Add uuid to the set of types supported by contrib/btree_gist.

Paul Jungwirth, reviewed and hacked on by Teodor Sigaev, Ildus
Kurbangaliev, Adam Brusselback, Chris Bandy, and myself.

Discussion: https://postgr.es/m/CA+renyUEE29=X01JXdz8_TQvo6n9=2XoEBBRnQ8rkLyr+kjPxQ@mail.gmail.com
Discussion: https://postgr.es/m/55F6EE82.8080209@sigaev.ru

contrib/btree_gist/Makefile

@@ -6,16 +6,16 @@ OBJS =  btree_gist.o btree_utils_num.o btree_utils_var.o btree_int2.o \
         btree_int4.o btree_int8.o btree_float4.o btree_float8.o btree_cash.o \
         btree_oid.o btree_ts.o btree_time.o btree_date.o btree_interval.o \
         btree_macaddr.o btree_inet.o btree_text.o btree_bytea.o btree_bit.o \
-        btree_numeric.o $(WIN32RES)
+        btree_numeric.o btree_uuid.o $(WIN32RES)
 
 EXTENSION = btree_gist
-DATA = btree_gist--1.2.sql btree_gist--1.1--1.2.sql btree_gist--1.0--1.1.sql \
-	btree_gist--unpackaged--1.0.sql
+DATA = btree_gist--unpackaged--1.0.sql btree_gist--1.0--1.1.sql \
+       btree_gist--1.1--1.2.sql btree_gist--1.2.sql btree_gist--1.2--1.3.sql
 PGFILEDESC = "btree_gist - B-tree equivalent GiST operator classes"
 
 REGRESS = init int2 int4 int8 float4 float8 cash oid timestamp timestamptz \
         time timetz date interval macaddr inet cidr text varchar char bytea \
-        bit varbit numeric not_equal
+        bit varbit numeric uuid not_equal
 
 SHLIB_LINK += $(filter -lm, $(LIBS))
 

contrib/btree_gist/btree_gist--1.2--1.3.sql

+/* contrib/btree_gist/btree_gist--1.2--1.3.sql */
+
+-- complain if script is sourced in psql, rather than via ALTER EXTENSION
+\echo Use "ALTER EXTENSION btree_gist UPDATE TO '1.3'" to load this file. \quit
+
+-- Add support for indexing UUID columns
+
+-- define the GiST support methods
+CREATE FUNCTION gbt_uuid_consistent(internal,uuid,int2,oid,internal)
+RETURNS bool
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_fetch(internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_compress(internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_penalty(internal,internal,internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_picksplit(internal, internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_union(internal, internal)
+RETURNS gbtreekey32
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+CREATE FUNCTION gbt_uuid_same(gbtreekey32, gbtreekey32, internal)
+RETURNS internal
+AS 'MODULE_PATHNAME'
+LANGUAGE C IMMUTABLE STRICT;
+
+-- Create the operator class
+CREATE OPERATOR CLASS gist_uuid_ops
+DEFAULT FOR TYPE uuid USING gist
+AS
+	OPERATOR	1	<   ,
+	OPERATOR	2	<=  ,
+	OPERATOR	3	=   ,
+	OPERATOR	4	>=  ,
+	OPERATOR	5	>   ,
+	FUNCTION	1	gbt_uuid_consistent (internal, uuid, int2, oid, internal),
+	FUNCTION	2	gbt_uuid_union (internal, internal),
+	FUNCTION	3	gbt_uuid_compress (internal),
+	FUNCTION	4	gbt_decompress (internal),
+	FUNCTION	5	gbt_uuid_penalty (internal, internal, internal),
+	FUNCTION	6	gbt_uuid_picksplit (internal, internal),
+	FUNCTION	7	gbt_uuid_same (gbtreekey32, gbtreekey32, internal),
+	STORAGE		gbtreekey32;
+
+-- These are "loose" in the opfamily for consistency with the rest of btree_gist
+ALTER OPERATOR FAMILY gist_uuid_ops USING gist ADD
+	OPERATOR	6	<>  (uuid, uuid) ,
+	FUNCTION	9 (uuid, uuid) gbt_uuid_fetch (internal) ;

contrib/btree_gist/btree_gist.control

 # btree_gist extension
 comment = 'support for indexing common datatypes in GiST'
-default_version = '1.2'
+default_version = '1.3'
 module_pathname = '$libdir/btree_gist'
 relocatable = true

contrib/btree_gist/btree_gist.h

@@ -31,7 +31,8 @@ enum gbtree_type
 	gbt_t_bpchar,
 	gbt_t_bytea,
 	gbt_t_bit,
-	gbt_t_inet
+	gbt_t_inet,
+	gbt_t_uuid
 };
 
