v19-0006-Add-builtin-collation-objects-PG_C_UTF8-and-PG_U.patch

From afefb373f8777f772c34f1b2daec88a33de74eb1 Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Mon, 19 Feb 2024 14:43:15 -0800
Subject: [PATCH v19 6/6] Add builtin collation objects PG_C_UTF8 and
 PG_UNICODE_FAST.

---
 doc/src/sgml/charset.sgml                  | 34 ++++++++++++++++
 src/include/catalog/pg_collation.dat       |  6 +++
 src/test/regress/expected/collate.utf8.out | 45 ++++++++++-----------
 src/test/regress/sql/collate.utf8.sql      | 47 ++++++++++------------
 4 files changed, 82 insertions(+), 50 deletions(-)

diff --git a/doc/src/sgml/charset.sgml b/doc/src/sgml/charset.sgml
index 2abd898115..47f137e1f0 100644
--- a/doc/src/sgml/charset.sgml
+++ b/doc/src/sgml/charset.sgml
@@ -874,6 +874,40 @@ SELECT * FROM test1 ORDER BY a || b COLLATE "fr_FR";
       </listitem>
      </varlistentry>
 
+     <varlistentry>
+      <term><literal>PG_C_UTF8</literal></term>
+      <listitem>
+       <para>
+        This collation sorts by Unicode code point.  Behavior is efficient and
+        stable within a <productname>Postgres</productname> major version.
+        For the functions <function>lower</function>,
+        <function>initcap</function>, and <function>upper</function> it uses
+        Unicode simple case mapping.  For pattern matching (including regular
+        expressions), it uses the POSIX Compatible variant of Unicode <ulink
+        url="https://www.unicode.org/reports/tr18/#Compatibility_Properties">Compatibility
+        Properties</ulink>.  This collation is only available for encoding
+        <literal>UTF8</literal>.
+       </para>
+      </listitem>
+     </varlistentry>
+
+     <varlistentry>
+      <term><literal>PG_UNICODE_FAST</literal></term>
+      <listitem>
+       <para>
+        This collation sorts by Unicode code point.  Behavior is efficient and
+        stable within a <productname>Postgres</productname> major version.
+        For the functions <function>lower</function>,
+        <function>initcap</function>, and <function>upper</function> it uses
+        Unicode full case mapping. For pattern matching (including regular
+        expressions), it uses the Standard variant of Unicode <ulink
+        url="https://www.unicode.org/reports/tr18/#Compatibility_Properties">Compatibility
+        Properties</ulink>.  It is only available for encoding
+        <literal>UTF8</literal>.
+       </para>
+      </listitem>
+     </varlistentry>
+
      <varlistentry>
       <term><literal>C</literal> (equivalent to <literal>POSIX</literal>)</term>
       <listitem>
diff --git a/src/include/catalog/pg_collation.dat b/src/include/catalog/pg_collation.dat
index 938432e8a4..a187b3dcef 100644
--- a/src/include/catalog/pg_collation.dat
+++ b/src/include/catalog/pg_collation.dat
@@ -30,5 +30,11 @@
   descr => 'sorts using the Unicode Collation Algorithm with default settings',
   collname => 'unicode', collprovider => 'i', collencoding => '-1',
   colllocale => 'und' },
+{ oid => '811', descr => 'sorts by Unicode code point; Unicode & POSIX character semantics',
+  collname => 'pg_c_utf8', collprovider => 'b', collencoding => '6',
+  colllocale => 'C.UTF-8' },
+{ oid => '812', descr => 'sorts by Unicode code point; Unicode character semantics',
+  collname => 'pg_unicode_fast', collprovider => 'b', collencoding => '6',
+  colllocale => 'PG_UNICODE_FAST' },
 
 ]
diff --git a/src/test/regress/expected/collate.utf8.out b/src/test/regress/expected/collate.utf8.out
index ecf5305105..14cbc0f870 100644
--- a/src/test/regress/expected/collate.utf8.out
+++ b/src/test/regress/expected/collate.utf8.out
@@ -11,10 +11,8 @@ SET client_encoding TO UTF8;
 --
 -- Test builtin PG_UNICODE_FAST locale.
 --
-CREATE COLLATION BUILTIN_UNICODE_FAST
-  ( provider = builtin, locale = 'PG_UNICODE_FAST' );
 CREATE TABLE builtin_test1 (
-  t TEXT COLLATE BUILTIN_UNICODE_FAST
+  t TEXT COLLATE PG_UNICODE_FAST
 );
 INSERT INTO builtin_test1 VALUES
   ('abc DEF'),
@@ -42,130 +40,129 @@ SELECT
 
