I noticed that hash_array_extended() does not pass down the collation to
the element's collation function, unlike hash_array(). As a
consequence, hash partitioning using text arrays as partition key fails.
The attached patch fixes this. I propose to backpatch this.
From 171c58bc37c211cab11f6b588220e9e773f67e67 Mon Sep 17 00:00:00 2001
From: Peter Eisentraut <pe...@eisentraut.org>
Date: Mon, 2 Nov 2020 08:24:28 +0100
Subject: [PATCH] Enable hash partitioning of text arrays
hash_array_extended() needs to pass PG_GET_COLLATION() to the hash
function of the element type. Otherwise, the hash function of a
collation-aware data type such as text will error out, since the
introduction of nondeterministic collation made hash functions require
a collation, too.
The consequence of this is that before this change, hash partitioning
using an array over text in the partition key would not work.
---
src/backend/utils/adt/arrayfuncs.c | 2 +-
.../regress/expected/collate.icu.utf8.out | 50 +++++++++++++++++++
src/test/regress/sql/collate.icu.utf8.sql | 26 ++++++++++
3 files changed, 77 insertions(+), 1 deletion(-)
diff --git a/src/backend/utils/adt/arrayfuncs.c
b/src/backend/utils/adt/arrayfuncs.c
index 392445ea03..a7ea7656c7 100644
--- a/src/backend/utils/adt/arrayfuncs.c
+++ b/src/backend/utils/adt/arrayfuncs.c
@@ -4071,7 +4071,7 @@ hash_array_extended(PG_FUNCTION_ARGS)
typalign = typentry->typalign;
InitFunctionCallInfoData(*locfcinfo,
&typentry->hash_extended_proc_finfo, 2,
- InvalidOid, NULL,
NULL);
+ PG_GET_COLLATION(),
NULL, NULL);
/* Loop over source data */
nitems = ArrayGetNItems(ndims, dims);
diff --git a/src/test/regress/expected/collate.icu.utf8.out
b/src/test/regress/expected/collate.icu.utf8.out
index 2b86ce9028..70133df804 100644
--- a/src/test/regress/expected/collate.icu.utf8.out
+++ b/src/test/regress/expected/collate.icu.utf8.out
@@ -1215,6 +1215,30 @@ SELECT * FROM test6 WHERE b = 'äbc' COLLATE ctest_nondet;
2 | äbc
(2 rows)
+-- same with arrays
+CREATE TABLE test6a (a int, b text[]);
+INSERT INTO test6a VALUES (1, ARRAY[U&'\00E4bc']);
+INSERT INTO test6a VALUES (2, ARRAY[U&'\0061\0308bc']);
+SELECT * FROM test6a;
+ a | b
+---+-------
+ 1 | {äbc}
+ 2 | {äbc}
+(2 rows)
+
+SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_det;
+ a | b
+---+-------
+ 1 | {äbc}
+(1 row)
+
+SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
+ a | b
+---+-------
+ 1 | {äbc}
+ 2 | {äbc}
+(2 rows)
+
CREATE COLLATION case_sensitive (provider = icu, locale = '');
CREATE COLLATION case_insensitive (provider = icu, locale =
'@colStrength=secondary', deterministic = false);
SELECT 'abc' <= 'ABC' COLLATE case_sensitive, 'abc' >= 'ABC' COLLATE
case_sensitive;
@@ -1842,6 +1866,19 @@ SELECT (SELECT count(*) FROM test22_0) = (SELECT
count(*) FROM test22_1);
t
(1 row)
+-- same with arrays
+CREATE TABLE test22a (a int, b text[] COLLATE case_sensitive) PARTITION BY
HASH (b);
+CREATE TABLE test22a_0 PARTITION OF test22a FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
+CREATE TABLE test22a_1 PARTITION OF test22a FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
+INSERT INTO test22a VALUES (1, ARRAY['def']);
+INSERT INTO test22a VALUES (2, ARRAY['DEF']);
+-- they end up in different partitions
+SELECT (SELECT count(*) FROM test22a_0) = (SELECT count(*) FROM test22a_1);
+ ?column?
