SQL2003-standard statistical aggregates, by Sergey Koposov. I've added only

the float8 versions of the aggregates, which is all that the standard requires.
Sergey's original patch also provided versions using numeric arithmetic,
but given the size and slowness of the code, I doubt we ought to include
those in core.

src/backend/utils/adt/int8.c

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/adt/int8.c,v 1.60 2006/03/05 15:58:42 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/adt/int8.c,v 1.61 2006/07/28 18:33:04 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -694,6 +694,28 @@ int8inc(PG_FUNCTION_ARGS)
 	}
 }
 
+/*
+ * These functions are exactly like int8inc but are used for aggregates that
+ * count only non-null values.  Since the functions are declared strict,
+ * the null checks happen before we ever get here, and all we need do is
+ * increment the state value.  We could actually make these pg_proc entries
+ * point right at int8inc, but then the opr_sanity regression test would
+ * complain about mismatched entries for a built-in function.
+ */
+
+Datum
+int8inc_any(PG_FUNCTION_ARGS)
+{
+	return int8inc(fcinfo);
+}
+
+Datum
+int8inc_float8_float8(PG_FUNCTION_ARGS)
+{
+	return int8inc(fcinfo);
+}
+
+
 Datum
 int8larger(PG_FUNCTION_ARGS)
 {

src/include/catalog/catversion.h

@@ -37,7 +37,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/catversion.h,v 1.342 2006/07/27 19:52:06 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/catversion.h,v 1.343 2006/07/28 18:33:04 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -53,6 +53,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	200607271
+#define CATALOG_VERSION_NO	200607281
 
 #endif

src/include/catalog/pg_aggregate.h

@@ -8,7 +8,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/pg_aggregate.h,v 1.56 2006/07/27 19:52:06 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/pg_aggregate.h,v 1.57 2006/07/28 18:33:04 tgl Exp $
  *
  * NOTES
  *	  the genbki.sh script reads this file and generates .bki
@@ -192,6 +192,20 @@ DATA(insert ( 2157	float4_accum	float8_stddev_samp	0	1022	"{0,0,0}" ));
 DATA(insert ( 2158	float8_accum	float8_stddev_samp	0	1022	"{0,0,0}" ));
 DATA(insert ( 2159	numeric_accum	numeric_stddev_samp	0	1231	"{0,0,0}" ));
 
+/* SQL2003 binary regression aggregates */
+DATA(insert ( 2818	int8inc_float8_float8		-				0	20		"0" ));
+DATA(insert ( 2819	float8_regr_accum	float8_regr_sxx			0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2820	float8_regr_accum	float8_regr_syy			0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2821	float8_regr_accum	float8_regr_sxy			0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2822	float8_regr_accum	float8_regr_avgx		0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2823	float8_regr_accum	float8_regr_avgy		0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2824	float8_regr_accum	float8_regr_r2			0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2825	float8_regr_accum	float8_regr_slope		0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2826	float8_regr_accum	float8_regr_intercept	0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2827	float8_regr_accum	float8_covar_pop		0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2828	float8_regr_accum	float8_covar_samp		0	1022	"{0,0,0,0,0,0}" ));
+DATA(insert ( 2829	float8_regr_accum	float8_corr				0	1022	"{0,0,0,0,0,0}" ));
+
 /* boolean-and and boolean-or */
 DATA(insert ( 2517	booland_statefunc	-			0	16		_null_ ));
 DATA(insert ( 2518	boolor_statefunc	-			0	16		_null_ ));

src/include/catalog/pg_proc.h

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/catalog/pg_proc.h,v 1.418 2006/07/27 19:52:06 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/catalog/pg_proc.h,v 1.419 2006/07/28 18:33:04 tgl Exp $
  *
  * NOTES
  *	  The script catalog/genbki.sh reads this file and generates .bki
@@ -1534,7 +1534,7 @@ DESCR("truncate interval to specified units");
 
