This is the mail archive of the
automake@gnu.org
mailing list for the automake project.
Re: automake and 'strict' Perl5
>>>>> "Rodolphe" == Rodolphe Ortalo <Rodolphe.Ortalo@cert.fr> writes:
Rodolphe> In the version I hacked, I 'promoted' 5 or 6 variables to
Rodolphe> true globals just to get rid of this.
I think I'd rather they remain local for now, but with a FIXME comment
so we can know to fix them later.
>> We've already agreed that the next automake can take advantage of
>> Perl 5 features.
Rodolphe> Even OO-modules, packages, etc. or just the new
Rodolphe> Perl5-specific syntax ?
At some point we will use some of the OO features. There are parts of
automake that would benefit from this, e.g. language support.
Rodolphe> Hmm. I can try. What is the development speed of the cvs
Rodolphe> version currently ? (Or, more directly: how much time such a
Rodolphe> patch would remain useful ?)
Things don't change too awfully fast. The bonus of doing a change in
a cvs sandbox is that you can use "cvs update" to merge our changes
with yours. This makes it easier for you to keep your patch
up-to-date as things change.
>> It would be nice to update aclocal as well. It is much shorter than
>> automake.
Rodolphe> Is this a criticism or an encouragment? :-))
Definitely encouragement!
Rodolphe> But, IIRC, FSF _requires_ a signed agreement from it ? This
Rodolphe> could be the problem ? The main problem I see is a delay
Rodolphe> for that (administrativa is rather heavy here...:-)). I'll
Rodolphe> ask questions... But I'm willing to get you these
Rodolphe> documents. Do you have a URL or something where I could
Rodolphe> check these issues precisely?
Yes, the FSF requires some signed paperwork.
See http://sourceware.cygnus.com/automake/contribute.html for details.
I realize this is a pain (and other automake contributors will confirm
that it is a slow process as well), but it is required.
Tom