 #endif

contrib/btree_gist/btree_uuid.c

+/*
+ * contrib/btree_gist/btree_uuid.c
+ */
+#include "postgres.h"
+
+#include "btree_gist.h"
+#include "btree_utils_num.h"
+#include "port/pg_bswap.h"
+#include "utils/uuid.h"
+
+typedef struct
+{
+	pg_uuid_t	lower,
+				upper;
+} uuidKEY;
+
+
+/*
+ * UUID ops
+ */
+PG_FUNCTION_INFO_V1(gbt_uuid_compress);
+PG_FUNCTION_INFO_V1(gbt_uuid_fetch);
+PG_FUNCTION_INFO_V1(gbt_uuid_union);
+PG_FUNCTION_INFO_V1(gbt_uuid_picksplit);
+PG_FUNCTION_INFO_V1(gbt_uuid_consistent);
+PG_FUNCTION_INFO_V1(gbt_uuid_penalty);
+PG_FUNCTION_INFO_V1(gbt_uuid_same);
+
+
+static int
+uuid_internal_cmp(const pg_uuid_t *arg1, const pg_uuid_t *arg2)
+{
+	return memcmp(arg1->data, arg2->data, UUID_LEN);
+}
+
+static bool
+gbt_uuidgt(const void *a, const void *b)
+{
+	return uuid_internal_cmp((const pg_uuid_t *) a, (const pg_uuid_t *) b) > 0;
+}
+
+static bool
+gbt_uuidge(const void *a, const void *b)
+{
+	return uuid_internal_cmp((const pg_uuid_t *) a, (const pg_uuid_t *) b) >= 0;
+}
+
+static bool
+gbt_uuideq(const void *a, const void *b)
+{
+	return uuid_internal_cmp((const pg_uuid_t *) a, (const pg_uuid_t *) b) == 0;
+}
+
+static bool
+gbt_uuidle(const void *a, const void *b)
+{
+	return uuid_internal_cmp((const pg_uuid_t *) a, (const pg_uuid_t *) b) <= 0;
+}
+
+static bool
+gbt_uuidlt(const void *a, const void *b)
+{
+	return uuid_internal_cmp((const pg_uuid_t *) a, (const pg_uuid_t *) b) < 0;
+}
+
+static int
+gbt_uuidkey_cmp(const void *a, const void *b)
+{
+	uuidKEY    *ia = (uuidKEY *) (((const Nsrt *) a)->t);
+	uuidKEY    *ib = (uuidKEY *) (((const Nsrt *) b)->t);
+	int			res;
+
+	res = uuid_internal_cmp(&ia->lower, &ib->lower);
+	if (res == 0)
+		res = uuid_internal_cmp(&ia->upper, &ib->upper);
+	return res;
+}
+
+
+static const gbtree_ninfo tinfo =
+{
+	gbt_t_uuid,
+	UUID_LEN,
+	32,							/* sizeof(gbtreekey32) */
+	gbt_uuidgt,
+	gbt_uuidge,
+	gbt_uuideq,
+	gbt_uuidle,
+	gbt_uuidlt,
+	gbt_uuidkey_cmp,
+	NULL
+};
+
+
+/**************************************************
+ * uuid ops
+ **************************************************/
+
+
+Datum
+gbt_uuid_compress(PG_FUNCTION_ARGS)
+{
+	GISTENTRY  *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
+	GISTENTRY  *retval;
+
+	if (entry->leafkey)
+	{
+		char	   *r = (char *) palloc(2 * UUID_LEN);
+		pg_uuid_t  *key = DatumGetUUIDP(entry->key);
+
+		retval = palloc(sizeof(GISTENTRY));
+
+		memcpy((void *) r, (void *) key, UUID_LEN);
+		memcpy((void *) (r + UUID_LEN), (void *) key, UUID_LEN);
+		gistentryinit(*retval, PointerGetDatum(r),
+					  entry->rel, entry->page,
+					  entry->offset, FALSE);
+	}
+	else
+		retval = entry;
+
+	PG_RETURN_POINTER(retval);
+}
+
+Datum
+gbt_uuid_fetch(PG_FUNCTION_ARGS)
+{
+	GISTENTRY  *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
+
+	PG_RETURN_POINTER(gbt_num_fetch(entry, &tinfo));
+}
+
+Datum
+gbt_uuid_consistent(PG_FUNCTION_ARGS)
+{
+	GISTENTRY  *entry = (GISTENTRY *) PG_GETARG_POINTER(0);
