kotovalexarian-likes-github/ruby--ruby
1
0
Fork 0
mirror of https://github.com/ruby/ruby.git synced 2022-11-09 12:17:21 -05:00
Code Releases Activity

merge revision(s) 50339,50340,50342,50343: [Backport #10705]

Browse source 
parser.rl: use StringValue

	* ext/json/parser/parser.rl (cParser_initialize): use StringValue
  instead of direct rb_convert_type and remove duplicate
  conversion.
	* ext/json/parser/parser.rl: raise with messages in UTF-8
	  encoding.  [ruby-core:67386] [Bug #10705]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_2@51571 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
nagachika 2015-08-14 06:20:41 +00:00
parent 9c03800575
commit 4fe4f46592
6 changed files with 186 additions and 137 deletions
Download patch file Download diff file Expand all files Collapse all files

5
ChangeLog
View file

@ -1,3 +1,8 @@
Fri Aug 14 15:09:34 2015 Nobuyoshi Nakada <nobu@ruby-lang.org>
* ext/json/parser/parser.rl: raise with messages in UTF-8
encoding. [ruby-core:67386] [Bug #10705]
Fri Aug 14 00:44:34 2015 Eric Wong <e@80x24.org> Fri Aug 14 00:44:34 2015 Eric Wong <e@80x24.org>
* io.c (rb_io_oflags_modestr): handle O_TRUNC correctly * io.c (rb_io_oflags_modestr): handle O_TRUNC correctly

2
ext/json/parser/extconf.rb
View file

@ -1,3 +1,5 @@
require 'mkmf' require 'mkmf'
have_func("rb_enc_raise", "ruby.h")
create_makefile 'json/ext/parser' create_makefile 'json/ext/parser'

