external
/
citus
mirror of https://github.com/citusdata/citus.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

try filtering diff

naisila/try_filtering_diff
naisila 2024-11-18 00:58:52 +03:00
parent 1cf690f377
commit ef2d026112
5 changed files with 23 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/test/regress/bin/normalize.sed
View File

@ -309,3 +309,9 @@ s/permission denied to terminate process/must be a superuser to terminate superu
s/permission denied to cancel query/must be a superuser to cancel superuser query/g
#endif /* PG_VERSION_NUM < PG_VERSION_16 */
# Adjusted subplan generation in local_table_join test due to PostgreSQL 17 changes
# ensuring consistency in the subquery condition format
# https://github.com/postgres/postgres/commit/b262ad44
# identify problematic line with "table_diff_filtering" name and change the filter to "WHERE true"
s/(.*) table_diff_filtering WHERE (.*)/\1 table_diff_filtering WHERE true/g

8
src/test/regress/expected/local_dist_join_mixed.out
View File

@ -1234,7 +1234,7 @@ DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT local.id, dis
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo3,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo2,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo10,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo1
(SELECT table_diff_filtering.id, table_diff_filtering.title FROM local table_diff_filtering, distributed WHERE table_diff_filtering.id = distributed.id ) as foo1
WHERE
foo1.id = foo9.id AND
foo1.id = foo8.id AND
@ -1265,9 +1265,9 @@ DEBUG: Wrapping relation "local" to a subquery
DEBUG: generating subplan XXX_8 for subquery SELECT id FROM local_dist_join_mixed.local WHERE true
DEBUG: Wrapping relation "local" to a subquery
DEBUG: generating subplan XXX_9 for subquery SELECT id FROM local_dist_join_mixed.local WHERE true
DEBUG: Wrapping relation "local" to a subquery
DEBUG: generating subplan XXX_10 for subquery SELECT id FROM local_dist_join_mixed.local WHERE (id IS NOT NULL)
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT foo1.id FROM (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo9, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_2'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo8, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_3'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo7, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_4'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo6, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_5'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo5, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_6'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo4, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_7'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo3, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_8'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo2, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_9'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo10, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_10'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo1 WHERE ((foo1.id OPERATOR(pg_catalog.=) foo9.id) AND (foo1.id OPERATOR(pg_catalog.=) foo8.id) AND (foo1.id OPERATOR(pg_catalog.=) foo7.id) AND (foo1.id OPERATOR(pg_catalog.=) foo6.id) AND (foo1.id OPERATOR(pg_catalog.=) foo5.id) AND (foo1.id OPERATOR(pg_catalog.=) foo4.id) AND (foo1.id OPERATOR(pg_catalog.=) foo3.id) AND (foo1.id OPERATOR(pg_catalog.=) foo2.id) AND (foo1.id OPERATOR(pg_catalog.=) foo10.id) AND (foo1.id OPERATOR(pg_catalog.=) foo1.id)) ORDER BY foo1.id
DEBUG: Wrapping relation "local" "table_diff_filtering" to a subquery
DEBUG: generating subplan XXX_10 for subquery SELECT id FROM local_dist_join_mixed.local table_diff_filtering WHERE true
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT foo1.id FROM (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo9, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_2'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo8, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_3'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo7, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_4'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo6, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_5'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo5, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_6'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo4, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_7'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo3, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_8'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo2, (SELECT local.id, local.title FROM (SELECT NULL::integer AS "dummy-1", local_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_9'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) local_1) local, local_dist_join_mixed.distributed WHERE (local.id OPERATOR(pg_catalog.=) distributed.id)) foo10, (SELECT table_diff_filtering.id, table_diff_filtering.title FROM (SELECT NULL::integer AS "dummy-1", table_diff_filtering_1.id, NULL::integer AS "dummy-3", NULL::text AS title, NULL::integer AS "dummy-5" FROM (SELECT intermediate_result.id FROM read_intermediate_result('XXX_10'::text, 'binary'::citus_copy_format) intermediate_result(id bigint)) table_diff_filtering_1) table_diff_filtering, local_dist_join_mixed.distributed WHERE (table_diff_filtering.id OPERATOR(pg_catalog.=) distributed.id)) foo1 WHERE ((foo1.id OPERATOR(pg_catalog.=) foo9.id) AND (foo1.id OPERATOR(pg_catalog.=) foo8.id) AND (foo1.id OPERATOR(pg_catalog.=) foo7.id) AND (foo1.id OPERATOR(pg_catalog.=) foo6.id) AND (foo1.id OPERATOR(pg_catalog.=) foo5.id) AND (foo1.id OPERATOR(pg_catalog.=) foo4.id) AND (foo1.id OPERATOR(pg_catalog.=) foo3.id) AND (foo1.id OPERATOR(pg_catalog.=) foo2.id) AND (foo1.id OPERATOR(pg_catalog.=) foo10.id) AND (foo1.id OPERATOR(pg_catalog.=) foo1.id)) ORDER BY foo1.id
id
---------------------------------------------------------------------
0

