1
0
mirror of https://github.com/postgres/postgres.git synced 2025-07-27 12:41:57 +03:00

Add strict_word_similarity to pg_trgm module

strict_word_similarity is similar to existing word_similarity function but
it takes into account word boundaries to compute similarity.

Author: Alexander Korotkov
Review by: David Steele, Liudmila Mantrova, me
Discussion: https://www.postgresql.org/message-id/flat/CY4PR17MB13207ED8310F847CF117EED0D85A0@CY4PR17MB1320.namprd17.prod.outlook.com
This commit is contained in:
Teodor Sigaev
2018-03-21 14:57:42 +03:00
parent f20b328534
commit be8a7a6866
10 changed files with 1461 additions and 61 deletions

View File

@ -221,6 +221,7 @@ gtrgm_consistent(PG_FUNCTION_ARGS)
{
case SimilarityStrategyNumber:
case WordSimilarityStrategyNumber:
case StrictWordSimilarityStrategyNumber:
qtrg = generate_trgm(VARDATA(query),
querysize - VARHDRSZ);
break;
@ -290,10 +291,11 @@ gtrgm_consistent(PG_FUNCTION_ARGS)
{
case SimilarityStrategyNumber:
case WordSimilarityStrategyNumber:
/* Similarity search is exact. Word similarity search is inexact */
*recheck = (strategy == WordSimilarityStrategyNumber);
nlimit = (strategy == SimilarityStrategyNumber) ?
similarity_threshold : word_similarity_threshold;
case StrictWordSimilarityStrategyNumber:
/* Similarity search is exact. (Strict) word similarity search is inexact */
*recheck = (strategy != SimilarityStrategyNumber);
nlimit = index_strategy_get_limit(strategy);
if (GIST_LEAF(entry))
{ /* all leafs contains orig trgm */
@ -468,7 +470,9 @@ gtrgm_distance(PG_FUNCTION_ARGS)
{
case DistanceStrategyNumber:
case WordDistanceStrategyNumber:
*recheck = strategy == WordDistanceStrategyNumber;
case StrictWordDistanceStrategyNumber:
/* Only plain trigram distance is exact */
*recheck = (strategy != DistanceStrategyNumber);
if (GIST_LEAF(entry))
{ /* all leafs contains orig trgm */