Hello!
I was surprised that it is allowed to create clusters/databases where
the default ICU collations do not actually work due to unsupported
encodings:
$ initdb --encoding SQL_ASCII --locale-provider icu --icu-locale en-US
-D data &&
pg_ctl -D data -l logfile start &&
psql -c "SELECT 'a' < 'b'" template1
...
waiting for server to start.... done
server started
ERROR: encoding "SQL_ASCII" not supported by ICU
$ createdb --encoding SQL_ASCII --locale-provider icu --icu-locale en-US
--template template0 mydb &&
psql -c "SELECT 'a' < 'b'" mydb
ERROR: encoding "SQL_ASCII" not supported by ICU
The patch diff_check_icu_encoding.patch prohibits the creation of such
objects...
--
Marina Polyakova
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
diff --git a/src/backend/commands/dbcommands.c b/src/backend/commands/dbcommands.c
index 6ff48bb18f3639ae45d9528b32df51a4aebc60c0..07758d15e8613d5a049537ddf2c5992e57ad6424 100644
--- a/src/backend/commands/dbcommands.c
+++ b/src/backend/commands/dbcommands.c
@@ -1042,11 +1042,16 @@ createdb(ParseState *pstate, const CreatedbStmt *stmt)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("ICU locale must be specified")));
- }
- if (dblocprovider == COLLPROVIDER_ICU)
check_icu_locale(dbiculocale);
+ if (!(is_encoding_supported_by_icu(encoding)))
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("encoding \"%s\" is not supported with ICU provider",
+ pg_encoding_to_char(encoding))));
+ }
+
/*
* Check that the new encoding and locale settings match the source
* database. We insist on this because we simply copy the source data ---
diff --git a/src/bin/initdb/initdb.c b/src/bin/initdb/initdb.c
index e00837ecacf4885cf2a176168c283f3e67c6eb53..8a762ced8340c9d8256f7832a4c19a43f1d5538a 100644
--- a/src/bin/initdb/initdb.c
+++ b/src/bin/initdb/initdb.c
@@ -2362,6 +2362,16 @@ setup_locale_encoding(void)
if (!check_locale_encoding(lc_ctype, encodingid) ||
!check_locale_encoding(lc_collate, encodingid))
exit(1); /* check_locale_encoding printed the error */
+
+ if (locale_provider == COLLPROVIDER_ICU &&
+ !(is_encoding_supported_by_icu(encodingid)))
+ {
+ pg_log_error("encoding \"%s\" is not supported with ICU provider",
+ pg_encoding_to_char(encodingid));
+ pg_log_error_hint("Rerun %s and choose a matching combination.",
+ progname);
+ exit(1);
+ }
}
diff --git a/src/bin/initdb/t/001_initdb.pl b/src/bin/initdb/t/001_initdb.pl
index a37f6dd9b334b6ee22d9fdd4d51422795cb54a39..e4bb3d0cdd9c23729c5fb97886374f8df558f239 100644
--- a/src/bin/initdb/t/001_initdb.pl
+++ b/src/bin/initdb/t/001_initdb.pl
@@ -118,6 +118,15 @@ if ($ENV{with_icu} eq 'yes')
],
qr/FATAL: could not open collator for locale/,
'fails for invalid ICU locale');
+
+ command_fails_like(
+ [
+ 'initdb', '--no-sync',
+ '--locale-provider=icu', '--icu-locale=en',
+ '--encoding=SQL_ASCII', "$tempdir/dataX"
+ ],
+ qr/error: encoding "SQL_ASCII" is not supported with ICU provider/,
+ 'encoding "SQL_ASCII" is not supported with ICU provider');
}
else
{
diff --git a/src/bin/scripts/t/020_createdb.pl b/src/bin/scripts/t/020_createdb.pl
index e91c1d013d08d8bd1e3a92f2aba958c5c7713ca6..eaab3caa32669ead068719d98bb953c5c6ff5a17 100644
--- a/src/bin/scripts/t/020_createdb.pl
+++ b/src/bin/scripts/t/020_createdb.pl
@@ -50,6 +50,16 @@ if ($ENV{with_icu} eq 'yes')
],
'fails for invalid ICU locale');
+ $node->command_fails_like(
+ [
+ 'createdb', '-T',
+ 'template0', '--locale-provider=icu',
+ '--icu-locale=en', '--encoding=SQL_ASCII',
+ 'foobarX'
+ ],
+ qr/ERROR: encoding "SQL_ASCII" is not supported with ICU provider/,
+ 'encoding "SQL_ASCII" is not supported with ICU provider');
+
# additional node, which uses the icu provider
my $node2 = PostgreSQL::Test::Cluster->new('icu');
$node2->init(extra => ['--locale-provider=icu', '--icu-locale=en']);