+	pg_uuid_t  *query = PG_GETARG_UUID_P(1);
+	StrategyNumber strategy = (StrategyNumber) PG_GETARG_UINT16(2);
+
+	/* Oid		subtype = PG_GETARG_OID(3); */
+	bool	   *recheck = (bool *) PG_GETARG_POINTER(4);
+	uuidKEY    *kkk = (uuidKEY *) DatumGetPointer(entry->key);
+	GBT_NUMKEY_R key;
+
+	/* All cases served by this function are exact */
+	*recheck = false;
+
+	key.lower = (GBT_NUMKEY *) &kkk->lower;
+	key.upper = (GBT_NUMKEY *) &kkk->upper;
+
+	PG_RETURN_BOOL(
+				   gbt_num_consistent(&key, (void *) query, &strategy,
+									  GIST_LEAF(entry), &tinfo)
+		);
+}
+
+Datum
+gbt_uuid_union(PG_FUNCTION_ARGS)
+{
+	GistEntryVector *entryvec = (GistEntryVector *) PG_GETARG_POINTER(0);
+	void	   *out = palloc(sizeof(uuidKEY));
+
+	*(int *) PG_GETARG_POINTER(1) = sizeof(uuidKEY);
+	PG_RETURN_POINTER(gbt_num_union((void *) out, entryvec, &tinfo));
+}
+
+/*
+ * Convert a uuid to a "double" value for estimating sizes of ranges.
+ */
+static double
+uuid_2_double(const pg_uuid_t *u)
+{
+	uint64		uu[2];
+	const double two64 = 18446744073709551616.0;		/* 2^64 */
+
+	/* Source data may not be suitably aligned, so copy */
+	memcpy(uu, u->data, UUID_LEN);
+
+	/*
+	 * uuid values should be considered as big-endian numbers, since that
+	 * corresponds to how memcmp will compare them.  On a little-endian
+	 * machine, byte-swap each half so we can use native uint64 arithmetic.
+	 */
+#ifndef WORDS_BIGENDIAN
+	uu[0] = BSWAP64(uu[0]);
+	uu[1] = BSWAP64(uu[1]);
+#endif
+
+	/*
+	 * 2^128 is about 3.4e38, which in theory could exceed the range of
+	 * "double" (POSIX only requires 1e37).  To avoid any risk of overflow,
+	 * put the decimal point between the two halves rather than treating the
+	 * uuid value as a 128-bit integer.
+	 */
+	return (double) uu[0] + (double) uu[1] / two64;
+}
+
+Datum
+gbt_uuid_penalty(PG_FUNCTION_ARGS)
+{
+	uuidKEY    *origentry = (uuidKEY *) DatumGetPointer(((GISTENTRY *) PG_GETARG_POINTER(0))->key);
+	uuidKEY    *newentry = (uuidKEY *) DatumGetPointer(((GISTENTRY *) PG_GETARG_POINTER(1))->key);
+	float	   *result = (float *) PG_GETARG_POINTER(2);
+	double		olower,
+				oupper,
+				nlower,
+				nupper;
+
+	olower = uuid_2_double(&origentry->lower);
+	oupper = uuid_2_double(&origentry->upper);
+	nlower = uuid_2_double(&newentry->lower);
+	nupper = uuid_2_double(&newentry->upper);
+
+	penalty_num(result, olower, oupper, nlower, nupper);
+
+	PG_RETURN_POINTER(result);
+}
+
+Datum
+gbt_uuid_picksplit(PG_FUNCTION_ARGS)
+{
+	PG_RETURN_POINTER(gbt_num_picksplit(
+									(GistEntryVector *) PG_GETARG_POINTER(0),
+									  (GIST_SPLITVEC *) PG_GETARG_POINTER(1),
+										&tinfo
+										));
+}
+
+Datum
+gbt_uuid_same(PG_FUNCTION_ARGS)
+{
+	uuidKEY    *b1 = (uuidKEY *) PG_GETARG_POINTER(0);
+	uuidKEY    *b2 = (uuidKEY *) PG_GETARG_POINTER(1);
+	bool	   *result = (bool *) PG_GETARG_POINTER(2);
+
+	*result = gbt_num_same((void *) b1, (void *) b2, &tinfo);
+	PG_RETURN_POINTER(result);
+}