 DROP TABLE builtin_test1;
 -- test Final_Sigma
-SELECT lower('ΑΣ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3
+SELECT lower('ΑΣ' COLLATE PG_UNICODE_FAST); -- 0391 03A3
  lower 
 -------
  ας
 (1 row)
 
-SELECT lower('ΑΣ0' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3 0030
+SELECT lower('ΑΣ0' COLLATE PG_UNICODE_FAST); -- 0391 03A3 0030
  lower 
 -------
  ας0
 (1 row)
 
-SELECT lower('ἈΣ̓' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0343 03A3 0343
+SELECT lower('ἈΣ̓' COLLATE PG_UNICODE_FAST); -- 0391 0343 03A3 0343
  lower 
 -------
  ἀς̓
 (1 row)
 
-SELECT lower('ᾼΣͅ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0345 03A3 0345
+SELECT lower('ᾼΣͅ' COLLATE PG_UNICODE_FAST); -- 0391 0345 03A3 0345
  lower 
 -------
  ᾳςͅ
 (1 row)
 
 -- test !Final_Sigma
-SELECT lower('Σ' COLLATE BUILTIN_UNICODE_FAST); -- 03A3
+SELECT lower('Σ' COLLATE PG_UNICODE_FAST); -- 03A3
  lower 
 -------
  σ
 (1 row)
 
-SELECT lower('0Σ' COLLATE BUILTIN_UNICODE_FAST); -- 0030 03A3
+SELECT lower('0Σ' COLLATE PG_UNICODE_FAST); -- 0030 03A3
  lower 
 -------
  0σ
 (1 row)
 
-SELECT lower('ΑΣΑ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3 0391
+SELECT lower('ΑΣΑ' COLLATE PG_UNICODE_FAST); -- 0391 03A3 0391
  lower 
 -------
  ασα
 (1 row)
 
-SELECT lower('ἈΣ̓Α' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0343 03A3 0343 0391
+SELECT lower('ἈΣ̓Α' COLLATE PG_UNICODE_FAST); -- 0391 0343 03A3 0343 0391
  lower 
 -------
  ἀσ̓α
 (1 row)
 
-SELECT lower('ᾼΣͅΑ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0345 03A3 0345 0391
+SELECT lower('ᾼΣͅΑ' COLLATE PG_UNICODE_FAST); -- 0391 0345 03A3 0345 0391
  lower 
 -------
  ᾳσͅα
 (1 row)
 
 -- properties
-SELECT 'xyz' ~ '[[:alnum:]]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xyz' ~ '[[:alnum:]]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'xyz' !~ '[[:upper:]]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xyz' !~ '[[:upper:]]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT '@' !~ '[[:alnum:]]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT '@' !~ '[[:alnum:]]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT '=' !~ '[[:punct:]]' COLLATE BUILTIN_UNICODE_FAST; -- symbols are not punctuation
+SELECT '=' !~ '[[:punct:]]' COLLATE PG_UNICODE_FAST; -- symbols are not punctuation
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'a8a' ~ '[[:digit:]]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'a8a' ~ '[[:digit:]]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT '൧' ~ '\d' COLLATE BUILTIN_UNICODE_FAST;
+SELECT '൧' ~ '\d' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
 -- case mapping
-SELECT 'xYz' ~* 'XyZ' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xYz' ~* 'XyZ' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'xAb' ~* '[W-Y]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xAb' ~* '[W-Y]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'xAb' !~* '[c-d]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xAb' !~* '[c-d]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'Δ' ~* '[α-λ]' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'Δ' ~* '[α-λ]' COLLATE PG_UNICODE_FAST;
  ?column? 
 ----------
  t
 (1 row)
 
-SELECT 'δ' ~* '[Γ-Λ]' COLLATE BUILTIN_UNICODE_FAST; -- same as above with cases reversed
+SELECT 'δ' ~* '[Γ-Λ]' COLLATE PG_UNICODE_FAST; -- same as above with cases reversed
  ?column? 
 ----------
  t
 (1 row)
 
-DROP COLLATION BUILTIN_UNICODE_FAST;
 --
 -- Test builtin C.UTF-8 locale.
 --
diff --git a/src/test/regress/sql/collate.utf8.sql b/src/test/regress/sql/collate.utf8.sql
index c0da1e7cb4..0c799b24a2 100644
--- a/src/test/regress/sql/collate.utf8.sql
+++ b/src/test/regress/sql/collate.utf8.sql
@@ -15,11 +15,8 @@ SET client_encoding TO UTF8;
 -- Test builtin PG_UNICODE_FAST locale.
 --
 
