Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit32032d4

Browse files
committed
Fix usage of char2wchar/wchar2char. Changes:
- pg_wchar and wchar_t could have different size, so char2wchar doesn't call pg_mb2wchar_with_len to prevent out-of-bound memory bug- make char2wchar/wchar2char symmetric, now they should not be called with C-locale because mbstowcs/wcstombs oftenly doesn't work correct with C-locale.- Text parser uses pg_mb2wchar_with_len directly in case of C-locale and multibyte encodingPer bug report by Hiroshi Inoue <inoue@tpf.co.jp> andfollowing discussion.Backpatch up to 8.2 when multybyte support was implemented in tsearch.
1 parent876b37d commit32032d4

File tree

2 files changed

+42
-34
lines changed

2 files changed

+42
-34
lines changed

‎src/backend/tsearch/wparser_def.c

Lines changed: 33 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -7,7 +7,7 @@
77
*
88
*
99
* IDENTIFICATION
10-
* $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.20 2009/01/15 16:33:59 teodor Exp $
10+
* $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.21 2009/03/02 15:10:09 teodor Exp $
1111
*
1212
*-------------------------------------------------------------------------
1313
*/
@@ -240,12 +240,12 @@ typedef struct TParser
240240
intlenstr;/* length of mbstring */
241241
#ifdefUSE_WIDE_UPPER_LOWER
242242
wchar_t*wstr;/* wide character string */
243-
intlenwstr;/* length of wsting */
243+
pg_wchar*pgwstr;/* wide character string for C-locale */
244+
boolusewide;
244245
#endif
245246

246247
/* State of parse */
247248
intcharmaxlen;
248-
boolusewide;
249249
TParserPosition*state;
250250
boolignore;
251251
boolwanthost;
@@ -299,13 +299,24 @@ TParserInit(char *str, int len)
299299
if (prs->charmaxlen>1)
300300
{
301301
prs->usewide= true;
302-
prs->wstr= (wchar_t*)palloc(sizeof(wchar_t)* (prs->lenstr+1));
303-
prs->lenwstr=char2wchar(prs->wstr,prs->lenstr+1,
304-
prs->str,prs->lenstr);
302+
if (lc_ctype_is_c() )
303+
{
304+
/*
305+
* char2wchar doesn't work for C-locale and
306+
* sizeof(pg_wchar) could be not equal to sizeof(wchar_t)
307+
*/
308+
prs->pgwstr= (pg_wchar*)palloc(sizeof(pg_wchar)* (prs->lenstr+1));
309+
pg_mb2wchar_with_len(prs->str,prs->pgwstr,prs->lenstr);
310+
}
311+
else
312+
{
313+
prs->wstr= (wchar_t*)palloc(sizeof(wchar_t)* (prs->lenstr+1));
314+
char2wchar(prs->wstr,prs->lenstr+1,prs->str,prs->lenstr);
315+
}
305316
}
306317
else
307-
#endif
308318
prs->usewide= false;
319+
#endif
309320

310321
prs->state=newTParserPosition(NULL);
311322
prs->state->state=TPS_Base;
@@ -331,17 +342,21 @@ TParserClose(TParser *prs)
331342
#ifdefUSE_WIDE_UPPER_LOWER
332343
if (prs->wstr)
333344
pfree(prs->wstr);
345+
if (prs->pgwstr)
346+
pfree(prs->pgwstr);
334347
#endif
335348

336349
pfree(prs);
337350
}
338351

339352
/*
340353
* Character-type support functions, equivalent to is* macros, but
341-
* working with any possible encodings and locales. Note,
342-
* that with multibyte encoding and C-locale isw* function may fail
343-
* or give wrong result. Note 2: multibyte encoding and C-locale
344-
* often are used for Asian languages
354+
* working with any possible encodings and locales. Notes:
355+
* - with multibyte encoding and C-locale isw* function may fail
356+
* or give wrong result.
357+
* - multibyte encoding and C-locale often are used for
358+
* Asian languages.
359+
* - if locale is C the we use pgwstr instead of wstr
345360
*/
346361