234
ext/json/parser/parser.c
View file

@ -3,6 +3,28 @@
#include "../fbuffer/fbuffer.h" #include "../fbuffer/fbuffer.h"
#include "parser.h" #include "parser.h"
#if defined HAVE_RUBY_ENCODING_H
# define EXC_ENCODING UTF_8,
# ifndef HAVE_RB_ENC_RAISE
static void
enc_raise(rb_encoding *enc, VALUE exc, const char *fmt, ...)
{
va_list args;
VALUE mesg;
va_start(args, fmt);
mesg = rb_enc_vsprintf(enc, fmt, args);
va_end(args);
rb_exc_raise(rb_exc_new3(exc, mesg));
}
# define rb_enc_raise enc_raise
# endif
#else
# define EXC_ENCODING /* nothing */
# define rb_enc_raise rb_raise
#endif
/* unicode */ /* unicode */
static const char digit_values[256] = { static const char digit_values[256] = {
@ -68,9 +90,7 @@ static int convert_UTF32_to_UTF8(char *buf, UTF32 ch)
} }
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
static VALUE CEncoding_ASCII_8BIT, CEncoding_UTF_8, CEncoding_UTF_16BE, static rb_encoding *UTF_8, *UTF_16BE, *UTF_16LE, *UTF_32BE, *UTF_32LE;
CEncoding_UTF_16LE, CEncoding_UTF_32BE, CEncoding_UTF_32LE;
static ID i_encoding, i_encode;
#else #else
static ID i_iconv; static ID i_iconv;
#endif #endif
@ -84,11 +104,11 @@ static ID i_json_creatable_p, i_json_create, i_create_id, i_create_additions,
i_match_string, i_aset, i_aref, i_leftshift; i_match_string, i_aset, i_aref, i_leftshift;
#line 110 "parser.rl" #line 130 "parser.rl"
#line 92 "parser.c" #line 112 "parser.c"
enum {JSON_object_start = 1}; enum {JSON_object_start = 1};
enum {JSON_object_first_final = 27}; enum {JSON_object_first_final = 27};
enum {JSON_object_error = 0}; enum {JSON_object_error = 0};
@ -96,7 +116,7 @@ enum {JSON_object_error = 0};
enum {JSON_object_en_main = 1}; enum {JSON_object_en_main = 1};
#line 151 "parser.rl" #line 171 "parser.rl"
static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result) static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *result)
@ -112,14 +132,14 @@ static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *resu
*result = NIL_P(object_class) ? rb_hash_new() : rb_class_new_instance(0, 0, object_class); *result = NIL_P(object_class) ? rb_hash_new() : rb_class_new_instance(0, 0, object_class);
#line 116 "parser.c" #line 136 "parser.c"
{ {
cs = JSON_object_start; cs = JSON_object_start;
} }
#line 166 "parser.rl" #line 186 "parser.rl"
#line 123 "parser.c" #line 143 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -147,7 +167,7 @@ case 2:
goto st2; goto st2;
goto st0; goto st0;
tr2: tr2:
#line 133 "parser.rl" #line 153 "parser.rl"
{ {
char *np; char *np;
json->parsing_name = 1; json->parsing_name = 1;
@ -160,7 +180,7 @@ st3:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof3; goto _test_eof3;
case 3: case 3:
#line 164 "parser.c" #line 184 "parser.c"
switch( (*p) ) { switch( (*p) ) {
case 13: goto st3; case 13: goto st3;
case 32: goto st3; case 32: goto st3;
@ -227,7 +247,7 @@ case 8:
goto st8; goto st8;
goto st0; goto st0;
tr11: tr11:
#line 118 "parser.rl" #line 138 "parser.rl"
{ {
VALUE v = Qnil; VALUE v = Qnil;
char *np = JSON_parse_value(json, p, pe, &v); char *np = JSON_parse_value(json, p, pe, &v);
@ -247,7 +267,7 @@ st9:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof9; goto _test_eof9;
case 9: case 9:
#line 251 "parser.c" #line 271 "parser.c"
switch( (*p) ) { switch( (*p) ) {
case 13: goto st9; case 13: goto st9;
case 32: goto st9; case 32: goto st9;
@ -336,14 +356,14 @@ case 18:
goto st9; goto st9;
goto st18; goto st18;
tr4: tr4:
#line 141 "parser.rl" #line 161 "parser.rl"
{ p--; {p++; cs = 27; goto _out;} } { p--; {p++; cs = 27; goto _out;} }
goto st27; goto st27;
st27: st27:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof27; goto _test_eof27;
case 27: case 27:
#line 347 "parser.c" #line 367 "parser.c"
goto st0; goto st0;
st19: st19:
if ( ++p == pe ) if ( ++p == pe )
@ -441,7 +461,7 @@ case 26:
_out: {} _out: {}
} }
#line 167 "parser.rl" #line 187 "parser.rl"
if (cs >= JSON_object_first_final) { if (cs >= JSON_object_first_final) {
if (json->create_additions) { if (json->create_additions) {
@ -466,7 +486,7 @@ case 26:
#line 470 "parser.c" #line 490 "parser.c"
enum {JSON_value_start = 1}; enum {JSON_value_start = 1};
enum {JSON_value_first_final = 21}; enum {JSON_value_first_final = 21};
enum {JSON_value_error = 0}; enum {JSON_value_error = 0};
@ -474,7 +494,7 @@ enum {JSON_value_error = 0};
enum {JSON_value_en_main = 1}; enum {JSON_value_en_main = 1};