 DATA(insert OID = 1219 (  int8inc		   PGNSP PGUID 12 f f t f i 1 20 "20" _null_ _null_ _null_	int8inc - _null_ ));
 DESCR("increment");
-DATA(insert OID = 2804 (  int8inc_any	   PGNSP PGUID 12 f f t f i 2 20 "20 2276" _null_ _null_ _null_	int8inc - _null_ ));
+DATA(insert OID = 2804 (  int8inc_any	   PGNSP PGUID 12 f f t f i 2 20 "20 2276" _null_ _null_ _null_	int8inc_any - _null_ ));
 DESCR("increment, ignores second argument");
 DATA(insert OID = 1230 (  int8abs		   PGNSP PGUID 12 f f t f i 1 20 "20" _null_ _null_ _null_	int8abs - _null_ ));
 DESCR("absolute value");
@@ -2730,6 +2730,32 @@ DATA(insert OID = 1963 (  int4_avg_accum   PGNSP PGUID 12 f f t f i 2 1016 "1016
 DESCR("AVG(int4) transition function");
 DATA(insert OID = 1964 (  int8_avg		   PGNSP PGUID 12 f f t f i 1 1700 "1016" _null_ _null_ _null_	int8_avg - _null_ ));
 DESCR("AVG(int) aggregate final function");
+DATA(insert OID = 2805 (  int8inc_float8_float8		PGNSP PGUID 12 f f t f i 3 20 "20 701 701" _null_ _null_ _null_ int8inc_float8_float8 - _null_ ));
+DESCR("REGR_COUNT(double, double) transition function");
+DATA(insert OID = 2806 (  float8_regr_accum			PGNSP PGUID 12 f f t f i 3 1022 "1022 701 701" _null_ _null_ _null_ float8_regr_accum - _null_ ));
+DESCR("REGR_...(double, double) transition function");
+DATA(insert OID = 2807 (  float8_regr_sxx			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_sxx - _null_ ));
+DESCR("REGR_SXX(double, double) aggregate final function");
+DATA(insert OID = 2808 (  float8_regr_syy			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_syy - _null_ ));
+DESCR("REGR_SYY(double, double) aggregate final function");
+DATA(insert OID = 2809 (  float8_regr_sxy			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_sxy - _null_ ));
+DESCR("REGR_SXY(double, double) aggregate final function");
+DATA(insert OID = 2810 (  float8_regr_avgx			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_avgx - _null_ ));
+DESCR("REGR_AVGX(double, double) aggregate final function");
+DATA(insert OID = 2811 (  float8_regr_avgy			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_avgy - _null_ ));
+DESCR("REGR_AVGY(double, double) aggregate final function");
+DATA(insert OID = 2812 (  float8_regr_r2			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_r2 - _null_ ));
+DESCR("REGR_R2(double, double) aggregate final function");
+DATA(insert OID = 2813 (  float8_regr_slope			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_slope - _null_ ));
+DESCR("REGR_SLOPE(double, double) aggregate final function");
+DATA(insert OID = 2814 (  float8_regr_intercept		PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_regr_intercept - _null_ ));
+DESCR("REGR_INTERCEPT(double, double) aggregate final function");
+DATA(insert OID = 2815 (  float8_covar_pop			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_covar_pop - _null_ ));
+DESCR("COVAR_POP(double, double) aggregate final function");
+DATA(insert OID = 2816 (  float8_covar_samp			PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_covar_samp - _null_ ));
+DESCR("COVAR_SAMP(double, double) aggregate final function");
+DATA(insert OID = 2817 (  float8_corr				PGNSP PGUID 12 f f t f i 1 701 "1022" _null_ _null_ _null_ float8_corr - _null_ ));
+DESCR("CORR(double, double) aggregate final function");
 
 /* To ASCII conversion */
 DATA(insert OID = 1845 ( to_ascii	PGNSP PGUID 12 f f t f i 1	25 "25" _null_ _null_ _null_	to_ascii_default - _null_ ));
@@ -3196,6 +3222,20 @@ DATA(insert OID = 2157 (  stddev			PGNSP PGUID 12 t f f f i 1 701 "700" _null_ _
 DATA(insert OID = 2158 (  stddev			PGNSP PGUID 12 t f f f i 1 701 "701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
 DATA(insert OID = 2159 (  stddev			PGNSP PGUID 12 t f f f i 1 1700 "1700" _null_ _null_ _null_ aggregate_dummy - _null_ ));
 
