Fix tests: use indexes and check various forms with tsquery

postgrespro · May 31, 2016 · ddbac4a · ddbac4a
1 parent 141f536
commit ddbac4a
Show file tree

Hide file tree

Showing 12 changed files with 463 additions and 331 deletions.
diff --git a/hunspell_de_de/expected/hunspell_de_de.out b/hunspell_de_de/expected/hunspell_de_de.out
@@ -1,55 +1,59 @@
 CREATE EXTENSION hunspell_de_de;
-SELECT ts_lexize('german_hunspell', 'überwachung');
-  ts_lexize   
---------------
- {überwachen}
-(1 row)
-
-SELECT ts_lexize('german_hunspell', 'etikettierung');
-   ts_lexize    
-----------------
- {etikettieren}
-(1 row)
-
-SELECT ts_lexize('german_hunspell', 'vernehmen');
-     ts_lexize      
---------------------
- {vernehmen,nehmen}
+CREATE TABLE table1(name varchar);
+INSERT INTO table1 VALUES ('etikettierung'), ('vernehmen'), ('verteilen'),
+						('bereits'), ('wahrnehmung');
+SELECT d.* FROM table1 AS t, LATERAL ts_debug('german_hunspell', t.name) AS d;
+   alias   |   description   |     token     |         dictionaries          |   dictionary    |            lexemes            
+-----------+-----------------+---------------+-------------------------------+-----------------+-------------------------------
+ asciiword | Word, all ASCII | etikettierung | {german_hunspell,german_stem} | german_hunspell | {etikettieren}
+ asciiword | Word, all ASCII | vernehmen     | {german_hunspell,german_stem} | german_hunspell | {vernehmen,nehmen}
+ asciiword | Word, all ASCII | verteilen     | {german_hunspell,german_stem} | german_hunspell | {verteilen,teilen,teile,teil}
+ asciiword | Word, all ASCII | bereits       | {german_hunspell,german_stem} | german_hunspell | {bereit}
+ asciiword | Word, all ASCII | wahrnehmung   | {german_hunspell,german_stem} | german_hunspell | {wahrnehmen}
+(5 rows)
+
+CREATE INDEX name_idx ON table1 USING GIN (to_tsvector('german_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'vernehmen');
+   name    
+-----------
+ vernehmen
 (1 row)
 
-SELECT ts_lexize('german_hunspell', 'verteilen');
-           ts_lexize           
--------------------------------
- {verteilen,teilen,teile,teil}
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'bereits');
+  name   
+---------
+ bereits
 (1 row)
 
-SELECT ts_lexize('german_hunspell', 'bereits');
- ts_lexize 
------------
- {bereit}
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'wahrnehmung');
+    name     
+-------------
+ wahrnehmung
 (1 row)
 
-SELECT ts_lexize('german_hunspell', 'üblichen');
- ts_lexize 
+DROP INDEX name_idx;
+CREATE INDEX name_idx ON table1 USING GIST (to_tsvector('german_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'vernehmen');
+   name    
 -----------
- {üblich}
+ vernehmen
 (1 row)
 
-SELECT ts_lexize('german_hunspell', 'wahrnehmung');
-  ts_lexize   
---------------
- {wahrnehmen}
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'bereits');
+  name   
+---------
+ bereits
 (1 row)
 
-SELECT ts_lexize('german_hunspell', 'maßen');
- ts_lexize 
------------
- {maß}
-(1 row)
-
-SELECT ts_lexize('german_hunspell', 'füße');
- ts_lexize 
------------
- {fuß}
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'wahrnehmung');
+    name     
+-------------
+ wahrnehmung
 (1 row)
 
diff --git a/hunspell_de_de/sql/hunspell_de_de.sql b/hunspell_de_de/sql/hunspell_de_de.sql
@@ -1,11 +1,24 @@
 CREATE EXTENSION hunspell_de_de;
 
