Skip to content

Commit 54cd4f0

Browse files
tglsfdctglsfdc
committed
Work around a subtle portability problem in use of printf %s format.
Depending on which spec you read, field widths and precisions in %s may be counted either in bytes or characters. Our code was assuming bytes, which is wrong at least for glibc's implementation, and in any case libc might have a different idea of the prevailing encoding than we do. Hence, for portable results we must avoid using anything more complex than just "%s" unless the string to be printed is known to be all-ASCII. This patch fixes the cases I could find, including the psql formatting failure reported by Hernan Gonzalez. In HEAD only, I also added comments to some places where it appears safe to continue using "%.*s".
1 parent 71a185a commit 54cd4f0

File tree

11 files changed

+111
-31
lines changed

11 files changed

+111
-31
lines changed

‎src/backend/lib/stringinfo.c

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@
99
* Portions Copyright (c) 1996-2010, PostgreSQL Global Development Group
1010
* Portions Copyright (c) 1994, Regents of the University of California
1111
*
12-
* $PostgreSQL: pgsql/src/backend/lib/stringinfo.c,v 1.52 2010/01/02 16:57:45 momjian Exp $
12+
* $PostgreSQL: pgsql/src/backend/lib/stringinfo.c,v 1.53 2010/05/08 16:39:49 tgl Exp $
1313
*
1414
*-------------------------------------------------------------------------
1515
*/
@@ -226,7 +226,8 @@ appendBinaryStringInfo(StringInfo str, const char *data, int datalen)
226226

227227
/*
228228
* Keep a trailing null in place, even though it's probably useless for
229-
* binary data...
229+
* binary data. (Some callers are dealing with text but call this
230+
* because their input isn't null-terminated.)
230231
*/
231232
str->data[str->len] = '\0';
232233
}

‎src/backend/parser/scansup.c

Lines changed: 13 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,7 @@
99
*
1010
*
1111
* IDENTIFICATION
12-
* $PostgreSQL: pgsql/src/backend/parser/scansup.c,v 1.39 2010/01/02 16:57:50 momjian Exp $
12+
* $PostgreSQL: pgsql/src/backend/parser/scansup.c,v 1.40 2010/05/08 16:39:49 tgl Exp $
1313
*
1414
*-------------------------------------------------------------------------
1515
*/
@@ -176,10 +176,20 @@ truncate_identifier(char *ident, int len, bool warn)
176176
{
177177
len = pg_mbcliplen(ident, len, NAMEDATALEN - 1);
178178
if (warn)
179+
{
180+
/*
181+
* Cannot use %.*s here because some machines interpret %s's
182+
* precision in characters, others in bytes.
183+
*/
184+
char buf[NAMEDATALEN];
185+
186+
memcpy(buf, ident, len);
187+
buf[len] = '\0';
179188
ereport(NOTICE,
180189
(errcode(ERRCODE_NAME_TOO_LONG),
181-
errmsg("identifier \"%s\" will be truncated to \"%.*s\"",
182-
ident, len, ident)));
190+
errmsg("identifier \"%s\" will be truncated to \"%s\"",
191+
ident, buf)));
192+
}
183193
ident[len] = '\0';
184194
}
185195
}

‎src/backend/tsearch/wparser_def.c

Lines changed: 8 additions & 1 deletion
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 1.30 2010/04/28 02:04:16 tgl Exp $
10+
* $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.31 2010/05/08 16:39:49 tgl Exp $
1111
*
1212
*-------------------------------------------------------------------------
1313
*/
@@ -322,6 +322,12 @@ TParserInit(char *str, int len)
322322
prs->state->state = TPS_Base;
323323

324324
#ifdef WPARSER_TRACE
325+
/*
326+
* Use of %.*s here is not portable when the string contains multibyte
327+
* characters: some machines interpret the length in characters, others
328+
* in bytes. Since it's only a debugging aid, we haven't bothered to
329+
* fix this.
330+
*/
325331
fprintf(stderr, "parsing \"%.*s\"\n", len, str);
326332
#endif
327333

@@ -361,6 +367,7 @@ TParserCopyInit(const TParser *orig)
361367
prs->state->state = TPS_Base;
362368

363369
#ifdef WPARSER_TRACE
370+
/* See note above about %.*s */
364371
fprintf(stderr, "parsing copy of \"%.*s\"\n", prs->lenstr, prs->str);
365372
#endif
366373

‎src/backend/utils/adt/datetime.c

Lines changed: 10 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
*
99
*
1010
* IDENTIFICATION
11-
* $PostgreSQL: pgsql/src/backend/utils/adt/datetime.c,v 1.210 2010/01/02 16:57:53 momjian Exp $
11+
* $PostgreSQL: pgsql/src/backend/utils/adt/datetime.c,v 1.211 2010/05/08 16:39:51 tgl Exp $
1212
*
1313
*-------------------------------------------------------------------------
1414
*/
@@ -3740,6 +3740,14 @@ EncodeDateTime(struct pg_tm * tm, fsec_t fsec, int *tzp, char **tzn, int style,
37403740