#line 271 "parser.rl" #line 291 "parser.rl"
static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result) static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *result)
@ -482,14 +502,14 @@ static char *JSON_parse_value(JSON_Parser *json, char *p, char *pe, VALUE *resul
int cs = EVIL; int cs = EVIL;
#line 486 "parser.c" #line 506 "parser.c"
{ {
cs = JSON_value_start; cs = JSON_value_start;
} }
#line 278 "parser.rl" #line 298 "parser.rl"
#line 493 "parser.c" #line 513 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -514,14 +534,14 @@ st0:
cs = 0; cs = 0;
goto _out; goto _out;
tr0: tr0:
#line 219 "parser.rl" #line 239 "parser.rl"
{ {
char *np = JSON_parse_string(json, p, pe, result); char *np = JSON_parse_string(json, p, pe, result);
if (np == NULL) { p--; {p++; cs = 21; goto _out;} } else {p = (( np))-1;} if (np == NULL) { p--; {p++; cs = 21; goto _out;} } else {p = (( np))-1;}
} }
goto st21; goto st21;
tr2: tr2:
#line 224 "parser.rl" #line 244 "parser.rl"
{ {
char *np; char *np;
if(pe > p + 9 - json->quirks_mode && !strncmp(MinusInfinity, p, 9)) { if(pe > p + 9 - json->quirks_mode && !strncmp(MinusInfinity, p, 9)) {
@ -530,7 +550,7 @@ tr2:
{p = (( p + 10))-1;} {p = (( p + 10))-1;}
p--; {p++; cs = 21; goto _out;} p--; {p++; cs = 21; goto _out;}
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
} }
} }
np = JSON_parse_float(json, p, pe, result); np = JSON_parse_float(json, p, pe, result);
@ -541,7 +561,7 @@ tr2:
} }
goto st21; goto st21;
tr5: tr5:
#line 242 "parser.rl" #line 262 "parser.rl"
{ {
char *np; char *np;
json->current_nesting++; json->current_nesting++;
@ -551,7 +571,7 @@ tr5:
} }
goto st21; goto st21;
tr9: tr9:
#line 250 "parser.rl" #line 270 "parser.rl"
{ {
char *np; char *np;
json->current_nesting++; json->current_nesting++;
@ -561,39 +581,39 @@ tr9:
} }
goto st21; goto st21;
tr16: tr16:
#line 212 "parser.rl" #line 232 "parser.rl"
{ {
if (json->allow_nan) { if (json->allow_nan) {
*result = CInfinity; *result = CInfinity;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p - 8); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p - 8);
} }
} }
goto st21; goto st21;
tr18: tr18:
#line 205 "parser.rl" #line 225 "parser.rl"
{ {
if (json->allow_nan) { if (json->allow_nan) {
*result = CNaN; *result = CNaN;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p - 2); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p - 2);
} }
} }
goto st21; goto st21;
tr22: tr22:
#line 199 "parser.rl" #line 219 "parser.rl"
{ {
*result = Qfalse; *result = Qfalse;
} }
goto st21; goto st21;
tr25: tr25:
#line 196 "parser.rl" #line 216 "parser.rl"
{ {
*result = Qnil; *result = Qnil;
} }
goto st21; goto st21;
tr28: tr28:
#line 202 "parser.rl" #line 222 "parser.rl"
{ {
*result = Qtrue; *result = Qtrue;
} }
@ -602,9 +622,9 @@ st21:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof21; goto _test_eof21;
case 21: case 21:
#line 258 "parser.rl" #line 278 "parser.rl"
{ p--; {p++; cs = 21; goto _out;} } { p--; {p++; cs = 21; goto _out;} }
#line 608 "parser.c" #line 628 "parser.c"
goto st0; goto st0;
st2: st2:
if ( ++p == pe ) if ( ++p == pe )
@ -765,7 +785,7 @@ case 20:
_out: {} _out: {}
} }
#line 279 "parser.rl" #line 299 "parser.rl"
if (cs >= JSON_value_first_final) { if (cs >= JSON_value_first_final) {
return p; return p;
@ -775,7 +795,7 @@ case 20:
} }
#line 779 "parser.c" #line 799 "parser.c"
enum {JSON_integer_start = 1}; enum {JSON_integer_start = 1};
enum {JSON_integer_first_final = 3}; enum {JSON_integer_first_final = 3};
enum {JSON_integer_error = 0}; enum {JSON_integer_error = 0};
@ -783,7 +803,7 @@ enum {JSON_integer_error = 0};
enum {JSON_integer_en_main = 1}; enum {JSON_integer_en_main = 1};
#line 295 "parser.rl" #line 315 "parser.rl"
static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result) static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *result)
@ -791,15 +811,15 @@ static char *JSON_parse_integer(JSON_Parser *json, char *p, char *pe, VALUE *res
int cs = EVIL; int cs = EVIL;
#line 795 "parser.c" #line 815 "parser.c"
{ {
cs = JSON_integer_start; cs = JSON_integer_start;
} }
#line 302 "parser.rl" #line 322 "parser.rl"
json->memo = p; json->memo = p;
#line 803 "parser.c" #line 823 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -833,14 +853,14 @@ case 3:
goto st0; goto st0;
goto tr4; goto tr4;
tr4: tr4:
#line 292 "parser.rl" #line 312 "parser.rl"
