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

Commit75a64ee

Browse files
committed
I made the patch that implements regexp_replace again.
The specification of this function is as follows.regexp_replace(source text, pattern text, replacement text, [flagstext])returns textReplace string that matches to regular expression in source text toreplacement text. - pattern is regular expression pattern. - replacement is replace string that can use '\1'-'\9', and '\&'. '\1'-'\9': back reference to the n'th subexpression. '\&' : entire matched string. - flags can use the following values: g: global (replace all) i: ignore case When the flags is not specified, case sensitive, replace the first instance only.Atsushi Ogawa
1 parent73e2431 commit75a64ee

File tree

10 files changed

+418
-55
lines changed

10 files changed

+418
-55
lines changed

‎doc/src/sgml/func.sgml

Lines changed: 21 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
<!--
2-
$PostgreSQL: pgsql/doc/src/sgml/func.sgml,v 1.263 2005/07/06 19:02:52 momjian Exp $
2+
$PostgreSQL: pgsql/doc/src/sgml/func.sgml,v 1.264 2005/07/10 04:54:30 momjian Exp $
33
PostgreSQL documentation
44
-->
55

@@ -1256,6 +1256,26 @@ PostgreSQL documentation
12561256
<entry><literal>'O''Reilly'</literal></entry>
12571257
</row>
12581258

1259+
<row>
1260+
<entry><literal><function>regexp_replace</function>(<parameter>source</parameter> <type>text</type>,
1261+
<parameter>pattern</parameter> <type>text</type>,
1262+
<parameter>replacement</parameter> <type>text</type>
1263+
<optional>, <parameter>flags</parameter> <type>text</type></optional>)</literal></entry>
1264+
<entry><type>text</type></entry>
1265+
<entry>Replace string that matches the regular expression
1266+
<parameter>pattern</parameter> in <parameter>source</parameter> to
1267+
<parameter>replacement</parameter>.
1268+
<parameter>replacement</parameter> can use <literal>\1</>-<literal>\9</> and <literal>\&amp;</>.
1269+
<literal>\1</>-<literal>\9</> is a back reference to the n'th subexpression, and
1270+
<literal>\&amp;</> is the entire matched string.
1271+
<parameter>flags</parameter> can use <literal>g</>(global) and <literal>i</>(ignore case).
1272+
When flags is not specified, case sensitive matching is used, and it replaces
1273+
only the instance.
1274+
</entry>
1275+
<entry><literal>regexp_replace('1112223333', '(\\d{3})(\\d{3})(\\d{4})', '(\\1) \\2-\\3')</literal></entry>
1276+
<entry><literal>(111) 222-3333</literal></entry>
1277+
</row>
1278+
12591279
<row>
12601280
<entry><literal><function>repeat</function>(<parameter>string</parameter> <type>text</type>, <parameter>number</parameter> <type>integer</type>)</literal></entry>
12611281
<entry><type>text</type></entry>

‎src/backend/regex/regexec.c

Lines changed: 7 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -27,7 +27,7 @@
2727
* OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
2828
* ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
2929
*
30-
* $PostgreSQL: pgsql/src/backend/regex/regexec.c,v 1.24 2003/11/29 19:51:55 pgsql Exp $
30+
* $PostgreSQL: pgsql/src/backend/regex/regexec.c,v 1.25 2005/07/10 04:54:30 momjian Exp $
3131
*
3232
*/
3333

@@ -110,6 +110,7 @@ struct vars
110110
regmatch_t*pmatch;
111111
rm_detail_t*details;
112112
chr*start;/* start of string */
113+
chr*search_start;/* search start of string */
113114
chr*stop;/* just past end of string */
114115
interr;/* error code if any (0 none) */
115116
regoff_t*mem;/* memory vector for backtracking */
@@ -168,6 +169,7 @@ int
168169
pg_regexec(regex_t*re,
169170
constchr*string,
170171
size_tlen,
172+
size_tsearch_start,
171173
rm_detail_t*details,
172174
size_tnmatch,
173175
regmatch_tpmatch[],
@@ -219,6 +221,7 @@ pg_regexec(regex_t *re,
219221
v->pmatch=pmatch;
220222
v->details=details;
221223
v->start= (chr*)string;
224+
v->search_start= (chr*)string+search_start;
222225
v->stop= (chr*)string+len;
223226
v->err=0;
224227
if (backref)
@@ -288,7 +291,8 @@ find(struct vars * v,
288291
NOERR();
289292
MDEBUG(("\nsearch at %ld\n",LOFF(v->start)));
290293
cold=NULL;
291-
close=shortest(v,s,v->start,v->start,v->stop,&cold, (int*)NULL);
294+
close=shortest(v,s,v->search_start,v->search_start,v->stop,
295+
&cold, (int*)NULL);
292296
freedfa(s);
293297
NOERR();
294298
if (v->g->cflags&REG_EXPECT)
@@ -415,7 +419,7 @@ cfindloop(struct vars * v,
415419

416420
assert(d!=NULL&&s!=NULL);
417421
cold=NULL;
418-
close=v->start;
422+
close=v->search_start;
419423
do
420424
{
421425
MDEBUG(("\ncsearch at %ld\n",LOFF(close)));

‎src/backend/utils/adt/regexp.c

Lines changed: 121 additions & 44 deletions
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,7 @@
88
*
99
*
1010
* IDENTIFICATION
11-
* $PostgreSQL: pgsql/src/backend/utils/adt/regexp.c,v 1.56 2004/12/31 22:01:22 pgsql Exp $
11+
* $PostgreSQL: pgsql/src/backend/utils/adt/regexp.c,v 1.57 2005/07/10 04:54:30 momjian Exp $
1212
*
1313
*Alistair Crooks added the code for the regex caching
1414
*agc - cached the regular expressions used - there's a good chance
@@ -81,38 +81,27 @@ static cached_re_str re_array[MAX_CACHED_RES];/* cached re's */
8181

8282

8383
/*
84-
*RE_compile_and_execute - compile and execute a RE, caching if possible
84+
*RE_compile_and_cache - compile a RE, caching if possible
8585
*
86-
* ReturnsTRUE on match, FALSE on no match
86+
* Returnsregex_t
8787
*
88-
*text_re --- the pattern, expressed as an *untoasted* TEXT object
89-
*dat --- the data to match against (need not be null-terminated)
90-
*dat_len --- the length of the data string
91-
*cflags --- compile options for the pattern
92-
*nmatch, pmatch--- optional return area for match details
88+
* text_re --- the pattern, expressed as an *untoasted* TEXT object
89+
* cflags --- compile options for the pattern
9390
*
94-
*Both pattern and data aregiven in the database encoding. We internally
95-
*convert toarray of pg_wchar which is what Spencer's regex package wants.
91+
*Pattern isgiven in the database encoding. We internally convert to
92+
* array of pg_wchar which is what Spencer's regex package wants.
9693
*/
97-
staticbool
98-
RE_compile_and_execute(text*text_re,unsignedchar*dat,intdat_len,
99-
intcflags,intnmatch,regmatch_t*pmatch)
94+
staticregex_t
95+
RE_compile_and_cache(text*text_re,intcflags)
10096
{
10197
inttext_re_len=VARSIZE(text_re);
102-
pg_wchar*data;
103-
size_tdata_len;
10498
pg_wchar*pattern;
10599
size_tpattern_len;
106100
inti;
107101
intregcomp_result;
108-
intregexec_result;
109102
cached_re_strre_temp;
110103
charerrMsg[100];
111104

112-
/* Convert data string to wide characters */
113-
data= (pg_wchar*)palloc((dat_len+1)*sizeof(pg_wchar));
114-
data_len=pg_mb2wchar_with_len(dat,data,dat_len);
115-
116105
/*
117106
* Look for a match among previously compiled REs.Since the data
118107
* structure is self-organizing with most-used entries at the front,
@@ -134,28 +123,7 @@ RE_compile_and_execute(text *text_re, unsigned char *dat, int dat_len,
134123
re_array[0]=re_temp;
135124
}
136125

137-
/* Perform RE match and return result */
138-
regexec_result=pg_regexec(&re_array[0].cre_re,
139-
data,
140-
data_len,
141-
NULL,/* no details */
142-
nmatch,
143-
pmatch,
144-
0);
145-
146-
pfree(data);
147-
148-
if (regexec_result!=REG_OKAY&&regexec_result!=REG_NOMATCH)
149-
{
150-
/* re failed??? */
151-
pg_regerror(regexec_result,&re_array[0].cre_re,
152-
errMsg,sizeof(errMsg));
153-
ereport(ERROR,
154-
(errcode(ERRCODE_INVALID_REGULAR_EXPRESSION),
155-
errmsg("regular expression failed: %s",errMsg)));
156-
}
157-
158-
return (regexec_result==REG_OKAY);
126+
returnre_array[0].cre_re;
159127
}
160128
}
161129

@@ -220,10 +188,45 @@ RE_compile_and_execute(text *text_re, unsigned char *dat, int dat_len,
220188
re_array[0]=re_temp;
221189
num_res++;
222190

