Paul Eggert wrote:
> > Since, as you said, it's not worth worrying about,
> > I would guess that there is no need to update the documentation?
>
> Yes, the posixtm bug's not worth worrying about. However the underlying
> localtime issue might be worth a mention.
Nice. I would not have know what to write here.
> MacOS X 10.5, Solaris 11.3.
Ah indeed, I still get the test failure on Mac OS X 10.5. Need to silence
it for this platform as well:
2021-01-01 Bruno Haible <br...@clisp.org>
posixtm tests: Disable part of the test on plaforms where it fails.
* tests/test-posixtm.c (T): Disable two tests on macOS as well.
diff --git a/tests/test-posixtm.c b/tests/test-posixtm.c
index 0d55608..ea20564 100644
--- a/tests/test-posixtm.c
+++ b/tests/test-posixtm.c
@@ -46,10 +46,10 @@ static struct posixtm_test const T[] =
{ "12131415.16", LY, 1, 0}, /* ??? Dec 13 14:15:16 ???? */
{ "12131415", LY, 1, 0}, /* ??? Dec 13 14:15:00 ???? */
-#if !defined __sun
- /* These two tests fail on 64-bit Solaris up through at least
- Solaris 11.3, which is off by one day for timestamps before
- 0001-01-01 00:00:00 UTC. */
+#if !((defined __APPLE__ && defined __MACH__) || defined __sun)
+ /* These two tests fail on 64-bit Mac OS X 10.5 and on 64-bit Solaris up
+ through at least Solaris 11.3, which is off by one day for timestamps
+ before 0001-01-01 00:00:00 UTC. */
{ "000001010000.00", LY, 1,
- INT64_C (62167219200)},/* Sat Jan 1 00:00:00 0 */
{ "000012312359.59", LY, 1,