{ p--; {p++; cs = 4; goto _out;} } { p--; {p++; cs = 4; goto _out;} }
goto st4; goto st4;
st4: st4:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof4; goto _test_eof4;
case 4: case 4:
#line 844 "parser.c" #line 864 "parser.c"
goto st0; goto st0;
st5: st5:
if ( ++p == pe ) if ( ++p == pe )
@ -859,7 +879,7 @@ case 5:
_out: {} _out: {}
} }
#line 304 "parser.rl" #line 324 "parser.rl"
if (cs >= JSON_integer_first_final) { if (cs >= JSON_integer_first_final) {
long len = p - json->memo; long len = p - json->memo;
@ -874,7 +894,7 @@ case 5:
} }
#line 878 "parser.c" #line 898 "parser.c"
enum {JSON_float_start = 1}; enum {JSON_float_start = 1};
enum {JSON_float_first_final = 8}; enum {JSON_float_first_final = 8};
enum {JSON_float_error = 0}; enum {JSON_float_error = 0};
@ -882,7 +902,7 @@ enum {JSON_float_error = 0};
enum {JSON_float_en_main = 1}; enum {JSON_float_en_main = 1};
#line 329 "parser.rl" #line 349 "parser.rl"
static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result) static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *result)
@ -890,15 +910,15 @@ static char *JSON_parse_float(JSON_Parser *json, char *p, char *pe, VALUE *resul
int cs = EVIL; int cs = EVIL;
#line 894 "parser.c" #line 914 "parser.c"
{ {
cs = JSON_float_start; cs = JSON_float_start;
} }
#line 336 "parser.rl" #line 356 "parser.rl"
json->memo = p; json->memo = p;
#line 902 "parser.c" #line 922 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -956,14 +976,14 @@ case 8:
goto st0; goto st0;
goto tr9; goto tr9;
tr9: tr9:
#line 323 "parser.rl" #line 343 "parser.rl"
{ p--; {p++; cs = 9; goto _out;} } { p--; {p++; cs = 9; goto _out;} }
goto st9; goto st9;
st9: st9:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof9; goto _test_eof9;
case 9: case 9:
#line 967 "parser.c" #line 987 "parser.c"
goto st0; goto st0;
st5: st5:
if ( ++p == pe ) if ( ++p == pe )
@ -1024,7 +1044,7 @@ case 7:
_out: {} _out: {}
} }
#line 338 "parser.rl" #line 358 "parser.rl"
if (cs >= JSON_float_first_final) { if (cs >= JSON_float_first_final) {
long len = p - json->memo; long len = p - json->memo;
@ -1040,7 +1060,7 @@ case 7:
#line 1044 "parser.c" #line 1064 "parser.c"
enum {JSON_array_start = 1}; enum {JSON_array_start = 1};
enum {JSON_array_first_final = 17}; enum {JSON_array_first_final = 17};
enum {JSON_array_error = 0}; enum {JSON_array_error = 0};
@ -1048,7 +1068,7 @@ enum {JSON_array_error = 0};
enum {JSON_array_en_main = 1}; enum {JSON_array_en_main = 1};
#line 381 "parser.rl" #line 401 "parser.rl"
static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result) static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *result)
@ -1062,14 +1082,14 @@ static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *resul
*result = NIL_P(array_class) ? rb_ary_new() : rb_class_new_instance(0, 0, array_class); *result = NIL_P(array_class) ? rb_ary_new() : rb_class_new_instance(0, 0, array_class);
#line 1066 "parser.c" #line 1086 "parser.c"
{ {
cs = JSON_array_start; cs = JSON_array_start;
} }
#line 394 "parser.rl" #line 414 "parser.rl"
#line 1073 "parser.c" #line 1093 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -1108,7 +1128,7 @@ case 2:
goto st2; goto st2;
goto st0; goto st0;
tr2: tr2:
#line 358 "parser.rl" #line 378 "parser.rl"
{ {
VALUE v = Qnil; VALUE v = Qnil;
char *np = JSON_parse_value(json, p, pe, &v); char *np = JSON_parse_value(json, p, pe, &v);
@ -1128,7 +1148,7 @@ st3:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof3; goto _test_eof3;
case 3: case 3:
#line 1132 "parser.c" #line 1152 "parser.c"
switch( (*p) ) { switch( (*p) ) {
case 13: goto st3; case 13: goto st3;
case 32: goto st3; case 32: goto st3;
@ -1228,14 +1248,14 @@ case 12:
goto st3; goto st3;
goto st12; goto st12;
tr4: tr4:
#line 373 "parser.rl" #line 393 "parser.rl"
{ p--; {p++; cs = 17; goto _out;} } { p--; {p++; cs = 17; goto _out;} }
goto st17; goto st17;
st17: st17:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof17; goto _test_eof17;
case 17: case 17:
#line 1239 "parser.c" #line 1259 "parser.c"
goto st0; goto st0;
st13: st13:
if ( ++p == pe ) if ( ++p == pe )
@ -1291,12 +1311,12 @@ case 16:
_out: {} _out: {}
} }
#line 395 "parser.rl" #line 415 "parser.rl"
if(cs >= JSON_array_first_final) { if(cs >= JSON_array_first_final) {
return p + 1; return p + 1;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return NULL; return NULL;
} }
} }