+DATA(insert OID = 2818 (  regr_count		PGNSP PGUID 12 t f f f i 2 20 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2819 (  regr_sxx			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2820 (  regr_syy			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2821 (  regr_sxy			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2822 (  regr_avgx			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2823 (  regr_avgy			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2824 (  regr_r2			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2825 (  regr_slope		PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2826 (  regr_intercept	PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+
+DATA(insert OID = 2827 (  covar_pop			PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2828 (  covar_samp		PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+DATA(insert OID = 2829 (  corr				PGNSP PGUID 12 t f f f i 2 701 "701 701" _null_ _null_ _null_  aggregate_dummy - _null_ ));
+
 DATA(insert OID = 2160 ( text_pattern_lt	 PGNSP PGUID 12 f f t f i 2 16 "25 25" _null_ _null_ _null_ text_pattern_lt - _null_ ));
 DATA(insert OID = 2161 ( text_pattern_le	 PGNSP PGUID 12 f f t f i 2 16 "25 25" _null_ _null_ _null_ text_pattern_le - _null_ ));
 DATA(insert OID = 2162 ( text_pattern_eq	 PGNSP PGUID 12 f f t f i 2 16 "25 25" _null_ _null_ _null_ text_pattern_eq - _null_ ));

src/include/utils/builtins.h

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/utils/builtins.h,v 1.280 2006/07/21 20:51:33 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/utils/builtins.h,v 1.281 2006/07/28 18:33:04 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -348,6 +348,18 @@ extern Datum float8_var_pop(PG_FUNCTION_ARGS);
 extern Datum float8_var_samp(PG_FUNCTION_ARGS);
 extern Datum float8_stddev_pop(PG_FUNCTION_ARGS);
 extern Datum float8_stddev_samp(PG_FUNCTION_ARGS);
+extern Datum float8_regr_accum(PG_FUNCTION_ARGS);
+extern Datum float8_regr_sxx(PG_FUNCTION_ARGS);
+extern Datum float8_regr_syy(PG_FUNCTION_ARGS);
+extern Datum float8_regr_sxy(PG_FUNCTION_ARGS);
+extern Datum float8_regr_avgx(PG_FUNCTION_ARGS);
+extern Datum float8_regr_avgy(PG_FUNCTION_ARGS);
+extern Datum float8_covar_pop(PG_FUNCTION_ARGS);
+extern Datum float8_covar_samp(PG_FUNCTION_ARGS);
+extern Datum float8_corr(PG_FUNCTION_ARGS);
+extern Datum float8_regr_r2(PG_FUNCTION_ARGS);
+extern Datum float8_regr_slope(PG_FUNCTION_ARGS);
+extern Datum float8_regr_intercept(PG_FUNCTION_ARGS);
 extern Datum float48pl(PG_FUNCTION_ARGS);
 extern Datum float48mi(PG_FUNCTION_ARGS);
 extern Datum float48mul(PG_FUNCTION_ARGS);

src/include/utils/int8.h

@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/utils/int8.h,v 1.44 2006/03/05 15:59:07 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/utils/int8.h,v 1.45 2006/07/28 18:33:04 tgl Exp $
  *
  * NOTES
  * These data types are supported on all 64-bit architectures, and may
@@ -74,6 +74,8 @@ extern Datum int8div(PG_FUNCTION_ARGS);
 extern Datum int8abs(PG_FUNCTION_ARGS);
 extern Datum int8mod(PG_FUNCTION_ARGS);
 extern Datum int8inc(PG_FUNCTION_ARGS);
+extern Datum int8inc_any(PG_FUNCTION_ARGS);
+extern Datum int8inc_float8_float8(PG_FUNCTION_ARGS);
 extern Datum int8larger(PG_FUNCTION_ARGS);
 extern Datum int8smaller(PG_FUNCTION_ARGS);
 

src/test/regress/expected/aggregates.out

@@ -137,6 +137,61 @@ SELECT stddev_pop(3.0::numeric), stddev_samp(4.0::numeric);
           0 |            
 (1 row)
 
+-- SQL2003 binary aggregates
+SELECT regr_count(b, a) FROM aggtest;
+ regr_count 
+------------
+          4
+(1 row)
+
+SELECT regr_sxx(b, a) FROM aggtest;
+ regr_sxx 
+----------
+     5099
+(1 row)
+
+SELECT regr_syy(b, a) FROM aggtest;
+     regr_syy     
+------------------
+ 68756.2156939293
+(1 row)
+
+SELECT regr_sxy(b, a) FROM aggtest;
+     regr_sxy     
+------------------
+ 2614.51582155004
+(1 row)
+
+SELECT regr_avgx(b, a), regr_avgy(b, a) FROM aggtest;
+ regr_avgx |    regr_avgy     
+-----------+------------------
+      49.5 | 107.943152273074
+(1 row)
+
+SELECT regr_r2(b, a) FROM aggtest;
+      regr_r2       
+--------------------
+ 0.0194977982031803
+(1 row)
+
+SELECT regr_slope(b, a), regr_intercept(b, a) FROM aggtest;
+    regr_slope     |  regr_intercept  
+-------------------+------------------
+ 0.512750700441271 | 82.5619926012309
+(1 row)
+
+SELECT covar_pop(b, a), covar_samp(b, a) FROM aggtest;
+    covar_pop    |    covar_samp    
+-----------------+------------------
+ 653.62895538751 | 871.505273850014
+(1 row)
+
+SELECT corr(b, a) FROM aggtest;
+       corr        
+-------------------
+ 0.139634516517873
+(1 row)
+
 SELECT count(four) AS cnt_1000 FROM onek;
  cnt_1000 
 ----------

