This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: C++/ObjC identifiers charset conversion cleanup
On Mon, 11 May 2009, David Ayers wrote:
> Original Mail: http://gcc.gnu.org/ml/gcc-patches/2009-05/msg00198.html
>
> Am Dienstag, den 05.05.2009, 21:25 +0000 schrieb Joseph S. Myers:
> > The ObjC changes are straightforward, and the changes to
> > objc_lookup_protocol to have two full diagnostic sentences instead of
> > building up a diagnostic from fragments make the code simpler as well
> > as making it i18n-friendly and making it work with extended
> > identifiers.
>
> [snip]
>
> > Bootstrapped with no regressions on x86_64-unknown-linux-gnu,
> > including Obj-C++. OK to commit?
>
> I'm not an ObjC(++) maintainer (cc'ed them) but I the patch looks good
> to me wrt ObjC... thanks for the i18n cleanups!
>
> If you can point to some analogous C test cases, I could try to come up
> with something for ObjC that tests many of the code paths.
For example, gcc.dg/ucnid-8.c. Put -fextended-identifiers in dg-options,
write code that generates a diagnostic including an identifier containing
a universal character name and verify that the universal character name
(in the \Unnnnnnnn form - which requires \\\\U to match it in
dg-warning/dg-error) appears in the diagnostic rather than the raw UTF-8
form (given that gcc-dg.exp sets LC_ALL=C so the testsuite runs with the
C locale).
--
Joseph S. Myers
joseph@codesourcery.com