This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [PATCH] Preliminary fix for codecvt_members_unicode_wchar_t
- From: Benjamin Kosnik <bkoz at redhat dot com>
- To: Paolo Carlini <pcarlini at unitus dot it>
- Cc: libstdc++ at gcc dot gnu dot org
- Date: Mon, 25 Mar 2002 15:12:38 -0800 (PST)
- Subject: Re: [PATCH] Preliminary fix for codecvt_members_unicode_wchar_t
> Thank you very much for these additional details.
> In the next few days I will try to learn more myself about all of this
> starting from these notes.
Thanks. I recommend the ISO C99 bits on the w* functions, and the Single
Unix Spec (Austin Group, IEEE 1003.1-2001) on langinfo for further
reading. Also, the glibc manual is pretty useful. I have yet to see a
graceful explanation of any of this, thus the desire to attempt docs...
Some of this stuff has evolved with time. For instance, the WCHAR_T
encoding is new. So, if possible, I suggest testing on x86 and S/390 with
the same versions of glibc. Since the testresults seem to both use
glibc-2.2.5, it looks like you are ok, but I thought I'd mention it.
> By the way, have you had at look at my "consistency" fix for
> collate_byname.cc? It is not strictly needed but changes those tests
> consistently with the current collate_members_xx.cc.
Yep, that should go in, thanks.
-benjamin