Re: [hjl@lucon.org: Re: c++/5577 (was: Re: c++/4862 and PRs...)]

["H . J . Lu" <hjl at lucon dot org>]

On Sun, Feb 03, 2002 at 08:00:10PM +0100, Paolo Carlini wrote:
> "H . J . Lu" wrote:
> 
> > I am sending this just for a record in the gcc mailing list archive
> > since my last mail was not archived.
> 
> Thanks H.J.
> 
> I have been meaning to regression test it on i686-pc-linux-gnu, but did'nt find the time,
> today. Perhaps you can do it?

See

http://gcc.gnu.org/ml/gcc-testresults/2002-02/msg00042.html

This is with my updated patch and

http://gcc.gnu.org/ml/gcc-testresults/2002-02/msg00030.html

This is with my original patch. I always have

XPASS: g++.dg/template/friend.C  (test for bogus messages, line 19)
XPASS: g++.dg/template/friend.C  (test for bogus messages, line 21)
XPASS: g++.dg/template/friend.C  (test for bogus messages, line 28)

I am not sure if it has anything to do with my original patch.

> 
> I know that Mark Mitchell obviously does not encourage fixing parser bugs for 3.1 but this is
> really a long awaited fix, IMHO, and it is worth proposing in "full glory" complete of
> regression testing done.
> 

I asked about it a long time ago and posted an incomplete patch along
with a testcase:

http://gcc.gnu.org/ml/gcc-patches/2001-08/msg01274.html
http://gcc.gnu.org/ml/gcc-patches/2001-08/msg01378.html

But there was no reponse. So I didn't bother with my complete patch.
Could someone please at least take a look at my testcase?

Thanks.


H.J.