_nesting => 20 + assert_kind_of Array, ok + ok = JSON.parse to_deep, :max_nesting => nil + assert_kind_of Array, ok + ok = JSON.parse to_deep, :max_nesting => false + assert_kind_of Array, ok + ok = JSON.parse to_deep, :max_nesting => 0 + assert_kind_of Array, ok + end +end + # vim: set et sw=2 ts=2: diff --git a/test/json/test_json_addition.rb b/test/json/test_json_addition.rb new file mode 100755 index 0000000000..e527f70a11 --- /dev/null +++ b/test/json/test_json_addition.rb @@ -0,0 +1,94 @@ +#!/usr/bin/env ruby + +require 'test/unit' +require 'json' + +class TC_JSONAddition < Test::Unit::TestCase + include JSON + + class A + def initialize(a) + @a = a + end + + attr_reader :a + + def ==(other) + a == other.a + end + + def self.json_create(object) + new(*object['args']) + end + + def to_json(*args) + { + 'json_class' => self.class, + 'args' => [ @a ], + }.to_json(*args) + end + end + + class B + def to_json(*args) + { + 'json_class' => self.class, + }.to_json(*args) + end + end + + class C + def to_json(*args) + { + 'json_class' => 'TC_JSONAddition::Nix', + }.to_json(*args) + end + end + + def setup + $KCODE = 'UTF8' + end + + def test_extended_json + a = A.new(666) + assert A.json_creatable? + json = generate(a) + a_again = JSON.parse(json) + assert_kind_of a.class, a_again + assert_equal a, a_again + end + + def test_extended_json_fail + b = B.new + assert !B.json_creatable? + json = generate(b) + assert_equal({ 'json_class' => B.name }, JSON.parse(json)) + end + + def test_extended_json_fail + c = C.new + assert !C.json_creatable? + json = generate(c) + assert_raises(ArgumentError) { JSON.parse(json) } + end + + def test_raw_strings + raw = '' + raw_array = [] + for i in 0..255 + raw << i + raw_array << i + end + json = raw.to_json_raw + json_raw_object = raw.to_json_raw_object + hash = { 'json_class' => 'String', 'raw'=> raw_array } + assert_equal hash, json_raw_object + json_raw = < 2, + 'b' => 3.141, + 'c' => 'c', + 'd' => [ 1, "b", 3.14 ], + 'e' => { 'foo' => 'bar' }, + 'g' => "\"\0\037", + 'h' => 1000.0, + 'i' => 0.001 + } + @json2 = '{"a":2,"b":3.141,"c":"c","d":[1,"b",3.14],"e":{"foo":"bar"},' + + '"g":"\\"\\u0000\\u001f","h":1000.0,"i":0.001}' + @json3 = <<'EOT'.chomp +{ + "a": 2, + "b": 3.141, + "c": "c", + "d": [ + 1, + "b", + 3.14 + ], + "e": { + "foo": "bar" + }, + "g": "\"\u0000\u001f", + "h": 1000.0, + "i": 0.001 +} +EOT + end + + def test_unparse + json = unparse(@hash) + assert_equal(@json2, json) + parsed_json = parse(json) + assert_equal(@hash, parsed_json) + json = generate({1=>2}) + assert_equal('{"1":2}', json) + parsed_json = parse(json) + assert_equal({"1"=>2}, parsed_json) + end + + def test_unparse_pretty + json = pretty_unparse(@hash) + assert_equal(@json3, json) + parsed_json = parse(json) + assert_equal(@hash, parsed_json) + json = pretty_generate({1=>2}) + assert_equal(<<'EOT'.chomp, json) +{ + "1": 2 +} +EOT + parsed_json = parse(json) + assert_equal({"1"=>2}, parsed_json) + end + + def test_states + json = generate({1=>2}, nil) + assert_equal('{"1":2}', json) + s = JSON.state.new(:check_circular => true) + #assert s.check_circular + h = { 1=>2 } + h[3] = h + assert_raises(JSON::CircularDatastructure) { generate(h, s) } + s = JSON.state.new(:check_circular => true) + #assert s.check_circular + a = [ 1, 2 ] + a << a + assert_raises(JSON::CircularDatastructure) { generate(a, s) } + end +end diff --git a/test/json/test_json_unicode.rb b/test/json/test_json_unicode.rb new file mode 100755 index 0000000000..a23e50edf9 --- /dev/null +++ b/test/json/test_json_unicode.rb @@ -0,0 +1,59 @@ +#!/usr/bin/env ruby + +require 'test/unit' +require 'json' + +class TC_JSONUnicode < Test::Unit::TestCase + include JSON + + def setup + $KCODE = 'UTF8' + end + + def test_unicode + assert_equal '""', ''.to_json + assert_equal '"\\b"', "\b".to_json + assert_equal '"\u0001"', 0x1.chr.to_json + assert_equal '"\u001f"', 0x1f.chr.to_json + assert_equal '" "', ' '.to_json + assert_equal "\"#{0x7f.chr}\"", 0x7f.chr.to_json + utf8 = [ "© ≠ €! \01" ] + json = '["\u00a9 \u2260 \u20ac! \u0001"]' + assert_equal json, utf8.to_json + assert_equal utf8, parse(json) + utf8 = ["\343\201\202\343\201\204\343\201\206\343\201\210\343\201\212"] + json = "[\"\\u3042\\u3044\\u3046\\u3048\\u304a\"]" + assert_equal json, utf8.to_json + assert_equal utf8, parse(json) + utf8 = ['საქართველო'] + json = "[\"\\u10e1\\u10d0\\u10e5\\u10d0\\u10e0\\u10d7\\u10d5\\u10d4\\u10da\\u10dd\"]" + assert_equal json, utf8.to_json + assert_equal utf8, parse(json) + assert_equal '["\\u00c3"]', JSON.generate(["Ã"]) + assert_equal ["€"], JSON.parse('["\u20ac"]') + utf8 = ["\xf0\xa0\x80\x81"] + json = '["\ud840\udc01"]' + assert_equal json, JSON.generate(utf8) + assert_equal utf8, JSON.parse(json) + end + + def test_chars + (0..0x7f).each do |i| + c = ('%c' % i)[0] # c is a character object + json = '["\u%04x"]' % i + assert_equal c, JSON.parse(json).first + if c == ?\b + generated = JSON.generate(["" << i]) + assert '["\b"]' == generated || '["\10"]' == generated + elsif [?\n, ?\r, ?\t, ?\f].include?(c) + assert_equal '[' << ('' << i).dump << ']', JSON.generate(["" << i]) + elsif i < 0x20 + assert_equal json, JSON.generate(["" << i]) + end + end + assert_raises(JSON::GeneratorError) do + JSON.generate(["" << 0x80]) + end + assert_equal "\302\200", JSON.parse('["\u0080"]').first + end +end -- cgit v1.2.3