Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commite452184

Browse files
Fix CLUSTER tuplesorts on abbreviated expressions.
CLUSTER sort won't use the datum1 SortTuple field when clusteringagainst an index whose leading key is an expression. This makes itunsafe to use the abbreviated keys optimization, which was missed by thelogic that sets up SortSupport state. Affected tuplesorts output tuplesin a completely bogus order as a result (the wrong SortSupport basedcomparator was used for the leading attribute).This issue is similar to the bug fixed on the master branch by recentcommitcc58eec. But it's a far older issue, that dates back to theintroduction of the abbreviated keys optimization by commit4ea51cd.Backpatch to all supported versions.Author: Peter Geoghegan <pg@bowt.ie>Author: Thomas Munro <thomas.munro@gmail.com>Discussion:https://postgr.es/m/CA+hUKG+bA+bmwD36_oDxAoLrCwZjVtST2fqe=b4=qZcmU7u89A@mail.gmail.comBackpatch: 10-
1 parente346329 commite452184

File tree

3 files changed

+23
-1
lines changed

3 files changed

+23
-1
lines changed

‎src/backend/utils/sort/tuplesort.c

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -975,6 +975,7 @@ tuplesort_begin_cluster(TupleDesc tupDesc,
975975
{
976976
Tuplesortstate*state=tuplesort_begin_common(workMem,coordinate,
977977
randomAccess);
978+
AttrNumberleading;
978979
BTScanInsertindexScanKey;
979980
MemoryContextoldcontext;
980981
inti;
@@ -1007,6 +1008,7 @@ tuplesort_begin_cluster(TupleDesc tupDesc,
10071008
state->abbrevNext=10;
10081009

10091010
state->indexInfo=BuildIndexInfo(indexRel);
1011+
leading=state->indexInfo->ii_IndexAttrNumbers[0];
10101012

10111013
state->tupDesc=tupDesc;/* assume we need not copy tupDesc */
10121014

@@ -1045,7 +1047,7 @@ tuplesort_begin_cluster(TupleDesc tupDesc,
10451047
(scanKey->sk_flags&SK_BT_NULLS_FIRST)!=0;
10461048
sortKey->ssup_attno=scanKey->sk_attno;
10471049
/* Convey if abbreviation optimization is applicable in principle */
1048-
sortKey->abbreviate= (i==0);
1050+
sortKey->abbreviate= (i==0&&leading!=0);
10491051

10501052
AssertState(sortKey->ssup_attno!=0);
10511053

‎src/test/regress/expected/cluster.out

Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -511,6 +511,13 @@ SELECT * FROM clstr_expression WHERE -a = -3 ORDER BY -a, b;
511511
COMMIT;
512512
-- and after clustering on clstr_expression_minus_a
513513
CLUSTER clstr_expression USING clstr_expression_minus_a;
514+
WITH rows AS
515+
(SELECT ctid, lag(a) OVER (ORDER BY ctid) AS la, a FROM clstr_expression)
516+
SELECT * FROM rows WHERE la < a;
517+
ctid | la | a
518+
------+----+---
519+
(0 rows)
520+
514521
BEGIN;
515522
SET LOCAL enable_seqscan = false;
516523
EXPLAIN (COSTS OFF) SELECT * FROM clstr_expression WHERE upper(b) = 'PREFIX3';
@@ -545,6 +552,13 @@ SELECT * FROM clstr_expression WHERE -a = -3 ORDER BY -a, b;
545552
COMMIT;
546553
-- and after clustering on clstr_expression_upper_b
547554
CLUSTER clstr_expression USING clstr_expression_upper_b;
555+
WITH rows AS
556+
(SELECT ctid, lag(b) OVER (ORDER BY ctid) AS lb, b FROM clstr_expression)
557+
SELECT * FROM rows WHERE upper(lb) > upper(b);
558+
ctid | lb | b
559+
------+----+---
560+
(0 rows)
561+
548562
BEGIN;
549563
SET LOCAL enable_seqscan = false;
550564
EXPLAIN (COSTS OFF) SELECT * FROM clstr_expression WHERE upper(b) = 'PREFIX3';

‎src/test/regress/sql/cluster.sql

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -245,6 +245,9 @@ COMMIT;
245245

246246
-- and after clustering on clstr_expression_minus_a
247247
CLUSTER clstr_expression USING clstr_expression_minus_a;
248+
WITH rowsAS
249+
(SELECT ctid, lag(a) OVER (ORDER BY ctid)AS la, aFROM clstr_expression)
250+
SELECT*FROM rowsWHERE la< a;
248251
BEGIN;
249252
SET LOCAL enable_seqscan= false;
250253
EXPLAIN (COSTS OFF)SELECT*FROM clstr_expressionWHEREupper(b)='PREFIX3';
@@ -255,6 +258,9 @@ COMMIT;
255258

256259
-- and after clustering on clstr_expression_upper_b
257260
CLUSTER clstr_expression USING clstr_expression_upper_b;
261+
WITH rowsAS
262+
(SELECT ctid, lag(b) OVER (ORDER BY ctid)AS lb, bFROM clstr_expression)
263+
SELECT*FROM rowsWHEREupper(lb)>upper(b);
258264
BEGIN;
259265
SET LOCAL enable_seqscan= false;
260266
EXPLAIN (COSTS OFF)SELECT*FROM clstr_expressionWHEREupper(b)='PREFIX3';

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp