Fix generation too long headline with ShortWords.

feodor · feodor · commit 8fd07a35bab3 · 2009-01-15T16:33:28.000Z
Per http://archives.postgresql.org/pgsql-hackers/2008-09/msg01088.php
diff --git 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 1.18 2009/01/01 17:23:48 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.19 2009/01/15 16:33:28 teodor Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -1921,6 +1921,7 @@ mark_hl_fragments(HeadlineParsedText *prs, TSQuery query, int highlight,
 	}
 	pfree(covers);
 }
+
 static void
 mark_hl_words(HeadlineParsedText *prs, TSQuery query, int highlight, 
 		int shortword, int min_words, int max_words)
@@ -1981,12 +1982,14 @@ mark_hl_words(HeadlineParsedText *prs, TSQuery query, int highlight,
 				if (curlen < min_words && i >= prs->curwords)
 				{				/* got end of text and our cover is shoter
 								 * than min_words */
-					for (i = p; i >= 0; i--)
+					for (i = p - 1; i >= 0; i--)
 					{
 						if (!NONWORDTOKEN(prs->words[i].type))
 							curlen++;
 						if (prs->words[i].item && !prs->words[i].repeated)
 							poslen++;
+						if ( curlen >= max_words )
+							break;
 						if (NOENDTOKEN(prs->words[i].type) || prs->words[i].len <= shortword)
 							continue;
 						if (curlen >= min_words)

Original file line number	Diff line number	Diff line change
`@@ -7,7 +7,7 @@`
`7`	`7`	`*`
`8`	`8`	`*`
`9`	`9`	`* IDENTIFICATION`
`10`		`- * $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.18 2009/01/01 17:23:48 momjian Exp $`
	`10`	`+ * $PostgreSQL: pgsql/src/backend/tsearch/wparser_def.c,v 1.19 2009/01/15 16:33:28 teodor Exp $`
`11`	`11`	`*`
`12`	`12`	`*-------------------------------------------------------------------------`
`13`	`13`	`*/`
`@@ -1921,6 +1921,7 @@ mark_hl_fragments(HeadlineParsedText *prs, TSQuery query, int highlight,`
`1921`	`1921`	`}`
`1922`	`1922`	`pfree(covers);`
`1923`	`1923`	`}`
	`1924`	`+`
`1924`	`1925`	`static void`
`1925`	`1926`	`mark_hl_words(HeadlineParsedText *prs, TSQuery query, int highlight,`
`1926`	`1927`	`int shortword, int min_words, int max_words)`
`@@ -1981,12 +1982,14 @@ mark_hl_words(HeadlineParsedText *prs, TSQuery query, int highlight,`
`1981`	`1982`	`if (curlen < min_words && i >= prs->curwords)`
`1982`	`1983`	`{ /* got end of text and our cover is shoter`
`1983`	`1984`	`* than min_words */`
`1984`		`- for (i = p; i >= 0; i--)`
	`1985`	`+ for (i = p - 1; i >= 0; i--)`
`1985`	`1986`	`{`
`1986`	`1987`	`if (!NONWORDTOKEN(prs->words[i].type))`
`1987`	`1988`	`curlen++;`
`1988`	`1989`	`if (prs->words[i].item && !prs->words[i].repeated)`
`1989`	`1990`	`poslen++;`
	`1991`	`+ if ( curlen >= max_words )`
	`1992`	`+ break;`
`1990`	`1993`	`if (NOENDTOKEN(prs->words[i].type) \|\| prs->words[i].len <= shortword)`
`1991`	`1994`	`continue;`
`1992`	`1995`	`if (curlen >= min_words)`