mirrors/postgres
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix regexp substring matching (substring(string from pattern)) for the corner

Browse Source 
case where there is a match to the pattern overall but the user has specified
a parenthesized subexpression and that subexpression hasn't got a match.
An example is substring('foo' from 'foo(bar)?').  This should return NULL,
since (bar) isn't matched, but it was mistakenly returning the whole-pattern
match instead (ie, 'foo').  Per bug #4044 from Rui Martins.

This has been broken since the beginning; patch in all supported versions.
The old behavior was sufficiently inconsistent that it's impossible to believe
anyone is depending on it.
This commit is contained in:
Tom Lane 2008-03-19 02:40:37 +00:00
parent 8436f9a036
commit 965a2a191a
1 changed files with 32 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

56
src/backend/utils/adt/regexp.c
View File

@ -8,7 +8,7 @@
* *
* *
* IDENTIFICATION * IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/utils/adt/regexp.c,v 1.78 2008/01/01 19:45:52 momjian Exp $ * $PostgreSQL: pgsql/src/backend/utils/adt/regexp.c,v 1.79 2008/03/19 02:40:37 tgl Exp $
* *
* Alistair Crooks added the code for the regex caching * Alistair Crooks added the code for the regex caching
* agc - cached the regular expressions used - there's a good chance * agc - cached the regular expressions used - there's a good chance
@ -576,8 +576,13 @@ textregexsubstr(PG_FUNCTION_ARGS)
{ {
text *s = PG_GETARG_TEXT_PP(0); text *s = PG_GETARG_TEXT_PP(0);
text *p = PG_GETARG_TEXT_PP(1); text *p = PG_GETARG_TEXT_PP(1);
bool match; regex_t *re;
regmatch_t pmatch[2]; regmatch_t pmatch[2];
int so,
eo;
/* Compile RE */
re = RE_compile_and_cache(p, regex_flavor);
/* /*
* We pass two regmatch_t structs to get info about the overall match and * We pass two regmatch_t structs to get info about the overall match and
@ -585,34 +590,37 @@ textregexsubstr(PG_FUNCTION_ARGS)
* is a parenthesized subexpression, we return what it matched; else * is a parenthesized subexpression, we return what it matched; else
* return what the whole regexp matched. * return what the whole regexp matched.
*/ */
match = RE_compile_and_execute(p, if (!RE_execute(re,
VARDATA_ANY(s), VARDATA_ANY(s), VARSIZE_ANY_EXHDR(s),
VARSIZE_ANY_EXHDR(s), 2, pmatch))
regex_flavor, PG_RETURN_NULL(); /* definitely no match */
2, pmatch);
/* match? then return the substring matching the pattern */ if (re->re_nsub > 0)
if (match)
{ {
int so, /* has parenthesized subexpressions, use the first one */
eo;
so = pmatch[1].rm_so; so = pmatch[1].rm_so;
eo = pmatch[1].rm_eo; eo = pmatch[1].rm_eo;
if (so < 0 || eo < 0) }
{ else
/* no parenthesized subexpression */ {
so = pmatch[0].rm_so; /* no parenthesized subexpression, use whole match */
eo = pmatch[0].rm_eo; so = pmatch[0].rm_so;
} eo = pmatch[0].rm_eo;
return DirectFunctionCall3(text_substr,
PointerGetDatum(s),
Int32GetDatum(so + 1),
Int32GetDatum(eo - so));
} }
PG_RETURN_NULL(); /*
* It is possible to have a match to the whole pattern but no match
* for a subexpression; for example 'foo(bar)?' is considered to match
* 'foo' but there is no subexpression match. So this extra test for
* match failure is not redundant.
*/
if (so < 0 || eo < 0)
PG_RETURN_NULL();
return DirectFunctionCall3(text_substr,
PointerGetDatum(s),
Int32GetDatum(so + 1),
Int32GetDatum(eo - so));
} }
/* /*