Piotr Caban (@piotr) commented about dlls/ucrtbase/tests/file.c:
+static void test_fopen(void) +{ + int i; + FILE *f; + wchar_t wpath[MAX_PATH]; + static const struct { + const char *loc; + const char *path; + int is_todo; + } tests[] = { + { "German.utf8", "t\xc3\xa4\xc3\x8f\xc3\xb6\xc3\x9f.txt", TRUE }, + { "Polish.utf8", "t\xc4\x99\xc5\x9b\xc4\x87.txt", TRUE }, + { "Turkish.utf8", "t\xc3\x87\xc4\x9e\xc4\xb1\xc4\xb0\xc5\x9e.txt", TRUE }, + { "Arabic.utf8", "t\xd8\xaa\xda\x86.txt", TRUE }, + { "Japanese.utf8", "t\xe3\x82\xaf\xe3\x83\xa4.txt", TRUE }, + { "Chinese.utf8", "t\xe4\xb8\x82\xe9\xbd\xab.txt", TRUE }, It would be good to also add a test for non utf-8 locale. It will confirm that CP_ACP should be used in other cases.
-- https://gitlab.winehq.org/wine/wine/-/merge_requests/6680#note_88140