Обсуждение: Regression test collate.icu.utf8 failed on REL_14_STABLE
Hi, hackers
I find when I compile PG 14 with --with-icu, collate.icu.utf8 and foreign_data
regression tests will failed. However it is OK on REL_15_STABLE and master.
I also test this on REL_13_STABLE, and it also failed. Here is the regression
diffs.
diff -U3 /home/japin/Codes/postgres/build/../src/test/regress/expected/collate.icu.utf8.out
/home/japin/Codes/postgres/build/src/test/regress/results/collate.icu.utf8.out
--- /home/japin/Codes/postgres/build/../src/test/regress/expected/collate.icu.utf8.out 2023-08-14 17:37:31.960448245
+0800
+++ /home/japin/Codes/postgres/build/src/test/regress/results/collate.icu.utf8.out 2023-08-14 21:30:44.335214886
+0800
@@ -1035,6 +1035,9 @@
quote_literal(current_setting('lc_ctype')) || ');';
END
$$;
+ERROR: collations with different collate and ctype values are not supported by ICU
+CONTEXT: SQL statement "CREATE COLLATION test1 (provider = icu, lc_collate = 'C.UTF-8', lc_ctype = 'en_US.UTF-8');"
+PL/pgSQL function inline_code_block line 3 at EXECUTE
CREATE COLLATION test3 (provider = icu, lc_collate = 'en_US.utf8'); -- fail, need lc_ctype
ERROR: parameter "lc_ctype" must be specified
CREATE COLLATION testx (provider = icu, locale = 'nonsense'); /* never fails with ICU */ DROP COLLATION testx;
@@ -1045,13 +1048,12 @@
collname
----------
test0
- test1
test5
-(3 rows)
+(2 rows)
ALTER COLLATION test1 RENAME TO test11;
+ERROR: collation "test1" for encoding "UTF8" does not exist
ALTER COLLATION test0 RENAME TO test11; -- fail
-ERROR: collation "test11" already exists in schema "collate_tests"
ALTER COLLATION test1 RENAME TO test22; -- fail
ERROR: collation "test1" for encoding "UTF8" does not exist
ALTER COLLATION test11 OWNER TO regress_test_role;
@@ -1059,18 +1061,19 @@
ERROR: role "nonsense" does not exist
ALTER COLLATION test11 SET SCHEMA test_schema;
COMMENT ON COLLATION test0 IS 'US English';
+ERROR: collation "test0" for encoding "UTF8" does not exist
SELECT collname, nspname, obj_description(pg_collation.oid, 'pg_collation')
FROM pg_collation JOIN pg_namespace ON (collnamespace = pg_namespace.oid)
WHERE collname LIKE 'test%'
ORDER BY 1;
collname | nspname | obj_description
----------+---------------+-----------------
- test0 | collate_tests | US English
test11 | test_schema |
test5 | collate_tests |
-(3 rows)
+(2 rows)
DROP COLLATION test0, test_schema.test11, test5;
+ERROR: collation "test0" for encoding "UTF8" does not exist
DROP COLLATION test0; -- fail
ERROR: collation "test0" for encoding "UTF8" does not exist
DROP COLLATION IF EXISTS test0;
@@ -1078,10 +1081,17 @@
SELECT collname FROM pg_collation WHERE collname LIKE 'test%';
collname
----------
-(0 rows)
+ test11
+ test5
+(2 rows)
DROP SCHEMA test_schema;
+ERROR: cannot drop schema test_schema because other objects depend on it
+DETAIL: collation test_schema.test11 depends on schema test_schema
+HINT: Use DROP ... CASCADE to drop the dependent objects too.
DROP ROLE regress_test_role;
+ERROR: role "regress_test_role" cannot be dropped because some objects depend on it
+DETAIL: owner of collation test_schema.test11
-- ALTER
ALTER COLLATION "en-x-icu" REFRESH VERSION;
NOTICE: version has not changed
diff -U3 /home/japin/Codes/postgres/build/../src/test/regress/expected/foreign_data.out
/home/japin/Codes/postgres/build/src/test/regress/results/foreign_data.out
--- /home/japin/Codes/postgres/build/../src/test/regress/expected/foreign_data.out 2023-08-14 17:37:31.964448260
+0800
+++ /home/japin/Codes/postgres/build/src/test/regress/results/foreign_data.out 2023-08-14 21:30:55.571170376 +0800
@@ -5,10 +5,13 @@
-- Suppress NOTICE messages when roles don't exist
SET client_min_messages TO 'warning';
DROP ROLE IF EXISTS regress_foreign_data_user, regress_test_role, regress_test_role2, regress_test_role_super,
regress_test_indirect,regress_unprivileged_role;
+ERROR: role "regress_test_role" cannot be dropped because some objects depend on it
+DETAIL: owner of collation test_schema.test11
RESET client_min_messages;
CREATE ROLE regress_foreign_data_user LOGIN SUPERUSER;
SET SESSION AUTHORIZATION 'regress_foreign_data_user';
CREATE ROLE regress_test_role;
+ERROR: role "regress_test_role" already exists
CREATE ROLE regress_test_role2;
CREATE ROLE regress_test_role_super SUPERUSER;
CREATE ROLE regress_test_indirect;
Is it a bug that fixed in REL_15_STABLE? If yes, why not backpatch?