-SELECT ts_lexize('german_hunspell', 'überwachung');
-SELECT ts_lexize('german_hunspell', 'etikettierung');
-SELECT ts_lexize('german_hunspell', 'vernehmen');
-SELECT ts_lexize('german_hunspell', 'verteilen');
-SELECT ts_lexize('german_hunspell', 'bereits');
-SELECT ts_lexize('german_hunspell', 'üblichen');
-SELECT ts_lexize('german_hunspell', 'wahrnehmung');
-SELECT ts_lexize('german_hunspell', 'maßen');
-SELECT ts_lexize('german_hunspell', 'füße');
+CREATE TABLE table1(name varchar);
+INSERT INTO table1 VALUES ('etikettierung'), ('vernehmen'), ('verteilen'),
+						('bereits'), ('wahrnehmung');
+
+SELECT d.* FROM table1 AS t, LATERAL ts_debug('german_hunspell', t.name) AS d;
+
+CREATE INDEX name_idx ON table1 USING GIN (to_tsvector('german_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'vernehmen');
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'bereits');
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'wahrnehmung');
+
+DROP INDEX name_idx;
+CREATE INDEX name_idx ON table1 USING GIST (to_tsvector('german_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'vernehmen');
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'bereits');
+SELECT * FROM table1 WHERE to_tsvector('german_hunspell', name)
+	@@ to_tsquery('german_hunspell', 'wahrnehmung');
diff --git a/hunspell_en_us/expected/hunspell_en_us.out b/hunspell_en_us/expected/hunspell_en_us.out
@@ -1,61 +1,77 @@
 CREATE EXTENSION hunspell_en_us;
-SELECT ts_lexize('english_hunspell', 'stories');
- ts_lexize 
------------
- {story}
-(1 row)
+CREATE TABLE table1(name varchar);
+INSERT INTO table1 VALUES ('leaves'), ('leaved'), ('leaving'),
+						('inability'), ('abilities'), ('disability'), ('ability');
+SELECT d.* FROM table1 AS t, LATERAL ts_debug('english_hunspell', t.name) AS d;
+   alias   |   description   |   token    |          dictionaries           |    dictionary    |  lexemes  
+-----------+-----------------+------------+---------------------------------+------------------+-----------
+ asciiword | Word, all ASCII | leaves     | {english_hunspell,english_stem} | english_hunspell | {leave}
+ asciiword | Word, all ASCII | leaved     | {english_hunspell,english_stem} | english_hunspell | {leave}
+ asciiword | Word, all ASCII | leaving    | {english_hunspell,english_stem} | english_hunspell | {leave}
+ asciiword | Word, all ASCII | inability  | {english_hunspell,english_stem} | english_hunspell | {ability}
+ asciiword | Word, all ASCII | abilities  | {english_hunspell,english_stem} | english_hunspell | {ability}
+ asciiword | Word, all ASCII | disability | {english_hunspell,english_stem} | english_hunspell | {ability}
+ asciiword | Word, all ASCII | ability    | {english_hunspell,english_stem} | english_hunspell | {ability}
+(7 rows)
 
-SELECT ts_lexize('english_hunspell', 'traveled');
-     ts_lexize     
--------------------
- {traveled,travel}
-(1 row)
+CREATE INDEX name_idx ON table1 USING GIN (to_tsvector('english_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'leaving');
+  name   
+---------
+ leaves
+ leaved
+ leaving
+(3 rows)
 
-SELECT ts_lexize('english_hunspell', 'eaten');
-  ts_lexize  
--------------
- {eaten,eat}
-(1 row)
-
-SELECT ts_lexize('english_hunspell', 'I''m');
- ts_lexize 
------------
- {i'm}
-(1 row)
-
-SELECT ts_lexize('english_hunspell', 'Saturdays');
- ts_lexize  
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'abilities');
+    name    
 ------------
- {saturday}
-(1 row)
-
-SELECT ts_lexize('english_hunspell', 'healthcare');
-  ts_lexize   
---------------
- {healthcare}
-(1 row)
+ inability
+ abilities
+ disability
+ ability
+(4 rows)
 
-SELECT ts_lexize('english_hunspell', 'generally');
- ts_lexize 
------------
- {general}
-(1 row)
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'ability');
+    name    
+------------
+ inability
+ abilities
+ disability
+ ability
+(4 rows)
 
-SELECT ts_lexize('english_hunspell', 'integrating');
-  ts_lexize  
--------------
- {integrate}
-(1 row)
+DROP INDEX name_idx;
+CREATE INDEX name_idx ON table1 USING GIST (to_tsvector('english_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'leaving');
+  name   
+---------
+ leaves
+ leaved
+ leaving
+(3 rows)
 
-SELECT ts_lexize('english_hunspell', 'lankiness''s');
-  ts_lexize  
--------------
- {lankiness}
-(1 row)
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'abilities');
+    name    
+------------
+ inability
+ abilities
+ disability
+ ability
+(4 rows)
 
-SELECT ts_lexize('english_hunspell', 'rewritten');
- ts_lexize 
------------
- {written}
-(1 row)
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'ability');
+    name    
+------------
+ inability
+ abilities
+ disability
+ ability
+(4 rows)
 
diff --git a/hunspell_en_us/sql/hunspell_en_us.sql b/hunspell_en_us/sql/hunspell_en_us.sql
@@ -1,12 +1,24 @@
 CREATE EXTENSION hunspell_en_us;
 
-SELECT ts_lexize('english_hunspell', 'stories');
-SELECT ts_lexize('english_hunspell', 'traveled');
-SELECT ts_lexize('english_hunspell', 'eaten');
-SELECT ts_lexize('english_hunspell', 'I''m');
-SELECT ts_lexize('english_hunspell', 'Saturdays');
-SELECT ts_lexize('english_hunspell', 'healthcare');
-SELECT ts_lexize('english_hunspell', 'generally');
-SELECT ts_lexize('english_hunspell', 'integrating');
-SELECT ts_lexize('english_hunspell', 'lankiness''s');
-SELECT ts_lexize('english_hunspell', 'rewritten');
+CREATE TABLE table1(name varchar);
+INSERT INTO table1 VALUES ('leaves'), ('leaved'), ('leaving'),
+						('inability'), ('abilities'), ('disability'), ('ability');
+
+SELECT d.* FROM table1 AS t, LATERAL ts_debug('english_hunspell', t.name) AS d;
+
+CREATE INDEX name_idx ON table1 USING GIN (to_tsvector('english_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'leaving');
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'abilities');
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'ability');
+
+DROP INDEX name_idx;
+CREATE INDEX name_idx ON table1 USING GIST (to_tsvector('english_hunspell', "name"));
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'leaving');
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'abilities');
+SELECT * FROM table1 WHERE to_tsvector('english_hunspell', name)
+	@@ to_tsquery('english_hunspell', 'ability');