This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: Common V3 coding bug
- To: Jason Merrill <jason_merrill at redhat dot com>
- Subject: Re: Common V3 coding bug
- From: Benjamin Kosnik <bkoz at redhat dot com>
- Date: Wed, 18 Jul 2001 09:59:07 -0700 (PDT)
- cc: Gabriel Dos Reis <gdr at codesourcery dot com>, Mark Mitchell <mark at codesourcery dot com>, libstdc++ at gcc dot gnu dot org
Thanks Mark.
> I am. People expect to get the version from their base class--thus the
> broken code in v3, which was written by experts. I have no objection to a
> flag to turn off such a warning, but I feel strongly that it should be on
> by default.
Me too.
>From the discussion following Mark's initial posting, this looks like
something that needs to be explained with some kind of intelligent
diagnostic message and an error. I suspect the discussion has already
pointed out the phrasing for the error.
For future reference, I'm in favor of a strict and
standards-conformant compiler by default so that bugs like this are
easily found.
If you keep posting bugs like this Mark, I'm going to have to try your
branch, sooner rather than later. Please post on this list when you want
people to bang on it.
;)
Also, let's fix this code up on the trunk.
Mark, you've already done several patches
to libstdc++-v3 as you've found broken items (two come to mind, including
this one.) I know you are working on a branch, but I'm trying to keep the
trunk the cannonical source for libstdc++-v3: as a courtesy to me and the
other libstdc++ developers, can you please check in both of these patches
to the trunk?
Thanks in advance,
benjamin