18
src/test/regress/expected/local_table_join.out
View File

@ -1424,22 +1424,22 @@ DEBUG: generating subplan XXX_1 for subquery SELECT typdefault FROM local_table
ERROR: direct joins between distributed and local tables are not supported
HINT: Use CTE's or subqueries to select from local tables and use them in joins
-- correlated sublinks are not yet supported because of #4470, unless we convert not-correlated table
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table using(key)
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table table_diff_filtering using(key)
WHERE d1.key IN (SELECT key FROM distributed_table WHERE d1.key = key and key = 5);
DEBUG: Wrapping relation "postgres_table" to a subquery
DEBUG: generating subplan XXX_1 for subquery SELECT key FROM local_table_join.postgres_table WHERE true
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT count(*) AS count FROM (local_table_join.distributed_table d1 JOIN (SELECT postgres_table_1.key, NULL::text AS value, NULL::jsonb AS value_2 FROM (SELECT intermediate_result.key FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(key integer)) postgres_table_1) postgres_table USING (key)) WHERE (d1.key OPERATOR(pg_catalog.=) ANY (SELECT distributed_table.key FROM local_table_join.distributed_table WHERE ((d1.key OPERATOR(pg_catalog.=) distributed_table.key) AND (distributed_table.key OPERATOR(pg_catalog.=) 5))))
DEBUG: Wrapping relation "postgres_table" "table_diff_filtering" to a subquery
DEBUG: generating subplan XXX_1 for subquery SELECT key FROM local_table_join.postgres_table table_diff_filtering WHERE true
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT count(*) AS count FROM (local_table_join.distributed_table d1 JOIN (SELECT table_diff_filtering_1.key, NULL::text AS value, NULL::jsonb AS value_2 FROM (SELECT intermediate_result.key FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(key integer)) table_diff_filtering_1) table_diff_filtering USING (key)) WHERE (d1.key OPERATOR(pg_catalog.=) ANY (SELECT distributed_table.key FROM local_table_join.distributed_table WHERE ((d1.key OPERATOR(pg_catalog.=) distributed_table.key) AND (distributed_table.key OPERATOR(pg_catalog.=) 5))))
count
---------------------------------------------------------------------
100
(1 row)
set citus.local_table_join_policy to 'prefer-distributed';
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table using(key)
WHERE d1.key IN (SELECT key FROM distributed_table WHERE d1.key = key and key = 5);
DEBUG: Wrapping relation "distributed_table" "d1" to a subquery
DEBUG: generating subplan XXX_1 for subquery SELECT key FROM local_table_join.distributed_table d1 WHERE true
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT count(*) AS count FROM ((SELECT d1_1.key, NULL::text AS value, NULL::jsonb AS value_2 FROM (SELECT intermediate_result.key FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(key integer)) d1_1) d1 JOIN local_table_join.postgres_table USING (key)) WHERE (d1.key OPERATOR(pg_catalog.=) ANY (SELECT distributed_table.key FROM local_table_join.distributed_table WHERE ((d1.key OPERATOR(pg_catalog.=) distributed_table.key) AND (distributed_table.key OPERATOR(pg_catalog.=) 5))))
SELECT COUNT(*) FROM distributed_table table_diff_filtering JOIN postgres_table using(key)
WHERE table_diff_filtering.key IN (SELECT key FROM distributed_table WHERE table_diff_filtering.key = key and key = 5);
DEBUG: Wrapping relation "distributed_table" "table_diff_filtering" to a subquery
DEBUG: generating subplan XXX_1 for subquery SELECT key FROM local_table_join.distributed_table table_diff_filtering WHERE true
DEBUG: Plan XXX query after replacing subqueries and CTEs: SELECT count(*) AS count FROM ((SELECT table_diff_filtering_1.key, NULL::text AS value, NULL::jsonb AS value_2 FROM (SELECT intermediate_result.key FROM read_intermediate_result('XXX_1'::text, 'binary'::citus_copy_format) intermediate_result(key integer)) table_diff_filtering_1) table_diff_filtering JOIN local_table_join.postgres_table USING (key)) WHERE (table_diff_filtering.key OPERATOR(pg_catalog.=) ANY (SELECT distributed_table.key FROM local_table_join.distributed_table WHERE ((table_diff_filtering.key OPERATOR(pg_catalog.=) distributed_table.key) AND (distributed_table.key OPERATOR(pg_catalog.=) 5))))
ERROR: direct joins between distributed and local tables are not supported
HINT: Use CTE's or subqueries to select from local tables and use them in joins
set citus.local_table_join_policy to 'auto';

2
src/test/regress/sql/local_dist_join_mixed.sql
View File

@ -330,7 +330,7 @@ SELECT id, name FROM local LEFT JOIN distributed USING (id) ORDER BY 1 LIMIT 1;
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo3,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo2,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo10,
(SELECT local.id, local.title FROM local, distributed WHERE local.id = distributed.id ) as foo1
(SELECT table_diff_filtering.id, table_diff_filtering.title FROM local table_diff_filtering, distributed WHERE table_diff_filtering.id = distributed.id ) as foo1
WHERE
foo1.id = foo9.id AND
foo1.id = foo8.id AND

6
src/test/regress/sql/local_table_join.sql
View File

@ -396,12 +396,12 @@ select typdefault from (
) as subq_2;
-- correlated sublinks are not yet supported because of #4470, unless we convert not-correlated table
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table using(key)
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table table_diff_filtering using(key)
WHERE d1.key IN (SELECT key FROM distributed_table WHERE d1.key = key and key = 5);
set citus.local_table_join_policy to 'prefer-distributed';
SELECT COUNT(*) FROM distributed_table d1 JOIN postgres_table using(key)
WHERE d1.key IN (SELECT key FROM distributed_table WHERE d1.key = key and key = 5);
SELECT COUNT(*) FROM distributed_table table_diff_filtering JOIN postgres_table using(key)
WHERE table_diff_filtering.key IN (SELECT key FROM distributed_table WHERE table_diff_filtering.key = key and key = 5);
set citus.local_table_join_policy to 'auto';
-- Some more subqueries