database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-07-30 16:24:05 +03:00
Code Activity

MDEV-26849: JSON Histograms: point selectivity estimates are off

Browse Source 
.. for non-existent values.

Handle this special case.
This commit is contained in:
Sergei Petrunia
2021-10-22 19:43:19 +03:00
parent f3f78bed85
commit 05877df472
8 changed files with 142 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

102
mysql-test/main/statistics_json.result
View File

@ -4148,6 +4148,7 @@ t1 id 1 17384 0.0000 4.0000 14.0000 {
explain select * from t1;
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 229376
drop table t0;
drop table t1;
set analyze_sample_percentage=@save_analyze_sample_percentage;
set histogram_size=@save_histogram_size;
@ -7530,3 +7531,104 @@ select c from t1 where c > '1';
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 16 16.00 75.00 75.00 Using where
drop table t1;
#
# MDEV-26849: JSON Histograms: point selectivity estimates are off for non-existent values
#
#
create table t0(a int);
insert into t0 (a) values (0),(1),(2),(3),(4),(5),(6),(7),(8),(9);
create table t1(a int);
insert into t1 select 100*A.a from t0 A, t0 B, t0 C;
select a, count(*) from t1 group by a order by a;
a count(*)
0 100
100 100
200 100
300 100
400 100
500 100
600 100
700 100
800 100
900 100
set histogram_type=json_hb, histogram_size=default;
analyze table t1 persistent for all;
Table Op Msg_type Msg_text
test.t1 analyze status Engine-independent statistics collected
test.t1 analyze status OK
select * from mysql.column_stats where table_name='t1';
db_name table_name column_name min_value max_value nulls_ratio avg_length avg_frequency hist_size hist_type histogram
test t1 a 0 900 0.0000 4.0000 100.0000 10 JSON_HB {
"histogram_hb_v2": [
{
"start": "0",
"size": 0.1,
"ndv": 1
},
{
"start": "100",
"size": 0.1,
"ndv": 1
},
{
"start": "200",
"size": 0.1,
"ndv": 1
},
{
"start": "300",
"size": 0.1,
"ndv": 1
},
{
"start": "400",
"size": 0.1,
"ndv": 1
},
{
"start": "500",
"size": 0.1,
"ndv": 1
},
{
"start": "600",
"size": 0.1,
"ndv": 1
},
{
"start": "700",
"size": 0.1,
"ndv": 1
},
{
"start": "800",
"size": 0.1,
"ndv": 1
},
{
"start": "900",
"end": "900",
"size": 0.1,
"ndv": 1
}
]
}
analyze select * from t1 where a=0;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 10.00 10.00 Using where
analyze select * from t1 where a=50;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 0.10 0.00 Using where
analyze select * from t1 where a=70;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 0.10 0.00 Using where
analyze select * from t1 where a=100;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 10.00 10.00 Using where
analyze select * from t1 where a=150;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 0.10 0.00 Using where
analyze select * from t1 where a=200;
id select_type table type possible_keys key key_len ref rows r_rows filtered r_filtered Extra
1 SIMPLE t1 ALL NULL NULL NULL NULL 1000 1000.00 10.00 10.00 Using where
drop table t0,t1;