--
Regrads,
Japin Li
DROP SCHEMA test_schema;
+ERROR: cannot drop schema test_schema because other objects depend on it
+DETAIL: collation test_schema.test11 depends on schema test_schema
+HINT: Use DROP ... CASCADE to drop the dependent objects too.
DROP ROLE regress_test_role;
+ERROR: role "regress_test_role" cannot be dropped because some objects depend on it
+DETAIL: owner of collation test_schema.test11
+ERROR: role "regress_test_role" cannot be dropped because some objects depend on it
+DETAIL: owner of collation test_schema.test11
+ERROR: role "regress_test_role" already exists
was a failure before this round of test? This looks to me that there
are some objects are not cleaned well before this run. you can try
'make installcheck' with a pretty clean setup or run 'make check'
directly to verify this.
--
Best Regards
Andy Fan
On Tue, 15 Aug 2023 at 08:49, Andy Fan <zhihui.fan1213@gmail.com> wrote:
>>
>>
>>
>> DROP SCHEMA test_schema;
>> +ERROR: cannot drop schema test_schema because other objects depend on it
>> +DETAIL: collation test_schema.test11 depends on schema test_schema
>> +HINT: Use DROP ... CASCADE to drop the dependent objects too.
>> DROP ROLE regress_test_role;
>> +ERROR: role "regress_test_role" cannot be dropped because some objects
>> depend on it
>> +DETAIL: owner of collation test_schema.test11
>> +ERROR: role "regress_test_role" cannot be dropped because some objects
>> depend on it
>> +DETAIL: owner of collation test_schema.test11
>>
>> +ERROR: role "regress_test_role" already exists
>>
>>
>>
>>
> Did you run 'make installcheck' rather than 'make check' and there
> was a failure before this round of test? This looks to me that there
> are some objects are not cleaned well before this run. you can try
> 'make installcheck' with a pretty clean setup or run 'make check'
> directly to verify this.
I used `make check` and cleanup the entire build directory. Here is my
compile & build script.
$ cat compile.sh
#!/bin/bash
set -e
rm -rf $(ls -I '*.sh')
../configure \
--prefix=$PWD/pg \
--enable-tap-tests \
--enable-debug \
--enable-cassert \
--enable-depend \
--enable-dtrace \
--with-icu \
--with-llvm \
--with-openssl \
--with-python \
--with-libxml \
--with-libxslt \
--with-lz4 \
--with-pam \
CFLAGS='-O0 -Wmissing-prototypes -Wincompatible-pointer-types' \
>configure.log 2>&1
make -j $(nproc) -s && make install -s
(cd contrib/ && make -j $(nproc) -s && make install -s)
--
Regrads,
Japin Li
On Tue, 15 Aug 2023 at 08:54, Japin Li <japinli@hotmail.com> wrote: > On Tue, 15 Aug 2023 at 08:49, Andy Fan <zhihui.fan1213@gmail.com> wrote: >>> >>> >>> >>> DROP SCHEMA test_schema; >>> +ERROR: cannot drop schema test_schema because other objects depend on it >>> +DETAIL: collation test_schema.test11 depends on schema test_schema >>> +HINT: Use DROP ... CASCADE to drop the dependent objects too. >>> DROP ROLE regress_test_role; >>> +ERROR: role "regress_test_role" cannot be dropped because some objects >>> depend on it >>> +DETAIL: owner of collation test_schema.test11 >>> +ERROR: role "regress_test_role" cannot be dropped because some objects >>> depend on it >>> +DETAIL: owner of collation test_schema.test11 >>> >>> +ERROR: role "regress_test_role" already exists >>> >>> >>> >>> >> Did you run 'make installcheck' rather than 'make check' and there >> was a failure before this round of test? This looks to me that there >> are some objects are not cleaned well before this run. you can try >> 'make installcheck' with a pretty clean setup or run 'make check' >> directly to verify this. > Thanks Andy, I think I find the root cause. In my environment, LANG has different setting from others. $ locale LANG=C.UTF-8 LANGUAGE= LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL=en_US.UTF-8 Then, I set LANG to en_US.UTF-8, all tests passed. What I'm curious about is why PG 15 can pass. -- Regrads, Japin Li