Rethink behavior of pg_import_system_collations().

Marco Atzeri reported that initdb would fail if "locale -a" reported
the same locale name more than once.  All previous versions of Postgres
implicitly de-duplicated the results of "locale -a", but the rewrite
to move the collation import logic into C had lost that property.
It had also lost the property that locale names matching built-in
collation names were silently ignored.

The simplest way to fix this is to make initdb run the function in
if-not-exists mode, which means that there's no real use-case for
non if-not-exists mode; we might as well just drop the boolean argument
and simplify the function's definition to be "add any collations not
already known".  This change also gets rid of some odd corner cases
caused by the fact that aliases were added in if-not-exists mode even
if the function argument said otherwise.

While at it, adjust the behavior so that pg_import_system_collations()
doesn't spew "collation foo already exists, skipping" messages during a
re-run; that's completely unhelpful, especially since there are often
hundreds of them.  And make it return a count of the number of collations
it did add, which seems like it might be helpful.

Also, re-integrate the previous coding's property that it would make a
deterministic selection of which alias to use if there were conflicting
possibilities.  This would only come into play if "locale -a" reports
multiple equivalent locale names, say "de_DE.utf8" and "de_DE.UTF-8",
but that hardly seems out of the question.

In passing, fix incorrect behavior in pg_import_system_collations()'s
ICU code path: it neglected CommandCounterIncrement, which would result
in failures if ICU returns duplicate names, and it would try to create
comments even if a new collation hadn't been created.

Also, reorder operations in initdb so that the 'ucs_basic' collation
is created before calling pg_import_system_collations() not after.
This prevents a failure if "locale -a" were to report a locale named
that.  There's no reason to think that that ever happens in the wild,
but the old coding would have survived it, so let's be equally robust.

Discussion: https://postgr.es/m/20c74bc3-d6ca-243d-1bbc-12f17fa4fe9a@gmail.com

doc/src/sgml/func.sgml

@@ -19711,9 +19711,9 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
       <row>
        <entry>
         <indexterm><primary>pg_import_system_collations</primary></indexterm>
-        <literal><function>pg_import_system_collations(<parameter>if_not_exists</> <type>boolean</>, <parameter>schema</> <type>regnamespace</>)</function></literal>
+        <literal><function>pg_import_system_collations(<parameter>schema</> <type>regnamespace</>)</function></literal>
        </entry>
-       <entry><type>void</type></entry>
+       <entry><type>integer</type></entry>
        <entry>Import operating system collations</entry>
       </row>
      </tbody>
@@ -19730,18 +19730,20 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
    </para>
 
    <para>
-    <function>pg_import_system_collations</> populates the system
-    catalog <literal>pg_collation</literal> with collations based on all the
-    locales it finds on the operating system.  This is
+    <function>pg_import_system_collations</> adds collations to the system
+    catalog <literal>pg_collation</literal> based on all the
+    locales it finds in the operating system.  This is
     what <command>initdb</command> uses;
     see <xref linkend="collation-managing"> for more details.  If additional
     locales are installed into the operating system later on, this function
-    can be run again to add collations for the new locales.  In that case, the
-    parameter <parameter>if_not_exists</parameter> should be set to true to
-    skip over existing collations.  The <parameter>schema</parameter>
-    parameter would typically be <literal>pg_catalog</literal>, but that is
-    not a requirement.  (Collation objects based on locales that are no longer
-    present on the operating system are never removed by this function.)
+    can be run again to add collations for the new locales.  Locales that
+    match existing entries in <literal>pg_collation</literal> will be skipped.
+    (But collation objects based on locales that are no longer
+    present in the operating system are not removed by this function.)
+    The <parameter>schema</parameter> parameter would typically
+    be <literal>pg_catalog</literal>, but that is not a requirement;
+    the collations could be installed into some other schema as well.
+    The function returns the number of new collation objects it created.
    </para>
 
   </sect2>

src/backend/catalog/pg_collation.c

@@ -37,6 +37,11 @@
  * CollationCreate
  *
  * Add a new tuple to pg_collation.
+ *
+ * if_not_exists: if true, don't fail on duplicate name, just print a notice
+ * and return InvalidOid.
+ * quiet: if true, don't fail on duplicate name, just silently return
+ * InvalidOid (overrides if_not_exists).
  */
 Oid
 CollationCreate(const char *collname, Oid collnamespace,
@@ -45,7 +50,8 @@ CollationCreate(const char *collname, Oid collnamespace,
 				int32 collencoding,
 				const char *collcollate, const char *collctype,
 				const char *collversion,
-				bool if_not_exists)
+				bool if_not_exists,
+				bool quiet)
 {
 	Relation	rel;
 	TupleDesc	tupDesc;
@@ -77,7 +83,9 @@ CollationCreate(const char *collname, Oid collnamespace,
 							  Int32GetDatum(collencoding),
 							  ObjectIdGetDatum(collnamespace)))
 	{
-		if (if_not_exists)
+		if (quiet)
+			return InvalidOid;
+		else if (if_not_exists)
 		{
 			ereport(NOTICE,
 					(errcode(ERRCODE_DUPLICATE_OBJECT),
@@ -119,7 +127,12 @@ CollationCreate(const char *collname, Oid collnamespace,
 							   Int32GetDatum(-1),
 							   ObjectIdGetDatum(collnamespace))))
 	{
-		if (if_not_exists)
+		if (quiet)
+		{
+			heap_close(rel, NoLock);
+			return InvalidOid;
+		}
+		else if (if_not_exists)
 		{
 			heap_close(rel, NoLock);
 			ereport(NOTICE,

src/bin/initdb/initdb.c

@@ -1636,10 +1636,16 @@ setup_description(FILE *cmdfd)
 static void
 setup_collation(FILE *cmdfd)
 {
-	PG_CMD_PUTS("SELECT pg_import_system_collations(if_not_exists => false, schema => 'pg_catalog');\n\n");
+	/*
+	 * Add an SQL-standard name.  We don't want to pin this, so it doesn't go
+	 * in pg_collation.h.  But add it before reading system collations, so
+	 * that it wins if libc defines a locale named ucs_basic.
+	 */
+	PG_CMD_PRINTF3("INSERT INTO pg_collation (collname, collnamespace, collowner, collprovider, collencoding, collcollate, collctype) VALUES ('ucs_basic', 'pg_catalog'::regnamespace, %u, '%c', %d, 'C', 'C');\n\n",
+				   BOOTSTRAP_SUPERUSERID, COLLPROVIDER_LIBC, PG_UTF8);
 
-	/* Add an SQL-standard name */
-	PG_CMD_PRINTF3("INSERT INTO pg_collation (collname, collnamespace, collowner, collprovider, collencoding, collcollate, collctype) VALUES ('ucs_basic', 'pg_catalog'::regnamespace, %u, '%c', %d, 'C', 'C');\n\n", BOOTSTRAP_SUPERUSERID, COLLPROVIDER_LIBC, PG_UTF8);
+	/* Now import all collations we can find in the operating system */
+	PG_CMD_PUTS("SELECT pg_import_system_collations('pg_catalog');\n\n");
 }
 
 /*

src/include/catalog/catversion.h

@@ -53,6 +53,6 @@
  */
 
 /*							yyyymmddN */
-#define CATALOG_VERSION_NO	201706202
+#define CATALOG_VERSION_NO	201706231
 
 #endif

src/include/catalog/pg_collation_fn.h

@@ -20,7 +20,8 @@ extern Oid CollationCreate(const char *collname, Oid collnamespace,
 				int32 collencoding,
 				const char *collcollate, const char *collctype,
 				const char *collversion,
-				bool if_not_exists);
+				bool if_not_exists,
+				bool quiet);
 extern void RemoveCollationById(Oid collationOid);
 
 #endif							/* PG_COLLATION_FN_H */

src/include/catalog/pg_proc.h

@@ -5462,11 +5462,12 @@ DESCR("pg_controldata recovery state information as a function");
 DATA(insert OID = 3444 ( pg_control_init PGNSP PGUID 12 1 0 0 0 f f f f t f v s 0 0 2249 "" "{23,23,23,23,23,23,23,23,23,16,16,23}" "{o,o,o,o,o,o,o,o,o,o,o,o}" "{max_data_alignment,database_block_size,blocks_per_segment,wal_block_size,bytes_per_wal_segment,max_identifier_length,max_index_columns,max_toast_chunk_size,large_object_chunk_size,float4_pass_by_value,float8_pass_by_value,data_page_checksum_version}" _null_ _null_ pg_control_init _null_ _null_ _null_ ));
 DESCR("pg_controldata init state information as a function");
 
-DATA(insert OID = 3445 ( pg_import_system_collations PGNSP PGUID 12 100 0 0 0 f f f f t f v r 2 0 2278 "16 4089" _null_ _null_ "{if_not_exists,schema}" _null_ _null_ pg_import_system_collations _null_ _null_ _null_ ));
+/* collation management functions */
+DATA(insert OID = 3445 ( pg_import_system_collations PGNSP PGUID 12 100 0 0 0 f f f f t f v r 1 0 23 "4089" _null_ _null_ _null_ _null_ _null_ pg_import_system_collations _null_ _null_ _null_ ));
 DESCR("import collations from operating system");
 
 DATA(insert OID = 3448 ( pg_collation_actual_version PGNSP PGUID 12 100 0 0 0 f f f f t f v s 1 0 25 "26" _null_ _null_ _null_ _null_ _null_ pg_collation_actual_version _null_ _null_ _null_ ));
-DESCR("import collations from operating system");
+DESCR("get actual version of collation from operating system");
 
 /* system management/monitoring related functions */
 DATA(insert OID = 3353 (  pg_ls_logdir				 PGNSP PGUID 12 10 20 0 0 f f f f t t v s 0 0 2249 "" "{25,20,1184}" "{o,o,o}" "{name,size,modification}" _null_ _null_ pg_ls_logdir _null_ _null_ _null_ ));