347362
#ifdefUSE_WIDE_UPPER_LOWER
@@ -352,14 +367,14 @@ p_is##type(TParser *prs) {\
352367
Assert( prs->state );\
353368
if ( prs->usewide )\
354369
{\
355-
if (lc_ctype_is_c() )\
356-
return is##type( 0xff & *( prs->wstr + prs->state->poschar) );\
370+
if (prs->pgwstr )\
371+
return is##type( 0xff & *( prs->pgwstr + prs->state->poschar) );\
357372
\
358373
return isw##type( *(wint_t*)( prs->wstr + prs->state->poschar ) );\
359374
}\
360375
\
361376
return is##type( *(unsigned char*)( prs->str + prs->state->posbyte ) ); \
362-
}\
377+
}\
363378
\
364379
static int\
365380
p_isnot##type(TParser *prs) {\
@@ -373,9 +388,9 @@ p_isalnum(TParser *prs)
373388

374389
if (prs->usewide)
375390
{
376-
if (lc_ctype_is_c())
391+
if (prs->pgwstr)
377392
{
378-
unsignedintc=*(prs->wstr+prs->state->poschar);
393+
unsignedintc=*(prs->pgwstr+prs->state->poschar);
379394

380395
/*
381396
* any non-ascii symbol with multibyte encoding with C-locale is
@@ -405,9 +420,9 @@ p_isalpha(TParser *prs)
405420

406421
if (prs->usewide)
407422
{
408-
if (lc_ctype_is_c())
423+
if (prs->pgwstr)
409424
{
410-
unsignedintc=*(prs->wstr+prs->state->poschar);
425+
unsignedintc=*(prs->pgwstr+prs->state->poschar);
411426

412427
/*
413428
* any non-ascii symbol with multibyte encoding with C-locale is

‎src/backend/utils/mb/mbutils.c

Lines changed: 9 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@
44
* (currently mule internal code (mic) is used)
55
* Tatsuo Ishii
66
*
7-
* $PostgreSQL: pgsql/src/backend/utils/mb/mbutils.c,v 1.78 2009/01/2210:09:48 mha Exp $
7+
* $PostgreSQL: pgsql/src/backend/utils/mb/mbutils.c,v 1.79 2009/03/02 15:10:09 teodor Exp $
88
*/
99
#include"postgres.h"
1010

@@ -601,7 +601,10 @@ wchar2char(char *to, const wchar_t *from, size_t tolen)
601601
}
602602
else
603603
#endif/* WIN32 */
604+
{
605+
Assert( !lc_ctype_is_c() );
604606
result=wcstombs(to,from,tolen);
607+
}
605608
returnresult;
606609
}
607610

@@ -647,22 +650,12 @@ char2wchar(wchar_t *to, size_t tolen, const char *from, size_t fromlen)
647650
else
648651
#endif/* WIN32 */
649652
{
650-
if (lc_ctype_is_c())
651-
{
652-
/*
653-
* pg_mb2wchar_with_len always adds trailing '\0', so 'to' should be
654-
* allocated with sufficient space
655-
*/
656-
result=pg_mb2wchar_with_len(from, (pg_wchar*)to,fromlen);
657-
}
658-
else
659-
{
660-
/* mbstowcs requires ending '\0' */
661-
char*str=pnstrdup(from,fromlen);
653+
/* mbstowcs requires ending '\0' */
654+
char*str=pnstrdup(from,fromlen);
662655

663-
result=mbstowcs(to,str,tolen);
664-
pfree(str);
665-
}
656+
Assert( !lc_ctype_is_c());
657+
result=mbstowcs(to,str,tolen);
658+
pfree(str);
666659
}
667660

668661
if (result==-1)

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp