aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorTeodor Sigaev <teodor@sigaev.ru>2009-01-15 17:06:03 +0000
committerTeodor Sigaev <teodor@sigaev.ru>2009-01-15 17:06:03 +0000
commit87042a196949eb63037658cc52a143d13b4ac06e (patch)
treedb23de6c321a4070adadb2f09e710e07ec016424
parent90208cc2fbc613ed7bdacd80c064de51378786c1 (diff)
downloadpostgresql-87042a196949eb63037658cc52a143d13b4ac06e.tar.gz
postgresql-87042a196949eb63037658cc52a143d13b4ac06e.zip
Fix URL generation in headline. Only tag lexeme will be replaced by space.
Per http://archives.postgresql.org/pgsql-bugs/2008-12/msg00013.php
-rw-r--r--src/backend/tsearch/ts_parse.c4
-rw-r--r--src/backend/tsearch/wparser_def.c19
-rw-r--r--src/include/tsearch/ts_public.h5
3 files changed, 16 insertions, 12 deletions
diff --git a/src/backend/tsearch/ts_parse.c b/src/backend/tsearch/ts_parse.c
index 22862bdb806..9f12d66bf6e 100644
--- a/src/backend/tsearch/ts_parse.c
+++ b/src/backend/tsearch/ts_parse.c
@@ -7,7 +7,7 @@
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/tsearch/ts_parse.c,v 1.7 2008/01/01 19:45:52 momjian Exp $
+ * $PostgreSQL: pgsql/src/backend/tsearch/ts_parse.c,v 1.7.2.1 2009/01/15 17:06:03 teodor Exp $
*
*-------------------------------------------------------------------------
*/
@@ -607,7 +607,7 @@ generateHeadline(HeadlineParsedText *prs)
*ptr = ' ';
ptr++;
}
- else
+ else if (!wrd->skip)
{
if (wrd->selected)
{
diff --git a/src/backend/tsearch/wparser_def.c b/src/backend/tsearch/wparser_def.c
index 24f371ab95d..0ab06ad0f15 100644
--- a/src/backend/tsearch/wparser_def.c
+++ b/src/backend/tsearch/wparser_def.c
@@ -7,7 +7,7 @@
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.14.2.2 2009/01/15 17:05:39 teodor Exp $
+ * $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.14.2.3 2009/01/15 17:06:03 teodor Exp $
*
*-------------------------------------------------------------------------
*/
@@ -1587,10 +1587,11 @@ prsd_end(PG_FUNCTION_ARGS)
#define COMPLEXTOKEN(x) ( (x)==URL_T || (x)==NUMHWORD || (x)==ASCIIHWORD || (x)==HWORD )
#define ENDPUNCTOKEN(x) ( (x)==SPACE )
-#define TS_IDIGNORE(x) ( (x)==TAG_T || (x)==PROTOCOL || (x)==SPACE || (x)==XMLENTITY )
-#define HLIDIGNORE(x) ( (x)==URL_T || (x)==TAG_T || (x)==NUMHWORD || (x)==ASCIIHWORD || (x)==HWORD )
-#define XMLHLIDIGNORE(x) ( (x)==URL_T || (x)==NUMHWORD || (x)==ASCIIHWORD || (x)==HWORD )
-#define NONWORDTOKEN(x) ( (x)==SPACE || HLIDIGNORE(x) )
+#define TS_IDIGNORE(x) ( (x)==TAG_T || (x)==PROTOCOL || (x)==SPACE || (x)==XMLENTITY )
+#define HLIDREPLACE(x) ( (x)==TAG_T )
+#define HLIDSKIP(x) ( (x)==URL_T || (x)==NUMHWORD || (x)==ASCIIHWORD || (x)==HWORD )
+#define XMLHLIDSKIP(x) ( (x)==URL_T || (x)==NUMHWORD || (x)==ASCIIHWORD || (x)==HWORD )
+#define NONWORDTOKEN(x) ( (x)==SPACE || HLIDREPLACE(x) || HLIDSKIP(x) )
#define NOENDTOKEN(x) ( NONWORDTOKEN(x) || (x)==SCIENTIFIC || (x)==VERSIONNUMBER || (x)==DECIMAL || (x)==SIGNEDINT || (x)==UNSIGNEDINT || TS_IDIGNORE(x) )
typedef struct
@@ -1867,13 +1868,15 @@ prsd_headline(PG_FUNCTION_ARGS)
prs->words[i].selected = 1;
if (highlight == 0)
{
- if (HLIDIGNORE(prs->words[i].type))
+ if (HLIDREPLACE(prs->words[i].type))
prs->words[i].replace = 1;
+ else if ( HLIDSKIP(prs->words[i].type) )
+ prs->words[i].skip = 1;
}
else
{
- if (XMLHLIDIGNORE(prs->words[i].type))
- prs->words[i].replace = 1;
+ if (XMLHLIDSKIP(prs->words[i].type))
+ prs->words[i].skip = 1;
}
prs->words[i].in = (prs->words[i].repeated) ? 0 : 1;
diff --git a/src/include/tsearch/ts_public.h b/src/include/tsearch/ts_public.h
index 8f21abf032c..071965b132a 100644
--- a/src/include/tsearch/ts_public.h
+++ b/src/include/tsearch/ts_public.h
@@ -6,7 +6,7 @@
*
* Copyright (c) 1998-2008, PostgreSQL Global Development Group
*
- * $PostgreSQL: pgsql/src/include/tsearch/ts_public.h,v 1.8 2008/01/01 19:45:59 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/tsearch/ts_public.h,v 1.8.2.1 2009/01/15 17:06:03 teodor Exp $
*
*-------------------------------------------------------------------------
*/
@@ -38,7 +38,8 @@ typedef struct
in:1,
replace:1,
repeated:1,
- unused:4,
+ skip:1,
+ unused:3,
type:8,
len:16;
char *word;