@ -1372,7 +1392,7 @@ static VALUE json_string_unescape(VALUE result, char *string, char *stringEnd)
} }
#line 1376 "parser.c" #line 1396 "parser.c"
enum {JSON_string_start = 1}; enum {JSON_string_start = 1};
enum {JSON_string_first_final = 8}; enum {JSON_string_first_final = 8};
enum {JSON_string_error = 0}; enum {JSON_string_error = 0};
@ -1380,7 +1400,7 @@ enum {JSON_string_error = 0};
enum {JSON_string_en_main = 1}; enum {JSON_string_en_main = 1};
#line 494 "parser.rl" #line 514 "parser.rl"
static int static int
@ -1402,15 +1422,15 @@ static char *JSON_parse_string(JSON_Parser *json, char *p, char *pe, VALUE *resu
*result = rb_str_buf_new(0); *result = rb_str_buf_new(0);
#line 1406 "parser.c" #line 1426 "parser.c"
{ {
cs = JSON_string_start; cs = JSON_string_start;
} }
#line 515 "parser.rl" #line 535 "parser.rl"
json->memo = p; json->memo = p;
#line 1414 "parser.c" #line 1434 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -1435,7 +1455,7 @@ case 2:
goto st0; goto st0;
goto st2; goto st2;
tr2: tr2:
#line 480 "parser.rl" #line 500 "parser.rl"
{ {
*result = json_string_unescape(*result, json->memo + 1, p); *result = json_string_unescape(*result, json->memo + 1, p);
if (NIL_P(*result)) { if (NIL_P(*result)) {
@ -1446,14 +1466,14 @@ tr2:
{p = (( p + 1))-1;} {p = (( p + 1))-1;}
} }
} }
#line 491 "parser.rl" #line 511 "parser.rl"
{ p--; {p++; cs = 8; goto _out;} } { p--; {p++; cs = 8; goto _out;} }
goto st8; goto st8;
st8: st8:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof8; goto _test_eof8;
case 8: case 8:
#line 1457 "parser.c" #line 1477 "parser.c"
goto st0; goto st0;
st3: st3:
if ( ++p == pe ) if ( ++p == pe )
@ -1529,7 +1549,7 @@ case 7:
_out: {} _out: {}
} }
#line 517 "parser.rl" #line 537 "parser.rl"
if (json->create_additions && RTEST(match_string = json->match_string)) { if (json->create_additions && RTEST(match_string = json->match_string)) {
VALUE klass; VALUE klass;
@ -1573,22 +1593,22 @@ static VALUE convert_encoding(VALUE source)
} }
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
{ {
VALUE encoding = rb_funcall(source, i_encoding, 0); rb_encoding *enc = rb_enc_get(source);
if (encoding == CEncoding_ASCII_8BIT) { if (enc == rb_ascii8bit_encoding()) {
if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) { if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_32BE); source = rb_str_conv_enc(source, UTF_32BE, rb_utf8_encoding());
} else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) { } else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_16BE); source = rb_str_conv_enc(source, UTF_16BE, rb_utf8_encoding());
} else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) { } else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_32LE); source = rb_str_conv_enc(source, UTF_32LE, rb_utf8_encoding());
} else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) { } else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_16LE); source = rb_str_conv_enc(source, UTF_16LE, rb_utf8_encoding());
} else { } else {
source = rb_str_dup(source); source = rb_str_dup(source);
FORCE_UTF8(source); FORCE_UTF8(source);
} }
} else { } else {
source = rb_funcall(source, i_encode, 1, CEncoding_UTF_8); source = rb_str_conv_enc(source, NULL, rb_utf8_encoding());
} }
} }
#else #else
@ -1716,12 +1736,11 @@ static VALUE cParser_initialize(int argc, VALUE *argv, VALUE self)
json->object_class = Qnil; json->object_class = Qnil;
json->array_class = Qnil; json->array_class = Qnil;
} }
source = rb_convert_type(source, T_STRING, "String", "to_str"); StringValue(source);
if (!json->quirks_mode) { if (!json->quirks_mode) {
source = convert_encoding(StringValue(source)); source = convert_encoding(source);
} }
json->current_nesting = 0; json->current_nesting = 0;
StringValue(source);
json->len = RSTRING_LEN(source); json->len = RSTRING_LEN(source);
json->source = RSTRING_PTR(source);; json->source = RSTRING_PTR(source);;
json->Vsource = source; json->Vsource = source;
@ -1729,7 +1748,7 @@ static VALUE cParser_initialize(int argc, VALUE *argv, VALUE self)
} }
#line 1733 "parser.c" #line 1752 "parser.c"
enum {JSON_start = 1}; enum {JSON_start = 1};
enum {JSON_first_final = 10}; enum {JSON_first_final = 10};
enum {JSON_error = 0}; enum {JSON_error = 0};
@ -1737,7 +1756,7 @@ enum {JSON_error = 0};
enum {JSON_en_main = 1}; enum {JSON_en_main = 1};
#line 740 "parser.rl" #line 759 "parser.rl"
