txtidx datatype for full text indexing with GiST.

From Oleg Bartunov and Teodor Sigaev.

1 files changed, 248 insertions, 0 deletions

diff --git a/contrib/tsearch/tsearch.sql.in b/contrib/tsearch/tsearch.sql.in
new file mode 100644
index 0000000000..bf31e1fc76
--- /dev/null
+++ b/contrib/tsearch/tsearch.sql.in
@@ -0,0 +1,248 @@
+BEGIN TRANSACTION;
+
+-- TXTIDX type
+
+CREATE FUNCTION txtidx_in(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION txtidx_out(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE TYPE txtidx (
+internallength = -1,
+input = txtidx_in,
+output = txtidx_out
+);
+
+CREATE FUNCTION txt2txtidx(text)
+RETURNS txtidx
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION txtidxsize(txtidx)
+RETURNS int4
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+--QUERYTYPES
+--without morphology
+CREATE FUNCTION qtxt_in(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION qtxt_out(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE TYPE query_txt (
+internallength = -1,
+input = qtxt_in,
+output = qtxt_out
+);
+
+--with morphology
+CREATE FUNCTION mqtxt_in(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE TYPE mquery_txt (
+internallength = -1,
+input = mqtxt_in,
+output = qtxt_out
+);
+
+--only for debug
+CREATE FUNCTION querytree(query_txt)
+RETURNS text
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION querytree(mquery_txt)
+RETURNS text
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+--operations
+CREATE FUNCTION execqtxt(txtidx, query_txt) RETURNS bool
+        AS 'MODULE_PATHNAME' LANGUAGE 'c' with (isstrict);
+
+COMMENT ON FUNCTION execqtxt(txtidx, query_txt) IS 'boolean operation with text index';
+
+CREATE FUNCTION execqtxt(txtidx, mquery_txt) RETURNS bool
+        AS 'MODULE_PATHNAME' LANGUAGE 'c' with (isstrict);
+
+COMMENT ON FUNCTION execqtxt(txtidx, mquery_txt) IS 'boolean operation with text index';
+
+CREATE FUNCTION rexecqtxt(query_txt, txtidx) RETURNS bool
+        AS 'MODULE_PATHNAME' LANGUAGE 'c' with (isstrict);
+
+COMMENT ON FUNCTION rexecqtxt(query_txt, txtidx) IS 'boolean operation with text index';
+
+CREATE FUNCTION rexecqtxt(mquery_txt, txtidx) RETURNS bool
+        AS 'MODULE_PATHNAME' LANGUAGE 'c' with (isstrict);
+
+COMMENT ON FUNCTION rexecqtxt(mquery_txt, txtidx) IS 'boolean operation with text index';
+
+CREATE OPERATOR @@ (
+   LEFTARG = txtidx, RIGHTARG = query_txt, PROCEDURE = execqtxt,
+   COMMUTATOR = '~@', RESTRICT = contsel, JOIN = contjoinsel
+);
+
+CREATE OPERATOR ~@ (
+   LEFTARG = query_txt, RIGHTARG = txtidx, PROCEDURE = rexecqtxt,
+   COMMUTATOR = '@@', RESTRICT = contsel, JOIN = contjoinsel
+);
+
+CREATE OPERATOR ## (
+   LEFTARG = txtidx, RIGHTARG = mquery_txt, PROCEDURE = execqtxt,
+   COMMUTATOR = '~#', RESTRICT = contsel, JOIN = contjoinsel
+);
+
+CREATE OPERATOR ~# (
+   LEFTARG = mquery_txt, RIGHTARG = txtidx, PROCEDURE = rexecqtxt,
+   COMMUTATOR = '##', RESTRICT = contsel, JOIN = contjoinsel
+);
+
+--Trigger
+create function tsearch() returns opaque as
+        'MODULE_PATHNAME'
+        language 'C';
+
+--GiST
+--GiST key type
+CREATE FUNCTION gtxtidx_in(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION gtxtidx_out(opaque)
+RETURNS opaque
+AS 'MODULE_PATHNAME'
+LANGUAGE 'c' with (isstrict);
+
+CREATE TYPE gtxtidx (
+internallength = -1,
+input = gtxtidx_in,
+output = gtxtidx_out
+);
+
+CREATE FUNCTION gtxtidx_consistent(gtxtidx,opaque,int4) RETURNS bool
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+CREATE FUNCTION gtxtidx_compress(opaque) RETURNS opaque
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+CREATE FUNCTION gtxtidx_decompress(opaque) RETURNS opaque
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+CREATE FUNCTION gtxtidx_penalty(opaque,opaque,opaque) RETURNS opaque
+        AS 'MODULE_PATHNAME' LANGUAGE 'c' with (isstrict);
+
+CREATE FUNCTION gtxtidx_picksplit(opaque, opaque) RETURNS opaque
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+CREATE FUNCTION gtxtidx_union(bytea, opaque) RETURNS _int4
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+CREATE FUNCTION gtxtidx_same(gtxtidx, gtxtidx, opaque) RETURNS opaque
+        AS 'MODULE_PATHNAME' LANGUAGE 'c';
+
+INSERT INTO pg_opclass (opcamid, opcname, opcintype, opcdefault, opckeytype)
+    VALUES (
+        (SELECT oid FROM pg_am WHERE amname = 'gist'),
+        'gist_txtidx_ops',
+        (SELECT oid FROM pg_type WHERE typname = 'txtidx'),
+        true,
+        (SELECT oid FROM pg_type WHERE typname = 'gtxtidx'));
+
+SELECT o.oid AS opoid, o.oprname
+INTO TEMP TABLE txtidx_ops_tmp
+FROM pg_operator o, pg_type t, pg_type tq
+WHERE o.oprleft = t.oid and o.oprright=tq.oid
+   and t.typname = 'txtidx'
+   and ( tq.typname='query_txt' or tq.typname='mquery_txt' );
+
+INSERT INTO pg_amop (amopclaid, amopstrategy, amopreqcheck, amopopr)
+   SELECT opcl.oid, 1, false, c.opoid
+   FROM pg_opclass opcl, txtidx_ops_tmp c
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and c.oprname = '@@';
+
+INSERT INTO pg_amop (amopclaid, amopstrategy, amopreqcheck, amopopr)
+   SELECT opcl.oid, 2, false, c.opoid
+   FROM pg_opclass opcl, txtidx_ops_tmp c
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and c.oprname = '##';
+
+DROP TABLE txtidx_ops_tmp;
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 1, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_consistent';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 2, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_union';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 3, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_compress';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 4, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_decompress';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 5, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_penalty';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 6, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_picksplit';
+
+INSERT INTO pg_amproc (amopclaid, amprocnum, amproc)
+   SELECT opcl.oid, 7, pro.oid
+   FROM pg_opclass opcl, pg_proc pro
+   WHERE
+      opcamid = (SELECT oid FROM pg_am WHERE amname = 'gist')
+      and opcname = 'gist_txtidx_ops'
+      and proname = 'gtxtidx_same';
+
+
+
+END TRANSACTION;