"Stephen J. Turnbull" <[EMAIL PROTECTED]> writes:

> You don't consider tool-chain-induced corruption serious?  I certainly
> do.

Like the autoconf maintainers, I don't consider the bulk of autoconf
bugs serious (and I'm quite familiar with it).  No evidence of serious
problems is apparent, just FUD.

> > > AFAIK 2.59 is happy enough with datarootdir-correct configure.in, so
> > > there's no need to bump the require once you've fixed configure.in.
> > 
> > > 2.59 knows nothing about datarootdir, so you can't trivially DTRT.

> "First, do no harm."  Even before DTRTing.

Please don't be insulting.  The `AFAIK' is wrong -- look at the patch.

> The question here is, does 2.59 barf on 2.61-correct code, or does it
> ignore it?

No, it isn't the question.

> However, I don't consider non-conformance a serious problem for
> non-GNU software, and so I recommend that Darcs not try to enforce
> use of 2.6x.

Notwithstanding words in the subject, I'm not trying to enforce GNU
maintenance standards generally:  this is a specific issue.  I'll
listen to the maintainers on the topic (who've accepted previous
configure changes to follow autotools conventions better).

> XEmacs has already dealt with the known problems, obviously.

Then I guess XEmacs cocked something up; Emacs development code
actually requires autoconf 2.61.  There doesn't seem to be a relevant
bug report against autoconf from XEmacs anyhow -- I actually looked.

> I just
> would prefer that Darcs not risk gratuitously imposing such pain on
> other projects when the autotools have such a horrible record of
> backward incompatibility.

This pain seems to be a figment of the imagination, and, as I said,
Debian has already imposed it on me and many others -- Debian's darcs
on my system is built with autoconf 2.60.
_______________________________________________
darcs-devel mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-devel

Reply via email to