static VALUE cParser_parse_strict(VALUE self) static VALUE cParser_parse_strict(VALUE self)
@ -1748,16 +1767,16 @@ static VALUE cParser_parse_strict(VALUE self)
GET_PARSER; GET_PARSER;
#line 1752 "parser.c" #line 1771 "parser.c"
{ {
cs = JSON_start; cs = JSON_start;
} }
#line 750 "parser.rl" #line 769 "parser.rl"
p = json->source; p = json->source;
pe = p + json->len; pe = p + json->len;
#line 1761 "parser.c" #line 1780 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -1813,7 +1832,7 @@ case 5:
goto st1; goto st1;
goto st5; goto st5;
tr3: tr3:
#line 729 "parser.rl" #line 748 "parser.rl"
{ {
char *np; char *np;
json->current_nesting = 1; json->current_nesting = 1;
@ -1822,7 +1841,7 @@ tr3:
} }
goto st10; goto st10;
tr4: tr4:
#line 722 "parser.rl" #line 741 "parser.rl"
{ {
char *np; char *np;
json->current_nesting = 1; json->current_nesting = 1;
@ -1834,7 +1853,7 @@ st10:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof10; goto _test_eof10;
case 10: case 10:
#line 1838 "parser.c" #line 1857 "parser.c"
switch( (*p) ) { switch( (*p) ) {
case 13: goto st10; case 13: goto st10;
case 32: goto st10; case 32: goto st10;
@ -1891,19 +1910,19 @@ case 9:
_out: {} _out: {}
} }
#line 753 "parser.rl" #line 772 "parser.rl"
if (cs >= JSON_first_final && p == pe) { if (cs >= JSON_first_final && p == pe) {
return result; return result;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return Qnil; return Qnil;
} }
} }
#line 1907 "parser.c" #line 1926 "parser.c"
enum {JSON_quirks_mode_start = 1}; enum {JSON_quirks_mode_start = 1};
enum {JSON_quirks_mode_first_final = 10}; enum {JSON_quirks_mode_first_final = 10};
enum {JSON_quirks_mode_error = 0}; enum {JSON_quirks_mode_error = 0};
@ -1911,7 +1930,7 @@ enum {JSON_quirks_mode_error = 0};
enum {JSON_quirks_mode_en_main = 1}; enum {JSON_quirks_mode_en_main = 1};
#line 778 "parser.rl" #line 797 "parser.rl"
static VALUE cParser_parse_quirks_mode(VALUE self) static VALUE cParser_parse_quirks_mode(VALUE self)
@ -1922,16 +1941,16 @@ static VALUE cParser_parse_quirks_mode(VALUE self)
GET_PARSER; GET_PARSER;
#line 1926 "parser.c" #line 1945 "parser.c"
{ {
cs = JSON_quirks_mode_start; cs = JSON_quirks_mode_start;
} }
#line 788 "parser.rl" #line 807 "parser.rl"
p = json->source; p = json->source;
pe = p + json->len; pe = p + json->len;
#line 1935 "parser.c" #line 1954 "parser.c"
{ {
if ( p == pe ) if ( p == pe )
goto _test_eof; goto _test_eof;
@ -1965,7 +1984,7 @@ st0:
cs = 0; cs = 0;
goto _out; goto _out;
tr2: tr2:
#line 770 "parser.rl" #line 789 "parser.rl"
{ {
char *np = JSON_parse_value(json, p, pe, &result); char *np = JSON_parse_value(json, p, pe, &result);
if (np == NULL) { p--; {p++; cs = 10; goto _out;} } else {p = (( np))-1;} if (np == NULL) { p--; {p++; cs = 10; goto _out;} } else {p = (( np))-1;}
@ -1975,7 +1994,7 @@ st10:
if ( ++p == pe ) if ( ++p == pe )
goto _test_eof10; goto _test_eof10;
case 10: case 10:
#line 1979 "parser.c" #line 1998 "parser.c"
switch( (*p) ) { switch( (*p) ) {
case 13: goto st10; case 13: goto st10;
case 32: goto st10; case 32: goto st10;
@ -2064,12 +2083,12 @@ case 9:
_out: {} _out: {}
} }
#line 791 "parser.rl" #line 810 "parser.rl"
if (cs >= JSON_quirks_mode_first_final && p == pe) { if (cs >= JSON_quirks_mode_first_final && p == pe) {
return result; return result;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return Qnil; return Qnil;
} }
} }
@ -2197,14 +2216,11 @@ void Init_parser(void)
i_aref = rb_intern("[]"); i_aref = rb_intern("[]");
i_leftshift = rb_intern("<<"); i_leftshift = rb_intern("<<");
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
CEncoding_UTF_8 = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-8")); UTF_8 = rb_utf8_encoding();
CEncoding_UTF_16BE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-16be")); UTF_16BE = rb_enc_find("utf-16be");
CEncoding_UTF_16LE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-16le")); UTF_16LE = rb_enc_find("utf-16le");
CEncoding_UTF_32BE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-32be")); UTF_32BE = rb_enc_find("utf-32be");
CEncoding_UTF_32LE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-32le")); UTF_32LE = rb_enc_find("utf-32le");
CEncoding_ASCII_8BIT = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("ascii-8bit"));
i_encoding = rb_intern("encoding");
i_encode = rb_intern("encode");
#else #else
i_iconv = rb_intern("iconv"); i_iconv = rb_intern("iconv");
#endif #endif

