1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
|
#
# MDEV-30660 COUNT DISTINCT seems unnecessarily slow when run on a PK
#
set @save_optimizer_trace = @@optimizer_trace;
SET optimizer_trace='enabled=on';
CREATE TABLE t1 (a INT NOT NULL PRIMARY KEY, b INT NOT NULL);
INSERT INTO t1 VALUES (1,1), (2,1), (3,1);
# Optimization is applied (aggregator=simple):
SELECT COUNT(DISTINCT a) FROM t1;
COUNT(DISTINCT a)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a)",
"aggregator_type": "simple"
}
]
SELECT AVG(DISTINCT a), SUM(DISTINCT b) FROM t1;
AVG(DISTINCT a) SUM(DISTINCT b)
2.0000 1
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "avg(distinct t1.a)",
"aggregator_type": "simple"
},
{
"function": "sum(distinct t1.b)",
"aggregator_type": "distinct"
}
]
# Only `a` is unique but it's enough to eliminate DISTINCT:
SELECT COUNT(DISTINCT b, a) FROM t1;
COUNT(DISTINCT b, a)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.b,t1.a)",
"aggregator_type": "simple"
}
]
SELECT COUNT(DISTINCT a, a + b) FROM t1;
COUNT(DISTINCT a, a + b)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a,t1.a + t1.b)",
"aggregator_type": "simple"
}
]
SELECT SUM(DISTINCT a), AVG(DISTINCT a), COUNT(DISTINCT a) FROM t1 WHERE a > 1;
SUM(DISTINCT a) AVG(DISTINCT a) COUNT(DISTINCT a)
5 2.5000 2
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "sum(distinct t1.a)",
"aggregator_type": "simple"
},
{
"function": "avg(distinct t1.a)",
"aggregator_type": "simple"
},
{
"function": "count(distinct t1.a)",
"aggregator_type": "simple"
}
]
# Optimization is not applied 'cause function argument is not a field
# (aggregator=distinct):
SELECT SUM(DISTINCT a + b) FROM t1;
SUM(DISTINCT a + b)
9
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "sum(distinct t1.a + t1.b)",
"aggregator_type": "distinct"
}
]
SELECT COUNT(DISTINCT b) FROM t1;
COUNT(DISTINCT b)
1
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.b)",
"aggregator_type": "distinct"
}
]
SELECT AVG(DISTINCT b / a) FROM t1;
AVG(DISTINCT b / a)
0.61110000
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "avg(distinct t1.b / t1.a)",
"aggregator_type": "distinct"
}
]
EXPLAIN SELECT COUNT(DISTINCT (SELECT a)) FROM t1;
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY t1 index NULL PRIMARY 4 NULL 3 Using index
2 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL No tables used
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct (/* select#2 */ select t1.a))",
"aggregator_type": "distinct"
}
]
CREATE TABLE t2 (a INT);
INSERT INTO t2 VALUES (1), (2);
# Optimization is not applied 'cause there is more than one table
SELECT COUNT(DISTINCT t1.a) FROM t1, t2;
COUNT(DISTINCT t1.a)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a)",
"aggregator_type": "distinct"
}
]
SELECT AVG(DISTINCT t1.a) FROM t1, t2;
AVG(DISTINCT t1.a)
2.0000
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "avg(distinct t1.a)",
"aggregator_type": "distinct"
}
]
# Const tables, optimization is applied
SELECT COUNT(DISTINCT a) FROM t1, (SELECT 1) AS t2;
COUNT(DISTINCT a)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a)",
"aggregator_type": "simple"
}
]
SELECT AVG(DISTINCT t1.a) FROM (SELECT 1 AS a) AS t2, t1, (SELECT 2 AS a) AS t3;
AVG(DISTINCT t1.a)
2.0000
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "avg(distinct t1.a)",
"aggregator_type": "simple"
}
]
SELECT COUNT(DISTINCT a) FROM t1, (SELECT 1 UNION SELECT 2) AS t2;
COUNT(DISTINCT a)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a)",
"aggregator_type": "distinct"
}
]
# Unique index on two columns
CREATE TABLE t3 (a INT NOT NULL, b INT NOT NULL);
INSERT INTO t3 VALUES (1,1), (1,2), (1,3), (2,1), (2,2), (3,1), (3,2);
CREATE UNIQUE INDEX t3_a_b ON t3 (a, b);
# Optimization is applied:
SELECT COUNT(DISTINCT a, b) FROM t3;
COUNT(DISTINCT a, b)
7
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t3.a,t3.b)",
"aggregator_type": "simple"
}
]
SELECT COUNT(DISTINCT b, a) FROM t3;
COUNT(DISTINCT b, a)
7
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t3.b,t3.a)",
"aggregator_type": "simple"
}
]
SELECT COUNT(DISTINCT b, a) FROM t3 WHERE a < 3;
COUNT(DISTINCT b, a)
5
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t3.b,t3.a)",
"aggregator_type": "simple"
}
]
# Optimization is applied to one of the functions:
SELECT COUNT(DISTINCT b), SUM(DISTINCT a), SUM(DISTINCT a + b) FROM t3 GROUP BY a;
COUNT(DISTINCT b) SUM(DISTINCT a) SUM(DISTINCT a + b)
3 1 9
2 2 7
2 3 9
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t3.b)",
"aggregator_type": "simple"
},
{
"function": "sum(distinct t3.a)",
"aggregator_type": "distinct"
},
{
"function": "sum(distinct t3.a + t3.b)",
"aggregator_type": "distinct"
}
]
# Can't apply optimization 'cause GROUP BY argument is not a field:
SELECT COUNT(DISTINCT b) FROM t3 GROUP BY a+b;
COUNT(DISTINCT b)
1
2
3
1
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t3.b)",
"aggregator_type": "distinct"
}
]
# Test merged view
CREATE VIEW v1 AS SELECT * FROM t1;
# Optimization is applied
SELECT COUNT(DISTINCT a, b) FROM v1;
COUNT(DISTINCT a, b)
3
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "count(distinct t1.a,t1.b)",
"aggregator_type": "simple"
}
]
# GROUP_CONCAT implements non-standard distinct aggregator
SELECT GROUP_CONCAT(b) FROM t1;
GROUP_CONCAT(b)
1,1,1
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "group_concat(t1.b separator ',')",
"aggregator_type": "simple"
}
]
SELECT GROUP_CONCAT(DISTINCT b) FROM t1;
GROUP_CONCAT(DISTINCT b)
1
SELECT JSON_DETAILED(JSON_EXTRACT(trace, '$**.prepare_sum_aggregators')) AS JS
FROM INFORMATION_SCHEMA.OPTIMIZER_TRACE;
JS
[
{
"function": "group_concat(distinct t1.b separator ',')",
"aggregator_type": "distinct"
}
]
DROP TABLE t1, t2, t3;
DROP VIEW v1;
SET optimizer_trace = @save_optimizer_trace;
#
# end of 10.5 tests
#
|
