Rémi Bernon (@rbernon) commented about dlls/kernel32/tests/actctx.c:
MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT),
MAKELANGID(LANG_NEUTRAL, SUBLANG_SYS_DEFAULT),
MAKELANGID(LANG_ENGLISH, SUBLANG_NEUTRAL),
MAKELANGID(LANG_ENGLISH, SUBLANG_DEFAULT),
MAKELANGID(LANG_INVARIANT, SUBLANG_NEUTRAL),
- };
- size_t i;
- for (i = 0; i < ARRAY_SIZE(langs); i++)
- {
winetest_push_context("[%Iu]lang=0x%04x", i, langs[i]);
subtest_valid_manifest_resources_locale(langs[i]);
winetest_pop_context();
- }
+}
All these tests seem a bit overkill to end up with the conclusion that only the first manifest resource is used. I think you should make them much simpler because with this amount of code, what they are trying to uncover becomes quite obscure.
And the remaining test failures after the last patch are also unrelated, and IIUC something instead related to loading localized assemblies, and which is commented as FIXME in `lookup_assembly`.