Re: New Object Access Type hooks

On 3/24/22 16:59, Andres Freund wrote:
> Hi,
>
> I just rebased the meson tree (atop 75b1521dae1) and the test_oat_hooks test
> fail on windows with meson. They don't with our "homegrown" buildsystem, but
> just because it won't run them.
>
> https://cirrus-ci.com/build/6101947223638016
> https://cirrus-ci.com/task/5869668815601664?logs=check_world#L67
> https://api.cirrus-ci.com/v1/artifact/task/5869668815601664/log/build/testrun/test_oat_hooks/pg_regress/regression.diffs
>
> diff -w -U3 C:/cirrus/src/test/modules/test_oat_hooks/expected/test_oat_hooks.out C:/cirrus/build/testrun/test_oat_hooks/pg_regress/results/test_oat_hooks.out
> --- C:/cirrus/src/test/modules/test_oat_hooks/expected/test_oat_hooks.out 2022-03-24 18:56:39.592048000 +0000
> +++ C:/cirrus/build/testrun/test_oat_hooks/pg_regress/results/test_oat_hooks.out 2022-03-24 19:03:33.910466700 +0000
> @@ -15,12 +15,6 @@
> NOTICE: in process utility: superuser finished CreateRoleStmt
> CREATE TABLE regress_test_table (t text);
> NOTICE: in process utility: superuser attempting CreateStmt
> -NOTICE: in object access: superuser attempting namespace search (subId=0) [no report on violation, allowed]
> -LINE 1: CREATE TABLE regress_test_table (t text);
> - ^
> -NOTICE: in object access: superuser finished namespace search (subId=0) [no report on violation, allowed]
> -LINE 1: CREATE TABLE regress_test_table (t text);
> - ^
> NOTICE: in object access: superuser attempting create (subId=0) [explicit]
> NOTICE: in object access: superuser finished create (subId=0) [explicit]
> NOTICE: in object access: superuser attempting create (subId=0) [explicit]
>
>
> I don't think this is meson specific...

Even if you use NO_LOCALE=1/ENCODING=UTF8 as the Makefile now does?

cheers

andrew

--
Andrew Dunstan
EDB: https://www.enterprisedb.com