GIN: Generalized Inverted iNdex.

text[], int4[], Tsearch2 support for GIN.

5 files changed, 243 insertions, 2 deletions

diff --git a/contrib/tsearch2/Makefile b/contrib/tsearch2/Makefile
index b7e4915ce7..3e322bb85c 100644
--- a/contrib/tsearch2/Makefile
+++ b/contrib/tsearch2/Makefile
@@ -1,4 +1,4 @@
-# $PostgreSQL: pgsql/contrib/tsearch2/Makefile,v 1.13 2006/01/27 16:32:31 teodor Exp $
+# $PostgreSQL: pgsql/contrib/tsearch2/Makefile,v 1.14 2006/05/02 11:28:54 teodor Exp $
 
 MODULE_big = tsearch2
 OBJS = dict_ex.o dict.o snmap.o stopword.o common.o prs_dcfg.o \
@@ -7,7 +7,7 @@ OBJS = dict_ex.o dict.o snmap.o stopword.o common.o prs_dcfg.o \
        ts_cfg.o tsvector.o query_cleanup.o crc32.o query.o gistidx.o \
        tsvector_op.o rank.o ts_stat.o \
        query_util.o query_support.o query_rewrite.o query_gist.o \
-       ts_locale.o
+       ts_locale.o ginidx.o
 
 SUBDIRS     := snowball ispell wordparser
 SUBDIROBJS  := $(SUBDIRS:%=%/SUBSYS.o)
diff --git a/contrib/tsearch2/expected/tsearch2.out b/contrib/tsearch2/expected/tsearch2.out
index 2a60e317cd..39a95b2f70 100644
--- a/contrib/tsearch2/expected/tsearch2.out
+++ b/contrib/tsearch2/expected/tsearch2.out
@@ -3001,3 +3001,42 @@ select a is null, a from test_tsvector order by a;
  t        |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            
 (514 rows)
 
+drop index wowidx;
+create index wowidx on test_tsvector using gin (a);
+set enable_seqscan=off;
+SELECT count(*) FROM test_tsvector WHERE a @@ 'wr|qh';
+ count 
+-------
+   158
+(1 row)
+
+SELECT count(*) FROM test_tsvector WHERE a @@ 'wr&qh';
+ count 
+-------
+    17
+(1 row)
+
+SELECT count(*) FROM test_tsvector WHERE a @@ 'eq&yt';
+ count 
+-------
+     6
+(1 row)
+
+SELECT count(*) FROM test_tsvector WHERE a @@ 'eq|yt';
+ count 
+-------
+    98
+(1 row)
+
+SELECT count(*) FROM test_tsvector WHERE a @@ '(eq&yt)|(wr&qh)';
+ count 
+-------
+    23
+(1 row)
+
+SELECT count(*) FROM test_tsvector WHERE a @@ '(eq|yt)&(wr|qh)';
+ count 
+-------
+    39
+(1 row)
+
diff --git a/contrib/tsearch2/ginidx.c b/contrib/tsearch2/ginidx.c
new file mode 100644
index 0000000000..0a4e04e64a
--- /dev/null
+++ b/contrib/tsearch2/ginidx.c
@@ -0,0 +1,145 @@
+#include "postgres.h"
+
+#include <float.h>
+
+#include "access/gist.h"
+#include "access/itup.h"
+#include "access/tuptoaster.h"
+#include "storage/bufpage.h"
+#include "utils/array.h"
+#include "utils/builtins.h"
+
+#include "tsvector.h"
+#include "query.h"
+#include "query_cleanup.h"
+
+PG_FUNCTION_INFO_V1(gin_extract_tsvector);
+Datum       gin_extract_tsvector(PG_FUNCTION_ARGS);
+
+Datum
+gin_extract_tsvector(PG_FUNCTION_ARGS) {
+	tsvector 	*vector = (tsvector *) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
+	uint32  	*nentries = (uint32*)PG_GETARG_POINTER(1);
+	Datum		*entries = NULL;
+
+	*nentries = 0;
+	if ( vector->size > 0 ) {
+		int i;
+		WordEntry	*we = ARRPTR( vector );
+
+		*nentries = (uint32)vector->size;
+		entries = (Datum*)palloc( sizeof(Datum) * vector->size );
+
+		for(i=0;i<vector->size;i++) {
+			text 	*txt = (text*)palloc( VARHDRSZ + we->len );
+
+			VARATT_SIZEP(txt) = VARHDRSZ + we->len; 
+			memcpy( VARDATA(txt), STRPTR( vector ) + we->pos, we->len ); 
+
+			entries[i] = PointerGetDatum( txt );
+
+			we++;
+		}
+	}
+
+	PG_FREE_IF_COPY(vector, 0);
+	PG_RETURN_POINTER(entries);
+}
+
+
+PG_FUNCTION_INFO_V1(gin_extract_tsquery);
+Datum       gin_extract_tsquery(PG_FUNCTION_ARGS);
+
+Datum
+gin_extract_tsquery(PG_FUNCTION_ARGS) {
+	QUERYTYPE *query = (QUERYTYPE*) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
+	uint32      *nentries = (uint32*)PG_GETARG_POINTER(1);
+	StrategyNumber	strategy = DatumGetUInt16( PG_GETARG_DATUM(2) ); 
+	Datum       *entries = NULL;
+
+	*nentries = 0;
+	if ( query->size > 0 ) {
+		int4 i, j=0, len;
+		ITEM *item;
+
+		item = clean_NOT_v2(GETQUERY(query), &len);
+		if ( !item )
+			elog(ERROR,"Query requires full scan, GIN doesn't support it");
+
+		item = GETQUERY(query);
+
+		for(i=0; i<query->size; i++)
+			if ( item[i].type == VAL )
+				(*nentries)++;
+
+		entries = (Datum*)palloc( sizeof(Datum) * (*nentries) );
+
+		for(i=0; i<query->size; i++)
+			if ( item[i].type == VAL ) {
+				text 	*txt;
+
+				txt = (text*)palloc( VARHDRSZ + item[i].length );
+
+				VARATT_SIZEP(txt) = VARHDRSZ + item[i].length; 
+				memcpy( VARDATA(txt), GETOPERAND( query ) + item[i].distance, item[i].length );
+
+				entries[j++] = PointerGetDatum( txt );
+
+				if ( strategy == 1 && item[i].weight != 0 )
+					elog(ERROR,"With class of lexeme restrictions use @@@ operation");
+			}
+
+	}
+
+	PG_FREE_IF_COPY(query, 0);
+	PG_RETURN_POINTER(entries);
+}
+
+typedef struct {
+	ITEM	*frst;
+	bool	*mapped_check;
+} GinChkVal; 
+
+static bool
+checkcondition_gin(void *checkval, ITEM * val) {
+	GinChkVal	*gcv = (GinChkVal*)checkval;
+
+	return gcv->mapped_check[ val - gcv->frst ];
+}
+
+PG_FUNCTION_INFO_V1(gin_ts_consistent);
+Datum       gin_ts_consistent(PG_FUNCTION_ARGS);
+
+Datum
+gin_ts_consistent(PG_FUNCTION_ARGS) {
+	bool    *check = (bool*)PG_GETARG_POINTER(0);
+	QUERYTYPE *query = (QUERYTYPE*) PG_DETOAST_DATUM(PG_GETARG_DATUM(2));
+	bool	res = FALSE;
+
+	if ( query->size > 0 ) {
+		int4 i, j=0;
+		ITEM *item;
+		GinChkVal	gcv;
+
+		gcv.frst = item = GETQUERY(query); 
+		gcv.mapped_check= (bool*)palloc( sizeof(bool) * query->size );
+
+		for(i=0; i<query->size; i++)
+			if ( item[i].type == VAL )
+				gcv.mapped_check[ i ] = check[ j++ ];
+
+
+		res = TS_execute(
+			GETQUERY(query),
+			&gcv,
+			true,
+			checkcondition_gin
+		);
+
+	}
+
+	PG_FREE_IF_COPY(query, 2);
+	PG_RETURN_BOOL(res);
+}
+
+
diff --git a/contrib/tsearch2/sql/tsearch2.sql b/contrib/tsearch2/sql/tsearch2.sql
index 82a62224df..c6c149c121 100644
--- a/contrib/tsearch2/sql/tsearch2.sql
+++ b/contrib/tsearch2/sql/tsearch2.sql
@@ -363,3 +363,14 @@ select * from ts_debug('Tsearch module for PostgreSQL 7.3.3');
 insert into test_tsvector values (null, null);
 select a is null, a from test_tsvector order by a;
 
