Implemented avg() window function.

It is based on the sum() function, thus much of the logic is shared.

Considering that there are 2 counters stored within the function, one
that handles the null value, while the other that handles the divisor
for the avg computation, it is possible to remove the counter from the
Item_sum_avg. I have not removed it in this patch as we may choose to
refactor the whole code into a separate class.
This remains to be dicussed.

1 files changed, 95 insertions, 0 deletions

diff --git a/mysql-test/r/win_avg.result b/mysql-test/r/win_avg.result
new file mode 100644
index 00000000000..7e539d933d8
--- /dev/null
+++ b/mysql-test/r/win_avg.result
@@ -0,0 +1,95 @@
+create table t1 (
+pk int primary key,
+a int,
+b int,
+c real
+);
+insert into t1 values
+(101 , 0, 10, 1.1),
+(102 , 0, 10, 2.1),
+(103 , 1, 10, 3.1),
+(104 , 1, 10, 4.1),
+(108 , 2, 10, 5.1),
+(105 , 2, 20, 6.1),
+(106 , 2, 20, 7.1),
+(107 , 2, 20, 8.15),
+(109 , 4, 20, 9.15),
+(110 , 4, 20, 10.15),
+(111 , 5, NULL, 11.15),
+(112 , 5, 1, 12.25),
+(113 , 5, NULL, 13.35),
+(114 , 5, NULL, 14.50),
+(115 , 5, NULL, 15.65),
+(116 , 6, 1, NULL),
+(117 , 6, 1, 10),
+(118 , 6, 1, 1.1),
+(119 , 6, 1, NULL),
+(120 , 6, 1, NULL),
+(121 , 6, 1, NULL),
+(122 , 6, 1, 2.2),
+(123 , 6, 1, 20.1),
+(124 , 6, 1, -10.4),
+(125 , 6, 1, NULL),
+(126 , 6, 1, NULL),
+(127 , 6, 1, NULL);
+select pk, a, b, avg(b) over (partition by a order by pk ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING)
+from t1;
+pk	a	b	avg(b) over (partition by a order by pk ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING)
+101	0	10	10.0000
+102	0	10	10.0000
+103	1	10	10.0000
+104	1	10	10.0000
+105	2	20	20.0000
+106	2	20	20.0000
+107	2	20	16.6667
+108	2	10	15.0000
+109	4	20	20.0000
+110	4	20	20.0000
+111	5	NULL	1.0000
+112	5	1	1.0000
+113	5	NULL	1.0000
+114	5	NULL	NULL
+115	5	NULL	NULL
+116	6	1	1.0000
+117	6	1	1.0000
+118	6	1	1.0000
+119	6	1	1.0000
+120	6	1	1.0000
+121	6	1	1.0000
+122	6	1	1.0000
+123	6	1	1.0000
+124	6	1	1.0000
+125	6	1	1.0000
+126	6	1	1.0000
+127	6	1	1.0000
+select pk, a, c, avg(c) over (partition by a order by pk ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING)
+from t1;
+pk	a	c	avg(c) over (partition by a order by pk ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING)
+101	0	1.1	1.6
+102	0	2.1	1.6
+103	1	3.1	3.5999999999999996
+104	1	4.1	3.5999999999999996
+105	2	6.1	6.6
+106	2	7.1	7.116666666666667
+107	2	8.15	6.783333333333334
+108	2	5.1	6.625000000000001
+109	4	9.15	9.65
+110	4	10.15	9.65
+111	5	11.15	11.7
+112	5	12.25	12.25
+113	5	13.35	13.366666666666667
+114	5	14.5	14.5
+115	5	15.65	15.075
+116	6	NULL	10
+117	6	10	5.55
+118	6	1.1	5.55
+119	6	NULL	1.0999999999999996
+120	6	NULL	NULL
+121	6	NULL	2.1999999999999997
+122	6	2.2	11.15
+123	6	20.1	3.966666666666667
+124	6	-10.4	4.85
+125	6	NULL	-10.400000000000002
+126	6	NULL	NULL
+127	6	NULL	NULL
+drop table t1;