This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix wide charater literals (was Re: [RFC] Unexpected compiler twist of the wcscoll issue)
- From: Benjamin Kosnik <bkoz at redhat dot com>
- To: gcc-patches at gcc dot gnu dot org, jakub at redhat dot com
- Date: Tue, 12 Mar 2002 08:44:43 -0800
- Subject: Re: [PATCH] Fix wide charater literals (was Re: [RFC] Unexpected compiler twist of the wcscoll issue)
Whoooohoooooo! Thanks Jakub. Thanks also for adding the testcase.
> Shouldn't testsuite framework use C locale for tests,
> or is there some way how to specify environment for a particular testcase?
You can use setlocale("C") to force to the "C" locale. If you look at
the libstdc++/testsuite/22_locale tests, most of them do this. (At
least for some tests.)
Anyway.
> Well, at least on powerpc-linux-gnu there are 2 more:
> codecvt_members_unicode_char.exe:
> codecvt_members_unicode_wchar_t.exe:
I've long been puzzled by these. They appear and then disappear for no
reason on powerpc: at first I thought they were endian issues with the
library code, but then I saw correct results from powerpc (before they
went bad again.)
alpha, ia64, others get these two tests right. I dunno what's up.
-benjamin