+----------
+ t
+(1 row)
+
CREATE TABLE test23 (a int, b text COLLATE case_insensitive) PARTITION BY HASH
(b);
CREATE TABLE test23_0 PARTITION OF test23 FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
CREATE TABLE test23_1 PARTITION OF test23 FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
@@ -1854,6 +1891,19 @@ SELECT (SELECT count(*) FROM test23_0) <> (SELECT
count(*) FROM test23_1);
t
(1 row)
+-- same with arrays
+CREATE TABLE test23a (a int, b text[] COLLATE case_insensitive) PARTITION BY
HASH (b);
+CREATE TABLE test23a_0 PARTITION OF test23a FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
+CREATE TABLE test23a_1 PARTITION OF test23a FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
+INSERT INTO test23a VALUES (1, ARRAY['def']);
+INSERT INTO test23a VALUES (2, ARRAY['DEF']);
+-- they end up in the same partition (but it's platform-dependent which one)
+SELECT (SELECT count(*) FROM test23a_0) <> (SELECT count(*) FROM test23a_1);
+ ?column?
+----------
+ t
+(1 row)
+
CREATE TABLE test30 (a int, b char(3) COLLATE case_insensitive) PARTITION BY
LIST (b);
CREATE TABLE test30_1 PARTITION OF test30 FOR VALUES IN ('abc');
INSERT INTO test30 VALUES (1, 'abc');
diff --git a/src/test/regress/sql/collate.icu.utf8.sql
b/src/test/regress/sql/collate.icu.utf8.sql
index 67de7d9794..9cee3d0042 100644
--- a/src/test/regress/sql/collate.icu.utf8.sql
+++ b/src/test/regress/sql/collate.icu.utf8.sql
@@ -483,6 +483,14 @@ CREATE TABLE test6 (a int, b text);
SELECT * FROM test6 WHERE b = 'äbc' COLLATE ctest_det;
SELECT * FROM test6 WHERE b = 'äbc' COLLATE ctest_nondet;
+-- same with arrays
+CREATE TABLE test6a (a int, b text[]);
+INSERT INTO test6a VALUES (1, ARRAY[U&'\00E4bc']);
+INSERT INTO test6a VALUES (2, ARRAY[U&'\0061\0308bc']);
+SELECT * FROM test6a;
+SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_det;
+SELECT * FROM test6a WHERE b = ARRAY['äbc'] COLLATE ctest_nondet;
+
CREATE COLLATION case_sensitive (provider = icu, locale = '');
CREATE COLLATION case_insensitive (provider = icu, locale =
'@colStrength=secondary', deterministic = false);
@@ -685,6 +693,15 @@ CREATE TABLE test22_1 PARTITION OF test22 FOR VALUES WITH
(MODULUS 2, REMAINDER
-- they end up in different partitions
SELECT (SELECT count(*) FROM test22_0) = (SELECT count(*) FROM test22_1);
+-- same with arrays
+CREATE TABLE test22a (a int, b text[] COLLATE case_sensitive) PARTITION BY
HASH (b);
+CREATE TABLE test22a_0 PARTITION OF test22a FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
+CREATE TABLE test22a_1 PARTITION OF test22a FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
+INSERT INTO test22a VALUES (1, ARRAY['def']);
+INSERT INTO test22a VALUES (2, ARRAY['DEF']);
+-- they end up in different partitions
+SELECT (SELECT count(*) FROM test22a_0) = (SELECT count(*) FROM test22a_1);
+
CREATE TABLE test23 (a int, b text COLLATE case_insensitive) PARTITION BY HASH
(b);
CREATE TABLE test23_0 PARTITION OF test23 FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
CREATE TABLE test23_1 PARTITION OF test23 FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
@@ -693,6 +710,15 @@ CREATE TABLE test23_1 PARTITION OF test23 FOR VALUES WITH
(MODULUS 2, REMAINDER
-- they end up in the same partition (but it's platform-dependent which one)
SELECT (SELECT count(*) FROM test23_0) <> (SELECT count(*) FROM test23_1);
+-- same with arrays
+CREATE TABLE test23a (a int, b text[] COLLATE case_insensitive) PARTITION BY
HASH (b);
+CREATE TABLE test23a_0 PARTITION OF test23a FOR VALUES WITH (MODULUS 2,
REMAINDER 0);
+CREATE TABLE test23a_1 PARTITION OF test23a FOR VALUES WITH (MODULUS 2,
REMAINDER 1);
+INSERT INTO test23a VALUES (1, ARRAY['def']);
+INSERT INTO test23a VALUES (2, ARRAY['DEF']);
+-- they end up in the same partition (but it's platform-dependent which one)
+SELECT (SELECT count(*) FROM test23a_0) <> (SELECT count(*) FROM test23a_1);
+
CREATE TABLE test30 (a int, b char(3) COLLATE case_insensitive) PARTITION BY
LIST (b);
CREATE TABLE test30_1 PARTITION OF test30 FOR VALUES IN ('abc');
INSERT INTO test30 VALUES (1, 'abc');
--
2.28.0
