database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-11-18 02:02:55 +03:00
Code Activity

Rework output format of pg_dependencies

Browse Source 
The existing format of pg_dependencies uses a single-object JSON
structure, with each key value embedding all the knowledge about the
set attributes tracked, like:
{"1 => 5": 1.000000, "5 => 1": 0.423130}

While this is a very compact format, it is confusing to read and it is
difficult to manipulate the values within the object, particularly when
tracking multiple attributes.

The new output format introduced in this commit is a JSON array of
objects, with:
- A key named "degree", with a float value.
- A key named "attributes", with an array of attribute numbers.
- A key named "dependency", with an attribute number.

The values use the same underlying type as previously when printed, with
a new output format that shows now as follows:
[{"degree": 1.000000, "attributes": [1], "dependency": 5},
 {"degree": 0.423130, "attributes": [5], "dependency": 1}]

This new format will become handy for a follow-up set of changes, so as
it becomes possible to inject extended statistics rather than require an
ANALYZE, like in a dump/restore sequence or after pg_upgrade on a new
cluster.

This format has been suggested by Tomas Vondra.  The key names are
defined in the header introduced by 1f927cce44, to ease the
integration of frontend-specific changes that are still under
discussion.  (Again a personal note: if anybody comes up with better
name for the keys, of course feel free.)

The bulk of the changes come from the regression tests, where
jsonb_pretty() is now used to make the outputs generated easier to
parse.

Author: Corey Huinker <corey.huinker@gmail.com>
Reviewed-by: Jian He <jian.universality@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Discussion: https://postgr.es/m/CADkLM=dpz3KFnqP-dgJ-zvRvtjsa8UZv8wDAQdqho=qN3kX0Zg@mail.gmail.com
This commit is contained in:
Michael Paquier
2025-11-17 10:44:26 +09:00
parent 1f927cce44
commit e76defbcf0
5 changed files with 141 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
doc/src/sgml/perform.sgml
View File

@@ -1485,12 +1485,27 @@ CREATE STATISTICS stts (dependencies) ON city, zip FROM zipcodes;
ANALYZE zipcodes;
SELECT stxname, stxkeys, stxddependencies
FROM pg_statistic_ext join pg_statistic_ext_data on (oid = stxoid)
SELECT stxkeys AS k, jsonb_pretty(stxddependencies::text::jsonb) AS dep
FROM pg_statistic_ext JOIN pg_statistic_ext_data ON (oid = stxoid)
WHERE stxname = 'stts';
stxname | stxkeys | stxddependencies
---------+---------+------------------------------------------
stts | 1 5 | {"1 => 5": 1.000000, "5 => 1": 0.423130}
-[ RECORD 1 ]--------------------
k | 1 5
dep | [ +
| { +
| "degree": 1.000000,+
| "attributes": [ +
| 1 +
| ], +
| "dependency": 5 +
| }, +
| { +
| "degree": 0.423130,+
| "attributes": [ +
| 5 +
| ], +
| "dependency": 1 +
| } +
| ]
(1 row)
</programlisting>
Here it can be seen that column 1 (zip code) fully determines column

31
src/backend/utils/adt/pg_dependencies.c
View File

@@ -16,6 +16,7 @@
#include "lib/stringinfo.h"
#include "statistics/extended_stats_internal.h"
#include "statistics/statistics_format.h"
#include "utils/fmgrprotos.h"
/*
@@ -46,34 +47,34 @@ pg_dependencies_out(PG_FUNCTION_ARGS)
{
bytea *data = PG_GETARG_BYTEA_PP(0);
MVDependencies *dependencies = statext_dependencies_deserialize(data);
int i,
j;
StringInfoData str;
initStringInfo(&str);
appendStringInfoChar(&str, '{');
appendStringInfoChar(&str, '[');
for (i = 0; i < dependencies->ndeps; i++)
for (int i = 0; i < dependencies->ndeps; i++)
{
MVDependency *dependency = dependencies->deps[i];
if (i > 0)
appendStringInfoString(&str, ", ");
appendStringInfoChar(&str, '"');
for (j = 0; j < dependency->nattributes; j++)
{
if (j == dependency->nattributes - 1)
appendStringInfoString(&str, " => ");
else if (j > 0)
appendStringInfoString(&str, ", ");
if (dependency->nattributes <= 1)
elog(ERROR, "invalid zero-length nattributes array in MVDependencies");
appendStringInfo(&str, "%d", dependency->attributes[j]);
}
appendStringInfo(&str, "\": %f", dependency->degree);
appendStringInfo(&str, "{\"" PG_DEPENDENCIES_KEY_ATTRIBUTES "\": [%d",
dependency->attributes[0]);
for (int j = 1; j < dependency->nattributes - 1; j++)
appendStringInfo(&str, ", %d", dependency->attributes[j]);
appendStringInfo(&str, "], \"" PG_DEPENDENCIES_KEY_DEPENDENCY "\": %d, "
"\"" PG_DEPENDENCIES_KEY_DEGREE "\": %f}",
dependency->attributes[dependency->nattributes - 1],
dependency->degree);
}
appendStringInfoChar(&str, '}');
appendStringInfoChar(&str, ']');
PG_RETURN_CSTRING(str.data);
}

15
src/include/statistics/statistics_format.h
View File

@@ -29,4 +29,19 @@
#define PG_NDISTINCT_KEY_ATTRIBUTES "attributes"
#define PG_NDISTINCT_KEY_NDISTINCT "ndistinct"
/* ----------
* pg_dependencies in human-readable format is a JSON array made of elements
* with a predefined set of keys, like:
*
* [{"degree": 1.000000, "attributes": [3], "dependency": 4},
* {"degree": 1.000000, "attributes": [3], "dependency": 6},
* ... ]
* ----------
*/
#define PG_DEPENDENCIES_KEY_ATTRIBUTES "attributes"
#define PG_DEPENDENCIES_KEY_DEPENDENCY "dependency"
#define PG_DEPENDENCIES_KEY_DEGREE "degree"
#endif /* STATISTICS_FORMAT_H */

95
src/test/regress/expected/stats_ext.out
View File

@@ -196,7 +196,8 @@ Statistics objects:
"public.ab1_a_b_stats" ON a, b FROM ab1; STATISTICS 0
ANALYZE ab1;
SELECT stxname, jsonb_pretty(d.stxdndistinct::text::jsonb) AS stxdndistinct, stxddependencies, stxdmcv, stxdinherit
SELECT stxname, jsonb_pretty(d.stxdndistinct::text::jsonb) AS stxdndistinct,
jsonb_pretty(d.stxddependencies::text::jsonb) AS stxddependencies, stxdmcv, stxdinherit
FROM pg_statistic_ext s LEFT JOIN pg_statistic_ext_data d ON (d.stxoid = s.oid)
WHERE s.stxname = 'ab1_a_b_stats';
stxname | stxdndistinct | stxddependencies | stxdmcv | stxdinherit
@@ -1433,10 +1434,48 @@ SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE
CREATE STATISTICS func_deps_stat (dependencies) ON a, b, c FROM functional_dependencies;
ANALYZE functional_dependencies;
-- print the detected dependencies
SELECT dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
dependencies
------------------------------------------------------------------------------------------------------------
{"3 => 4": 1.000000, "3 => 6": 1.000000, "4 => 6": 1.000000, "3, 4 => 6": 1.000000, "3, 6 => 4": 1.000000}
SELECT jsonb_pretty(dependencies::text::jsonb) AS dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
dependencies
-----------------------------
[ +
{ +
"degree": 1.000000,+
"attributes": [ +
3 +
], +
"dependency": 4 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
3 +
], +
"dependency": 6 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
4 +
], +
"dependency": 6 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
3, +
4 +
], +
"dependency": 6 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
3, +
6 +
], +
"dependency": 4 +
} +
]
(1 row)
SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE a = 1 AND b = ''1''');
@@ -1775,10 +1814,48 @@ SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE
CREATE STATISTICS func_deps_stat (dependencies) ON (a * 2), upper(b), (c + 1) FROM functional_dependencies;
ANALYZE functional_dependencies;
-- print the detected dependencies
SELECT dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
dependencies
------------------------------------------------------------------------------------------------------------------------
{"-1 => -2": 1.000000, "-1 => -3": 1.000000, "-2 => -3": 1.000000, "-1, -2 => -3": 1.000000, "-1, -3 => -2": 1.000000}
SELECT jsonb_pretty(dependencies::text::jsonb) AS dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
dependencies
-----------------------------
[ +
{ +
"degree": 1.000000,+
"attributes": [ +
-1 +
], +
"dependency": -2 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
-1 +
], +
"dependency": -3 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
-2 +
], +
"dependency": -3 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
-1, +
-2 +
], +
"dependency": -3 +
}, +
{ +
"degree": 1.000000,+
"attributes": [ +
-1, +
-3 +
], +
"dependency": -2 +
} +
]
(1 row)
SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE (a * 2) = 2 AND upper(b) = ''1''');

7
src/test/regress/sql/stats_ext.sql
View File

@@ -125,7 +125,8 @@ ALTER TABLE ab1 ALTER a SET STATISTICS -1;
ALTER STATISTICS ab1_a_b_stats SET STATISTICS 0;
\d ab1
ANALYZE ab1;
SELECT stxname, jsonb_pretty(d.stxdndistinct::text::jsonb) AS stxdndistinct, stxddependencies, stxdmcv, stxdinherit
SELECT stxname, jsonb_pretty(d.stxdndistinct::text::jsonb) AS stxdndistinct,
jsonb_pretty(d.stxddependencies::text::jsonb) AS stxddependencies, stxdmcv, stxdinherit
FROM pg_statistic_ext s LEFT JOIN pg_statistic_ext_data d ON (d.stxoid = s.oid)
WHERE s.stxname = 'ab1_a_b_stats';
ALTER STATISTICS ab1_a_b_stats SET STATISTICS -1;
@@ -708,7 +709,7 @@ CREATE STATISTICS func_deps_stat (dependencies) ON a, b, c FROM functional_depen
ANALYZE functional_dependencies;
-- print the detected dependencies
SELECT dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
SELECT jsonb_pretty(dependencies::text::jsonb) AS dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE a = 1 AND b = ''1''');
@@ -844,7 +845,7 @@ CREATE STATISTICS func_deps_stat (dependencies) ON (a * 2), upper(b), (c + 1) FR
ANALYZE functional_dependencies;
-- print the detected dependencies
SELECT dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
SELECT jsonb_pretty(dependencies::text::jsonb) AS dependencies FROM pg_stats_ext WHERE statistics_name = 'func_deps_stat';
SELECT * FROM check_estimated_rows('SELECT * FROM functional_dependencies WHERE (a * 2) = 2 AND upper(b) = ''1''');