191+
returnre_array[0].cre_re;
192+
}
193+
194+
/*
195+
* RE_compile_and_execute - compile and execute a RE
196+
*
197+
* Returns TRUE on match, FALSE on no match
198+
*
199+
*text_re --- the pattern, expressed as an *untoasted* TEXT object
200+
*dat --- the data to match against (need not be null-terminated)
201+
*dat_len --- the length of the data string
202+
*cflags --- compile options for the pattern
203+
*nmatch, pmatch--- optional return area for match details
204+
*
205+
* Both pattern and data are given in the database encoding. We internally
206+
* convert to array of pg_wchar which is what Spencer's regex package wants.
207+
*/
208+
staticbool
209+
RE_compile_and_execute(text*text_re,unsignedchar*dat,intdat_len,
210+
intcflags,intnmatch,regmatch_t*pmatch)
211+
{
212+
pg_wchar*data;
213+
size_tdata_len;
214+
intregexec_result;
215+
regex_tre;
216+
charerrMsg[100];
217+
218+
/* Convert data string to wide characters */
219+
data= (pg_wchar*)palloc((dat_len+1)*sizeof(pg_wchar));
220+
data_len=pg_mb2wchar_with_len(dat,data,dat_len);
221+
222+
/* Compile RE */
223+
re=RE_compile_and_cache(text_re,cflags);
224+
223225
/* Perform RE match and return result */
224226
regexec_result=pg_regexec(&re_array[0].cre_re,
225227
data,
226228
data_len,
229+
0,
227230
NULL,/* no details */
228231
nmatch,
229232
pmatch,
@@ -428,15 +431,89 @@ textregexsubstr(PG_FUNCTION_ARGS)
428431
eo=pmatch[0].rm_eo;
429432
}
430433

431-
return(DirectFunctionCall3(text_substr,
434+
returnDirectFunctionCall3(text_substr,
432435
PointerGetDatum(s),
433436
Int32GetDatum(so+1),
434-
Int32GetDatum(eo-so)));
437+
Int32GetDatum(eo-so));
435438
}
436439

437440
PG_RETURN_NULL();
438441
}
439442

443+
/*
444+
* textregexreplace_noopt()
445+
* Return a replace string matched by a regular expression.
446+
*This function is a version that doesn't specify the option of
447+
*textregexreplace. This is case sensitive, replace the first
448+
*instance only.
449+
*/
450+
Datum
451+
textregexreplace_noopt(PG_FUNCTION_ARGS)
452+
{
453+
text*s=PG_GETARG_TEXT_P(0);
454+
text*p=PG_GETARG_TEXT_P(1);
455+
text*r=PG_GETARG_TEXT_P(2);
456+
regex_tre;
457+
458+
re=RE_compile_and_cache(p,regex_flavor);
459+
460+
returnDirectFunctionCall4(replace_text_regexp,
461+
PointerGetDatum(s),
462+
PointerGetDatum(&re),
463+
PointerGetDatum(r),
464+
BoolGetDatum(false));
465+
}
466+
467+
/*
468+
* textregexreplace()
469+
* Return a replace string matched by a regular expression.
470+
*/
471+
Datum
472+
textregexreplace(PG_FUNCTION_ARGS)
473+
{
474+
text*s=PG_GETARG_TEXT_P(0);
475+
text*p=PG_GETARG_TEXT_P(1);
476+
text*r=PG_GETARG_TEXT_P(2);
477+
text*opt=PG_GETARG_TEXT_P(3);
478+
char*opt_p=VARDATA(opt);
479+
intopt_len= (VARSIZE(opt)-VARHDRSZ);
480+
inti;
481+
boolglobal= false;
482+
boolignorecase= false;
483+
regex_tre;
484+
485+
/* parse options */
486+
for (i=0;i<opt_len;i++)
487+
{
488+
switch (opt_p[i])
489+
{
490+
case'i':
491+
ignorecase= true;
492+
break;
493+
case'g':
494+
global= true;
495+
break;
496+
default:
497+
ereport(ERROR,
498+
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
499+
errmsg("invalid option of regexp_replace: %c",
500+
opt_p[i])));
501+
break;
502+
}
503+
}
504+
505+
if (ignorecase)
506+
re=RE_compile_and_cache(p,regex_flavor |REG_ICASE);
507+
else
508+
re=RE_compile_and_cache(p,regex_flavor);
509+
510+
returnDirectFunctionCall4(replace_text_regexp,
511+
PointerGetDatum(s),
512+
PointerGetDatum(&re),
513+
PointerGetDatum(r),
514+
BoolGetDatum(global));
515+
}
516+
440517
/* similar_escape()
441518
* Convert a SQL99 regexp pattern to POSIX style, so it can be used by
442519
* our regexp engine.

0 commit comments

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