70
ext/json/parser/parser.rl
View file

@ -1,6 +1,28 @@
#include "../fbuffer/fbuffer.h" #include "../fbuffer/fbuffer.h"
#include "parser.h" #include "parser.h"
#if defined HAVE_RUBY_ENCODING_H
# define EXC_ENCODING UTF_8,
# ifndef HAVE_RB_ENC_RAISE
static void
enc_raise(rb_encoding *enc, VALUE exc, const char *fmt, ...)
{
va_list args;
VALUE mesg;
va_start(args, fmt);
mesg = rb_enc_vsprintf(enc, fmt, args);
va_end(args);
rb_exc_raise(rb_exc_new3(exc, mesg));
}
# define rb_enc_raise enc_raise
# endif
#else
# define EXC_ENCODING /* nothing */
# define rb_enc_raise rb_raise
#endif
/* unicode */ /* unicode */
static const char digit_values[256] = { static const char digit_values[256] = {
@ -66,9 +88,7 @@ static int convert_UTF32_to_UTF8(char *buf, UTF32 ch)
} }
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
static VALUE CEncoding_ASCII_8BIT, CEncoding_UTF_8, CEncoding_UTF_16BE, static rb_encoding *UTF_8, *UTF_16BE, *UTF_16LE, *UTF_32BE, *UTF_32LE;
CEncoding_UTF_16LE, CEncoding_UTF_32BE, CEncoding_UTF_32LE;
static ID i_encoding, i_encode;
#else #else
static ID i_iconv; static ID i_iconv;
#endif #endif
@ -206,14 +226,14 @@ static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *resu
if (json->allow_nan) { if (json->allow_nan) {
*result = CNaN; *result = CNaN;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p - 2); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p - 2);
} }
} }
action parse_infinity { action parse_infinity {
if (json->allow_nan) { if (json->allow_nan) {
*result = CInfinity; *result = CInfinity;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p - 8); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p - 8);
} }
} }
action parse_string { action parse_string {
@ -229,7 +249,7 @@ static char *JSON_parse_object(JSON_Parser *json, char *p, char *pe, VALUE *resu
fexec p + 10; fexec p + 10;
fhold; fbreak; fhold; fbreak;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
} }
} }
np = JSON_parse_float(json, fpc, pe, result); np = JSON_parse_float(json, fpc, pe, result);
@ -396,7 +416,7 @@ static char *JSON_parse_array(JSON_Parser *json, char *p, char *pe, VALUE *resul
if(cs >= JSON_array_first_final) { if(cs >= JSON_array_first_final) {
return p + 1; return p + 1;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return NULL; return NULL;
} }
} }
@ -557,22 +577,22 @@ static VALUE convert_encoding(VALUE source)
} }
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
{ {
VALUE encoding = rb_funcall(source, i_encoding, 0); rb_encoding *enc = rb_enc_get(source);
if (encoding == CEncoding_ASCII_8BIT) { if (enc == rb_ascii8bit_encoding()) {
if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) { if (len >= 4 && ptr[0] == 0 && ptr[1] == 0 && ptr[2] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_32BE); source = rb_str_conv_enc(source, UTF_32BE, rb_utf8_encoding());
} else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) { } else if (len >= 4 && ptr[0] == 0 && ptr[2] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_16BE); source = rb_str_conv_enc(source, UTF_16BE, rb_utf8_encoding());
} else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) { } else if (len >= 4 && ptr[1] == 0 && ptr[2] == 0 && ptr[3] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_32LE); source = rb_str_conv_enc(source, UTF_32LE, rb_utf8_encoding());
} else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) { } else if (len >= 4 && ptr[1] == 0 && ptr[3] == 0) {
source = rb_funcall(source, i_encode, 2, CEncoding_UTF_8, CEncoding_UTF_16LE); source = rb_str_conv_enc(source, UTF_16LE, rb_utf8_encoding());
} else { } else {
source = rb_str_dup(source); source = rb_str_dup(source);
FORCE_UTF8(source); FORCE_UTF8(source);
} }
} else { } else {
source = rb_funcall(source, i_encode, 1, CEncoding_UTF_8); source = rb_str_conv_enc(source, NULL, rb_utf8_encoding());
} }
} }
#else #else
@ -700,12 +720,11 @@ static VALUE cParser_initialize(int argc, VALUE *argv, VALUE self)
json->object_class = Qnil; json->object_class = Qnil;
json->array_class = Qnil; json->array_class = Qnil;
} }
source = rb_convert_type(source, T_STRING, "String", "to_str"); StringValue(source);
if (!json->quirks_mode) { if (!json->quirks_mode) {
source = convert_encoding(StringValue(source)); source = convert_encoding(source);
} }
json->current_nesting = 0; json->current_nesting = 0;
StringValue(source);
json->len = RSTRING_LEN(source); json->len = RSTRING_LEN(source);
json->source = RSTRING_PTR(source);; json->source = RSTRING_PTR(source);;
json->Vsource = source; json->Vsource = source;
@ -754,7 +773,7 @@ static VALUE cParser_parse_strict(VALUE self)
if (cs >= JSON_first_final && p == pe) { if (cs >= JSON_first_final && p == pe) {
return result; return result;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return Qnil; return Qnil;
} }
} }
@ -792,7 +811,7 @@ static VALUE cParser_parse_quirks_mode(VALUE self)
if (cs >= JSON_quirks_mode_first_final && p == pe) { if (cs >= JSON_quirks_mode_first_final && p == pe) {
return result; return result;
} else { } else {
rb_raise(eParserError, "%u: unexpected token at '%s'", __LINE__, p); rb_enc_raise(EXC_ENCODING eParserError, "%u: unexpected token at '%s'", __LINE__, p);
return Qnil; return Qnil;
} }
} }
@ -920,14 +939,11 @@ void Init_parser()
i_aref = rb_intern("[]"); i_aref = rb_intern("[]");
i_leftshift = rb_intern("<<"); i_leftshift = rb_intern("<<");
#ifdef HAVE_RUBY_ENCODING_H #ifdef HAVE_RUBY_ENCODING_H
CEncoding_UTF_8 = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-8")); UTF_8 = rb_utf8_encoding();
CEncoding_UTF_16BE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-16be")); UTF_16BE = rb_enc_find("utf-16be");
CEncoding_UTF_16LE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-16le")); UTF_16LE = rb_enc_find("utf-16le");
CEncoding_UTF_32BE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-32be")); UTF_32BE = rb_enc_find("utf-32be");
CEncoding_UTF_32LE = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("utf-32le")); UTF_32LE = rb_enc_find("utf-32le");
CEncoding_ASCII_8BIT = rb_funcall(rb_path2class("Encoding"), rb_intern("find"), 1, rb_str_new2("ascii-8bit"));
i_encoding = rb_intern("encoding");
i_encode = rb_intern("encode");
#else #else
i_iconv = rb_intern("iconv"); i_iconv = rb_intern("iconv");
#endif #endif

