mirror of
https://git.postgresql.org/git/postgresql.git
synced 2024-08-27 01:47:19 +02:00
The GUC settings lc_collate and lc_ctype are from a time when those locale settings were cluster-global. When those locale settings were made per-database (PG 8.4), the settings were kept as read-only. As of PG 15, you can use ICU as the per-database locale provider, so examining these settings is already less meaningful and possibly confusing, since you need to look into pg_database to find out what is really happening, and they would likely become fully obsolete in the future anyway. Reviewed-by: Jeff Davis <pgsql@j-davis.com> Discussion: https://www.postgresql.org/message-id/696054d1-bc88-b6ab-129a-18b8bce6a6f0@enterprisedb.com
154 lines
2.6 KiB
Plaintext
154 lines
2.6 KiB
Plaintext
/*
|
|
* This test must be run in a database with UTF-8 encoding
|
|
* and a Unicode-aware locale.
|
|
*
|
|
* Also disable this file for ICU, because the test for the the
|
|
* Turkish dotted I is not correct for many ICU locales. citext always
|
|
* uses the default collation, so it's not easy to restrict the test
|
|
* to the "tr-TR-x-icu" collation where it will succeed.
|
|
*/
|
|
SELECT getdatabaseencoding() <> 'UTF8' OR
|
|
(SELECT (datlocprovider = 'c' AND datctype = 'C') OR datlocprovider = 'i'
|
|
FROM pg_database
|
|
WHERE datname=current_database())
|
|
AS skip_test \gset
|
|
\if :skip_test
|
|
\quit
|
|
\endif
|
|
set client_encoding = utf8;
|
|
-- CREATE EXTENSION IF NOT EXISTS citext;
|
|
-- Multibyte sanity tests.
|
|
SELECT 'À'::citext = 'À'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'À'::citext = 'à'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'À'::text = 'à'::text AS f; -- text wins.
|
|
f
|
|
---
|
|
f
|
|
(1 row)
|
|
|
|
SELECT 'À'::citext <> 'B'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
-- Test combining characters making up canonically equivalent strings.
|
|
SELECT 'Ä'::text <> 'Ä'::text AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'Ä'::citext <> 'Ä'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
-- Test the Turkish dotted I. The lowercase is a single byte while the
|
|
-- uppercase is multibyte. This is why the comparison code can't be optimized
|
|
-- to compare string lengths.
|
|
SELECT 'i'::citext = 'İ'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
-- Regression.
|
|
SELECT 'láska'::citext <> 'laská'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'Ask Bjørn Hansen'::citext = 'Ask Bjørn Hansen'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'Ask Bjørn Hansen'::citext = 'ASK BJØRN HANSEN'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'Ask Bjørn Hansen'::citext <> 'Ask Bjorn Hansen'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT 'Ask Bjørn Hansen'::citext <> 'ASK BJORN HANSEN'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT citext_cmp('Ask Bjørn Hansen'::citext, 'Ask Bjørn Hansen'::citext) = 0 AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT citext_cmp('Ask Bjørn Hansen'::citext, 'ask bjørn hansen'::citext) = 0 AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT citext_cmp('Ask Bjørn Hansen'::citext, 'ASK BJØRN HANSEN'::citext) = 0 AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT citext_cmp('Ask Bjørn Hansen'::citext, 'Ask Bjorn Hansen'::citext) > 0 AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
SELECT citext_cmp('Ask Bjorn Hansen'::citext, 'Ask Bjørn Hansen'::citext) < 0 AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
-- Test ~<~ and ~<=~
|
|
SELECT 'à'::citext ~<~ 'À'::citext AS f;
|
|
f
|
|
---
|
|
f
|
|
(1 row)
|
|
|
|
SELECT 'à'::citext ~<=~ 'À'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|
|
-- Test ~>~ and ~>=~
|
|
SELECT 'à'::citext ~>~ 'À'::citext AS f;
|
|
f
|
|
---
|
|
f
|
|
(1 row)
|
|
|
|
SELECT 'à'::citext ~>=~ 'À'::citext AS t;
|
|
t
|
|
---
|
|
t
|
|
(1 row)
|
|
|