+drop index wowidx;
+create index wowidx on test_tsvector using gin (a);
+set enable_seqscan=off;
+
+SELECT count(*) FROM test_tsvector WHERE a @@ 'wr|qh';
+SELECT count(*) FROM test_tsvector WHERE a @@ 'wr&qh';
+SELECT count(*) FROM test_tsvector WHERE a @@ 'eq&yt';
+SELECT count(*) FROM test_tsvector WHERE a @@ 'eq|yt';
+SELECT count(*) FROM test_tsvector WHERE a @@ '(eq&yt)|(wr&qh)';
+SELECT count(*) FROM test_tsvector WHERE a @@ '(eq|yt)&(wr|qh)';
+
diff --git a/contrib/tsearch2/tsearch.sql.in b/contrib/tsearch2/tsearch.sql.in
index 0ab0887b45..76b4c5bb9e 100644
--- a/contrib/tsearch2/tsearch.sql.in
+++ b/contrib/tsearch2/tsearch.sql.in
@@ -1146,8 +1146,54 @@ AS
         FUNCTION        7       gtsq_same (gtsq, gtsq, internal),
         STORAGE         gtsq;
 
+--GIN support function
+CREATE FUNCTION gin_extract_tsvector(tsvector,internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C RETURNS NULL ON NULL INPUT;
+
+CREATE FUNCTION gin_extract_tsquery(tsquery,internal,internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C RETURNS NULL ON NULL INPUT;
+
+CREATE FUNCTION gin_ts_consistent(internal,internal,tsquery)
+RETURNS bool
+AS 'MODULE_PATHNAME'
+LANGUAGE C RETURNS NULL ON NULL INPUT;
+
+CREATE OPERATOR @@@ (
+        LEFTARG = tsvector,
+        RIGHTARG = tsquery,
+        PROCEDURE = exectsq,
+        COMMUTATOR = '@@@',
+        RESTRICT = contsel,
+        JOIN = contjoinsel
+);
+CREATE OPERATOR @@@ (
+        LEFTARG = tsquery,
+        RIGHTARG = tsvector,
+        PROCEDURE = rexectsq,
+        COMMUTATOR = '@@@',
+        RESTRICT = contsel,
+        JOIN = contjoinsel
+);
+
+CREATE OPERATOR CLASS gin_tsvector_ops
+DEFAULT FOR TYPE tsvector USING gin
+AS
+        OPERATOR        1       @@ (tsvector, tsquery),
+        OPERATOR        2       @@@ (tsvector, tsquery) RECHECK,
+        FUNCTION        1       bttextcmp(text, text),
+        FUNCTION        2       gin_extract_tsvector(tsvector,internal),
+        FUNCTION        3       gin_extract_tsquery(tsquery,internal,internal),
+        FUNCTION        4       gin_ts_consistent(internal,internal,tsquery),
+        STORAGE         text;
+
+
 --example of ISpell dictionary
 --update pg_ts_dict set dict_initoption='DictFile="/usr/local/share/ispell/russian.dict" ,AffFile ="/usr/local/share/ispell/russian.aff", StopFile="/usr/local/share/ispell/russian.stop"' where dict_name='ispell_template';
 --example of synonym dict
 --update pg_ts_dict set dict_initoption='/usr/local/share/ispell/english.syn' where dict_id=5;
+
 END;