37413741
AppendTimestampSeconds(str + strlen(str), tm, fsec);
37423742

3743+
/*
3744+
* Note: the uses of %.*s in this function would be unportable
3745+
* if the timezone names ever contain non-ASCII characters,
3746+
* since some platforms think the string length is measured
3747+
* in characters not bytes. However, all TZ abbreviations in
3748+
* the Olson database are plain ASCII.
3749+
*/
3750+
37433751
if (tzp != NULL && tm->tm_isdst >= 0)
37443752
{
37453753
if (*tzn != NULL)
@@ -4091,6 +4099,7 @@ CheckDateTokenTable(const char *tablename, const datetkn *base, int nel)
40914099
{
40924100
if (strncmp(base[i - 1].token, base[i].token, TOKMAXLEN) >= 0)
40934101
{
4102+
/* %.*s is safe since all our tokens are ASCII */
40944103
elog(LOG, "ordering error in %s table: \"%.*s\" >= \"%.*s\"",
40954104
tablename,
40964105
TOKMAXLEN, base[i - 1].token,

‎src/backend/utils/error/elog.c

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,7 @@
4242
*
4343
*
4444
* IDENTIFICATION
45-
* $PostgreSQL: pgsql/src/backend/utils/error/elog.c,v 1.223 2010/02/26 02:01:12 momjian Exp $
45+
* $PostgreSQL: pgsql/src/backend/utils/error/elog.c,v 1.224 2010/05/08 16:39:51 tgl Exp $
4646
*
4747
*-------------------------------------------------------------------------
4848
*/
@@ -1871,7 +1871,7 @@ log_line_prefix(StringInfo buf, ErrorData *edata)
18711871
int displen;
18721872

18731873
psdisp = get_ps_display(&displen);
1874-
appendStringInfo(buf, "%.*s", displen, psdisp);
1874+
appendBinaryStringInfo(buf, psdisp, displen);
18751875
}
18761876
break;
18771877
case 'r':
@@ -2029,7 +2029,7 @@ write_csvlog(ErrorData *edata)
20292029
initStringInfo(&msgbuf);
20302030

20312031
psdisp = get_ps_display(&displen);
2032-
appendStringInfo(&msgbuf, "%.*s", displen, psdisp);
2032+
appendBinaryStringInfo(&msgbuf, psdisp, displen);
20332033
appendCSVLiteral(&buf, msgbuf.data);
20342034

20352035
pfree(msgbuf.data);

‎src/bin/psql/command.c

Lines changed: 8 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
*
44
* Copyright (c) 2000-2010, PostgreSQL Global Development Group
55
*
6-
* $PostgreSQL: pgsql/src/bin/psql/command.c,v 1.218 2010/04/03 20:55:57 tgl Exp $
6+
* $PostgreSQL: pgsql/src/bin/psql/command.c,v 1.219 2010/05/08 16:39:51 tgl Exp $
77
*/
88
#include "postgres_fe.h"
99
#include "command.h"
@@ -651,6 +651,13 @@ exec_command(const char *cmd,
651651
{
652652
char *opt = psql_scan_slash_option(scan_state,
653653
OT_WHOLE_LINE, NULL, false);
654+
size_t len;
655+
656+
/* strip any trailing spaces and semicolons */
657+
len = strlen(opt);
658+
while (len > 0 &&
659+
(isspace((unsigned char) opt[len - 1]) || opt[len - 1] == ';'))
660+
opt[--len] = '\0';
654661

655662
helpSQL(opt, pset.popt.topt.pager);
656663
free(opt);

‎src/bin/psql/help.c

Lines changed: 9 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
*
44
* Copyright (c) 2000-2010, PostgreSQL Global Development Group
55
*
6-
* $PostgreSQL: pgsql/src/bin/psql/help.c,v 1.157 2010/03/07 17:02:34 mha Exp $
6+
* $PostgreSQL: pgsql/src/bin/psql/help.c,v 1.158 2010/05/08 16:39:51 tgl Exp $
77
*/
88
#include "postgres_fe.h"
99

@@ -284,6 +284,7 @@ slashUsage(unsigned short int pager)
284284
/*
285285
* helpSQL -- help with SQL commands
286286
*
287+
* Note: we assume caller removed any trailing spaces in "topic".
287288
*/
288289
void
289290
helpSQL(const char *topic, unsigned short int pager)
@@ -352,17 +353,16 @@ helpSQL(const char *topic, unsigned short int pager)
352353
wordlen;
353354
int nl_count = 0;
354355

355-
/* User gets two chances: exact match, then the first word */
356-
357-
/* First pass : strip trailing spaces and semicolons */
356+
/*
357+
* We first try exact match, then first + second words, then first
358+
* word only.
359+
*/
358360
len = strlen(topic);
359-
while (topic[len - 1] == ' ' || topic[len - 1] == ';')
360-
len--;
361361