10
test/json/test_json.rb
View file

@ -544,4 +544,14 @@ EOT
JSON::Parser.new(source) JSON::Parser.new(source)
assert_equal Encoding::ASCII_8BIT, source.encoding assert_equal Encoding::ASCII_8BIT, source.encoding
end if defined?(Encoding::ASCII_8BIT) end if defined?(Encoding::ASCII_8BIT)
def test_error_message_encoding
bug10705 = '[ruby-core:67386] [Bug #10705]'
json = "\"\xE2\x88\x9A\"".force_encoding(Encoding::UTF_8)
e = assert_raise(JSON::ParserError) {
JSON.parse(json)
}
assert_equal(Encoding::UTF_8, e.message.encoding, bug10705)
assert_include(e.message, json, bug10705)
end if defined?(Encoding::UTF_8)
end end

2
version.h
View file

@ -1,6 +1,6 @@
#define RUBY_VERSION "2.2.3" #define RUBY_VERSION "2.2.3"
#define RUBY_RELEASE_DATE "2015-08-14" #define RUBY_RELEASE_DATE "2015-08-14"
#define RUBY_PATCHLEVEL 164 #define RUBY_PATCHLEVEL 165
#define RUBY_RELEASE_YEAR 2015 #define RUBY_RELEASE_YEAR 2015
#define RUBY_RELEASE_MONTH 8 #define RUBY_RELEASE_MONTH 8