contrib/btree_gist/expected/uuid.out

+-- uuid check
+CREATE TABLE uuidtmp (a uuid);
+\copy uuidtmp from 'data/uuid.data'
+SET enable_seqscan=on;
+SELECT count(*) FROM uuidtmp WHERE a <  '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+ count 
+-------
+   227
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a <= '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+ count 
+-------
+   228
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a  = '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+ count 
+-------
+     1
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a >= '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+ count 
+-------
+   376
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a >  '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+ count 
+-------
+   375
+(1 row)
+
+CREATE INDEX uuididx ON uuidtmp USING gist ( a );
+SET enable_seqscan=off;
+SELECT count(*) FROM uuidtmp WHERE a <  '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+ count 
+-------
+   227
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a <= '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+ count 
+-------
+   228
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a  = '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+ count 
+-------
+     1
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a >= '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+ count 
+-------
+   376
+(1 row)
+
+SELECT count(*) FROM uuidtmp WHERE a >  '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+ count 
+-------
+   375
+(1 row)
+

contrib/btree_gist/sql/uuid.sql

+-- uuid check
+
+CREATE TABLE uuidtmp (a uuid);
+
+\copy uuidtmp from 'data/uuid.data'
+
+SET enable_seqscan=on;
+
+SELECT count(*) FROM uuidtmp WHERE a <  '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+
+SELECT count(*) FROM uuidtmp WHERE a <= '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+
+SELECT count(*) FROM uuidtmp WHERE a  = '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+
+SELECT count(*) FROM uuidtmp WHERE a >= '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+
+SELECT count(*) FROM uuidtmp WHERE a >  '55e65ca2-4136-4a4b-ba78-cd3fe4678203';
+
+CREATE INDEX uuididx ON uuidtmp USING gist ( a );
+
+SET enable_seqscan=off;
+
+SELECT count(*) FROM uuidtmp WHERE a <  '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+
+SELECT count(*) FROM uuidtmp WHERE a <= '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+
+SELECT count(*) FROM uuidtmp WHERE a  = '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+
+SELECT count(*) FROM uuidtmp WHERE a >= '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;
+
+SELECT count(*) FROM uuidtmp WHERE a >  '55e65ca2-4136-4a4b-ba78-cd3fe4678203'::uuid;

doc/src/sgml/btree-gist.sgml

@@ -16,7 +16,8 @@
   <type>time without time zone</>, <type>date</>, <type>interval</>,
   <type>oid</>, <type>money</>, <type>char</>,
   <type>varchar</>, <type>text</>, <type>bytea</>, <type>bit</>,
-  <type>varbit</>, <type>macaddr</>, <type>inet</>, and <type>cidr</>.
+  <type>varbit</>, <type>macaddr</>, <type>inet</>, <type>cidr</>,
+  and <type>uuid</>.
  </para>
 
  <para>
@@ -99,8 +100,9 @@ INSERT 0 1
 
   <para>
    Teodor Sigaev (<email>teodor@stack.net</email>),
-   Oleg Bartunov (<email>oleg@sai.msu.su</email>), and
-   Janko Richter (<email>jankorichter@yahoo.de</email>).  See
+   Oleg Bartunov (<email>oleg@sai.msu.su</email>),
+   Janko Richter (<email>jankorichter@yahoo.de</email>), and
+   Paul Jungwirth (<email>pj@illuminatedcomputing.com</email>).  See
    <ulink url="http://www.sai.msu.su/~megera/postgres/gist/"></ulink>
    for additional information.
   </para>

src/backend/utils/adt/uuid.c

@@ -22,15 +22,6 @@
 #include "utils/sortsupport.h"
 #include "utils/uuid.h"
 
-/* uuid size in bytes */
-#define UUID_LEN 16
-
-/* pg_uuid_t is declared to be struct pg_uuid_t in uuid.h */
-struct pg_uuid_t
-{
-	unsigned char data[UUID_LEN];
-};
-
 /* sortsupport for uuid */
 typedef struct
 {

src/include/utils/uuid.h

@@ -14,11 +14,13 @@
 #ifndef UUID_H
 #define UUID_H
 
-/* guid size in bytes */
+/* uuid size in bytes */
 #define UUID_LEN 16
 
-/* opaque struct; defined in uuid.c */
-typedef struct pg_uuid_t pg_uuid_t;
+typedef struct pg_uuid_t
+{
+	unsigned char data[UUID_LEN];
+} pg_uuid_t;
 
 /* fmgr interface macros */
 #define UUIDPGetDatum(X)		PointerGetDatum(X)