-CREATE COLLATION BUILTIN_UNICODE_FAST
-  ( provider = builtin, locale = 'PG_UNICODE_FAST' );
-
 CREATE TABLE builtin_test1 (
-  t TEXT COLLATE BUILTIN_UNICODE_FAST
+  t TEXT COLLATE PG_UNICODE_FAST
 );
 INSERT INTO builtin_test1 VALUES
   ('abc DEF'),
@@ -40,36 +37,34 @@ SELECT
 DROP TABLE builtin_test1;
 
 -- test Final_Sigma
-SELECT lower('ΑΣ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3
-SELECT lower('ΑΣ0' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3 0030
-SELECT lower('ἈΣ̓' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0343 03A3 0343
-SELECT lower('ᾼΣͅ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0345 03A3 0345
+SELECT lower('ΑΣ' COLLATE PG_UNICODE_FAST); -- 0391 03A3
+SELECT lower('ΑΣ0' COLLATE PG_UNICODE_FAST); -- 0391 03A3 0030
+SELECT lower('ἈΣ̓' COLLATE PG_UNICODE_FAST); -- 0391 0343 03A3 0343
+SELECT lower('ᾼΣͅ' COLLATE PG_UNICODE_FAST); -- 0391 0345 03A3 0345
 
 -- test !Final_Sigma
-SELECT lower('Σ' COLLATE BUILTIN_UNICODE_FAST); -- 03A3
-SELECT lower('0Σ' COLLATE BUILTIN_UNICODE_FAST); -- 0030 03A3
-SELECT lower('ΑΣΑ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 03A3 0391
-SELECT lower('ἈΣ̓Α' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0343 03A3 0343 0391
-SELECT lower('ᾼΣͅΑ' COLLATE BUILTIN_UNICODE_FAST); -- 0391 0345 03A3 0345 0391
+SELECT lower('Σ' COLLATE PG_UNICODE_FAST); -- 03A3
+SELECT lower('0Σ' COLLATE PG_UNICODE_FAST); -- 0030 03A3
+SELECT lower('ΑΣΑ' COLLATE PG_UNICODE_FAST); -- 0391 03A3 0391
+SELECT lower('ἈΣ̓Α' COLLATE PG_UNICODE_FAST); -- 0391 0343 03A3 0343 0391
+SELECT lower('ᾼΣͅΑ' COLLATE PG_UNICODE_FAST); -- 0391 0345 03A3 0345 0391
 
 -- properties
 
-SELECT 'xyz' ~ '[[:alnum:]]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT 'xyz' !~ '[[:upper:]]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT '@' !~ '[[:alnum:]]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT '=' !~ '[[:punct:]]' COLLATE BUILTIN_UNICODE_FAST; -- symbols are not punctuation
-SELECT 'a8a' ~ '[[:digit:]]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT '൧' ~ '\d' COLLATE BUILTIN_UNICODE_FAST;
+SELECT 'xyz' ~ '[[:alnum:]]' COLLATE PG_UNICODE_FAST;
+SELECT 'xyz' !~ '[[:upper:]]' COLLATE PG_UNICODE_FAST;
+SELECT '@' !~ '[[:alnum:]]' COLLATE PG_UNICODE_FAST;
+SELECT '=' !~ '[[:punct:]]' COLLATE PG_UNICODE_FAST; -- symbols are not punctuation
+SELECT 'a8a' ~ '[[:digit:]]' COLLATE PG_UNICODE_FAST;
+SELECT '൧' ~ '\d' COLLATE PG_UNICODE_FAST;
 
 -- case mapping
 
-SELECT 'xYz' ~* 'XyZ' COLLATE BUILTIN_UNICODE_FAST;
-SELECT 'xAb' ~* '[W-Y]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT 'xAb' !~* '[c-d]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT 'Δ' ~* '[α-λ]' COLLATE BUILTIN_UNICODE_FAST;
-SELECT 'δ' ~* '[Γ-Λ]' COLLATE BUILTIN_UNICODE_FAST; -- same as above with cases reversed
-
-DROP COLLATION BUILTIN_UNICODE_FAST;
+SELECT 'xYz' ~* 'XyZ' COLLATE PG_UNICODE_FAST;
+SELECT 'xAb' ~* '[W-Y]' COLLATE PG_UNICODE_FAST;
+SELECT 'xAb' !~* '[c-d]' COLLATE PG_UNICODE_FAST;
+SELECT 'Δ' ~* '[α-λ]' COLLATE PG_UNICODE_FAST;
+SELECT 'δ' ~* '[Γ-Λ]' COLLATE PG_UNICODE_FAST; -- same as above with cases reversed
 
 --
 -- Test builtin C.UTF-8 locale.
-- 
2.34.1