src/test/regress/expected/opr_sanity.out

@@ -66,15 +66,15 @@ WHERE p1.oid != p2.oid AND
 -- of the same internal function (ie, matching prosrc fields).  It's OK to
 -- have several entries with different pronames for the same internal function,
 -- but conflicts in the number of arguments and other critical items should
--- be complained of.
--- Ignore aggregates, since they all use "aggregate_dummy".
--- As of 8.2, this finds int8inc and int8inc_any, which are OK.
+-- be complained of.  (We don't check data types here; see next query.)
+-- Note: ignore aggregate functions here, since they all point to the same
+-- dummy built-in function.
 SELECT p1.oid, p1.proname, p2.oid, p2.proname
 FROM pg_proc AS p1, pg_proc AS p2
 WHERE p1.oid < p2.oid AND
     p1.prosrc = p2.prosrc AND
     p1.prolang = 12 AND p2.prolang = 12 AND
-    p1.proisagg = false AND p2.proisagg = false AND
+    (p1.proisagg = false OR p2.proisagg = false) AND
     (p1.prolang != p2.prolang OR
      p1.proisagg != p2.proisagg OR
      p1.prosecdef != p2.prosecdef OR
@@ -83,9 +83,8 @@ WHERE p1.oid < p2.oid AND
      p1.provolatile != p2.provolatile OR
      p1.pronargs != p2.pronargs);
  oid | proname | oid | proname 
-------+---------+------+-------------
- 1219 | int8inc | 2804 | int8inc_any
-(1 row)
+-----+---------+-----+---------
+(0 rows)
 
 -- Look for uses of different type OIDs in the argument/result type fields
 -- for different aliases of the same built-in function.

src/test/regress/sql/aggregates.sql

@@ -39,6 +39,17 @@ SELECT var_samp(b::numeric) FROM aggtest;
 SELECT var_pop(1.0), var_samp(2.0);
 SELECT stddev_pop(3.0::numeric), stddev_samp(4.0::numeric);
 
+-- SQL2003 binary aggregates
+SELECT regr_count(b, a) FROM aggtest;
+SELECT regr_sxx(b, a) FROM aggtest;
+SELECT regr_syy(b, a) FROM aggtest;
+SELECT regr_sxy(b, a) FROM aggtest;
+SELECT regr_avgx(b, a), regr_avgy(b, a) FROM aggtest;
+SELECT regr_r2(b, a) FROM aggtest;
+SELECT regr_slope(b, a), regr_intercept(b, a) FROM aggtest;
+SELECT covar_pop(b, a), covar_samp(b, a) FROM aggtest;
+SELECT corr(b, a) FROM aggtest;
+
 SELECT count(four) AS cnt_1000 FROM onek;
 SELECT count(DISTINCT four) AS cnt_4 FROM onek;
 

src/test/regress/sql/opr_sanity.sql

@@ -68,17 +68,16 @@ WHERE p1.oid != p2.oid AND
 -- of the same internal function (ie, matching prosrc fields).  It's OK to
 -- have several entries with different pronames for the same internal function,
 -- but conflicts in the number of arguments and other critical items should
--- be complained of.
--- Ignore aggregates, since they all use "aggregate_dummy".
-
--- As of 8.2, this finds int8inc and int8inc_any, which are OK.
+-- be complained of.  (We don't check data types here; see next query.)
+-- Note: ignore aggregate functions here, since they all point to the same
+-- dummy built-in function.
 
 SELECT p1.oid, p1.proname, p2.oid, p2.proname
 FROM pg_proc AS p1, pg_proc AS p2
 WHERE p1.oid < p2.oid AND
     p1.prosrc = p2.prosrc AND
     p1.prolang = 12 AND p2.prolang = 12 AND
-    p1.proisagg = false AND p2.proisagg = false AND
+    (p1.proisagg = false OR p2.proisagg = false) AND
     (p1.prolang != p2.prolang OR
      p1.proisagg != p2.proisagg OR
      p1.prosecdef != p2.prosecdef OR