diff options
| author | Teodor Sigaev | 2009-01-15 16:33:59 +0000 |
|---|---|---|
| committer | Teodor Sigaev | 2009-01-15 16:33:59 +0000 |
| commit | 4921bde61906314125edae959b81cbc80c1b4e83 (patch) | |
| tree | 974fd7dce35ad7d1e9720a4de75e3ae24fd50526 | |
| parent | f6ffaf8045664bc5b86b3d9cd9175b5f321d658e (diff) | |
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.c | 2 | ||||
| -rw-r--r-- | src/backend/tsearch/wparser_def.c | 25 | ||||
| -rw-r--r-- | src/include/tsearch/ts_public.h | 3 |
3 files changed, 18 insertions, 12 deletions
diff --git a/src/backend/tsearch/ts_parse.c b/src/backend/tsearch/ts_parse.c index d717bb3dce..20f010f604 100644 --- a/src/backend/tsearch/ts_parse.c +++ b/src/backend/tsearch/ts_parse.c @@ -625,7 +625,7 @@ 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 94ff4087ed..190b882581 100644 --- a/src/backend/tsearch/wparser_def.c +++ b/src/backend/tsearch/wparser_def.c @@ -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 @@ -1695,13 +1696,15 @@ mark_fragment(HeadlineParsedText *prs, int highlight, int startpos, int endpos) 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; @@ -2050,13 +2053,15 @@ mark_hl_words(HeadlineParsedText *prs, TSQuery query, int highlight, 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 c2d4be0105..0558ae36e2 100644 --- a/src/include/tsearch/ts_public.h +++ b/src/include/tsearch/ts_public.h @@ -38,7 +38,8 @@ typedef struct in:1, replace:1, repeated:1, - unused:4, + skip:1, + unused:3, type:8, len:16; char *word; |