362-
for (x = 1; x <= 3; x++) /* Three chances to guess that word... */
362+
for (x = 1; x <= 3; x++)
363363
{
364364
if (x > 1) /* Nothing on first pass - try the opening
365-
* words */
365+
* word(s) */
366366
{
367367
wordlen = j = 1;
368368
while (topic[j] != ' ' && j++ < len)
@@ -423,7 +423,7 @@ helpSQL(const char *topic, unsigned short int pager)
423423
}
424424

425425
if (!help_found)
426-
fprintf(output, _("No help available for \"%-.*s\".\nTry \\h with no arguments to see available help.\n"), (int) len, topic);
426+
fprintf(output, _("No help available for \"%s\".\nTry \\h with no arguments to see available help.\n"), topic);
427427

428428
/* Only close if we used the pager */
429429
if (output != stdout)

‎src/bin/psql/print.c

Lines changed: 21 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
*
44
* Copyright (c) 2000-2010, PostgreSQL Global Development Group
55
*
6-
* $PostgreSQL: pgsql/src/bin/psql/print.c,v 1.124 2010/03/01 21:27:26 heikki Exp $
6+
* $PostgreSQL: pgsql/src/bin/psql/print.c,v 1.125 2010/05/08 16:39:52 tgl Exp $
77
*/
88
#include "postgres_fe.h"
99

@@ -252,6 +252,20 @@ format_numeric_locale(const char *my_str)
252252
}
253253

254254

255+
/*
256+
* fputnbytes: print exactly N bytes to a file
257+
*
258+
* Think not to use fprintf with a %.*s format for this. Some machines
259+
* believe %s's precision is measured in characters, others in bytes.
260+
*/
261+
static void
262+
fputnbytes(FILE *f, const char *str, size_t n)
263+
{
264+
while (n-- > 0)
265+
fputc(*str++, f);
266+
}
267+
268+
255269
/*************************/
256270
/* Unaligned text */
257271
/*************************/
@@ -913,14 +927,16 @@ print_aligned_text(const printTableContent *cont, FILE *fout)
913927
{
914928
/* spaces first */
915929
fprintf(fout, "%*s", width_wrap[j] - chars_to_output, "");
916-
fprintf(fout, "%.*s", bytes_to_output,
917-
this_line->ptr + bytes_output[j]);
930+
fputnbytes(fout,
931+
this_line->ptr + bytes_output[j],
932+
bytes_to_output);
918933
}
919934
else /* Left aligned cell */
920935
{
921936
/* spaces second */
922-
fprintf(fout, "%.*s", bytes_to_output,
923-
this_line->ptr + bytes_output[j]);
937+
fputnbytes(fout,
938+
this_line->ptr + bytes_output[j],
939+
bytes_to_output);
924940
}
925941

926942
bytes_output[j] += bytes_to_output;

‎src/interfaces/ecpg/ecpglib/error.c

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
/* $PostgreSQL: pgsql/src/interfaces/ecpg/ecpglib/error.c,v 1.25 2010/03/08 12:15:24 meskes Exp $ */
1+
/* $PostgreSQL: pgsql/src/interfaces/ecpg/ecpglib/error.c,v 1.26 2010/05/08 16:39:52 tgl Exp $ */
22

33
#define POSTGRES_ECPG_INTERNAL
44
#include "postgres_fe.h"
@@ -332,6 +332,7 @@ ecpg_raise_backend(int line, PGresult *result, PGconn *conn, int compat)
332332
else
333333
sqlca->sqlcode = ECPG_PGSQL;
334334

335+
/* %.*s is safe here as long as sqlstate is all-ASCII */
335336
ecpg_log("raising sqlstate %.*s (sqlcode %d): %s\n",
336337
sizeof(sqlca->sqlstate), sqlca->sqlstate, sqlca->sqlcode, sqlca->sqlerrm.sqlerrmc);
337338

‎src/interfaces/ecpg/pgtypeslib/dt_common.c

Lines changed: 9 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
/* $PostgreSQL: pgsql/src/interfaces/ecpg/pgtypeslib/dt_common.c,v 1.51 2009/06/11 14:49:13 momjian Exp $ */
1+
/* $PostgreSQL: pgsql/src/interfaces/ecpg/pgtypeslib/dt_common.c,v 1.52 2010/05/08 16:39:52 tgl Exp $ */
22

33
#include "postgres_fe.h"
44

@@ -855,6 +855,14 @@ EncodeDateTime(struct tm * tm, fsec_t fsec, int *tzp, char **tzn, int style, cha
855855
if (tm->tm_year <= 0)
856856
sprintf(str + strlen(str), " BC");
857857

858+
/*
859+
* Note: the uses of %.*s in this function would be unportable
860+
* if the timezone names ever contain non-ASCII characters,
861+
* since some platforms think the string length is measured
862+
* in characters not bytes. However, all TZ abbreviations in
863+
* the Olson database are plain ASCII.
864+
*/
865+
858866
if (tzp != NULL && tm->tm_isdst >= 0)
859867
{
860868
if (*tzn != NULL)

0 commit comments

Comments
 (0)