perl-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stas Bekman <s...@stason.org>
Subject Re: [mp1] Volunteering to be RM for mod_perl 1.28
Date Fri, 23 May 2003 04:11:52 GMT
Randy Kobes wrote:
> On Fri, 23 May 2003, Stas Bekman wrote:
> 
> 
>>Geoffrey Young wrote:
>>
>>
>>>>So, unless anybody objects today, I'll be posting the current mp1's
>>>>status on (dev|modperl)@perl.apache.org to see if people can come up
>>>>with other issues that could/should be slapped into 1.28.
>>>
>>>the Apache::test stuff definitely needs to be in there.
>>
>>so we need to rename Apache/test.pm with Apache/testold.pm and
>>adjust all the tests.
>>
>>we need to ask Brian Behlendorf to add the following
>>/home/cvs/CVSROOT/modules entry for mod_perl:
>>
>>modperl    modperl &apachetest-alias
>>
>>so Apache-Test will be bundled with modperl-1.x.
>>
>>However this is going to be a problem with PAUSE indexer, since
>>we don't yet have tools for hiding dual-life CPAN packages from
>>PAUSE indexer (not yet a problem with mp2, since it's not
>>indexed). So we probably need to mangle 'make dist' to move
>>Apache-Test into inc/Apache-Test, which hides it from the
>>indexer, but also need to make sure that Makefile.PL picks up
>>these files. Or may be we should not distribute Apache-Test
>>with 1.xx... what do you think?
> 
> 
> Is the worry that the Apache-Test/* in mp1 might get indexed by
> PAUSE, and subsequently confuse that with the Apache-Test
> distribution? I don't think that'd be a problem, as long as
> Apache::Test is registered to a different CPAN author than the
> one who submits mod_perl-1 (I'm not sure what happens in a
> case where two people share ownership of two modules with the
> same name ...). As an example, I have a module that uses an eval
> with
>    $eval = qq{
>         package ExtUtils::MakeMaker::_version;
>         blah;
>    };
> to do something, and the PAUSE indexer sees that as
> an attempt to create a real ExtUtils::MakeMaker::_version,
> but then denies such an attempt, as the real package
> is registered to a different author.

True, but since we are probably going to get to the point where several 
committers share the ownership, that would be a problem. Doug currently owns 
mod_perl-1.x distro modules, but he also co-owns Apache-Test distro's modules. 
So we do have a trouble.

> As for mp2, is the following what you're referring to, Stas?
> Right now mod_perl 2 isn't indexed by PAUSE, due to _ in the
> distribution. But when it's officially released, there'll be
> likely problems with the PAUSE indexer - mod_perl 2 contains
> Apache::Connection, Apache::Filter, Apache::Module, and
> Apache::SubProcess, all of which appear (and are recognized as
> such by PAUSE) in different CPAN packages (this actually presents
> a potential confusing point now - if one has mod_perl 2
> installed, then CPAN.pm recommends, eg, upgrading Apache::Filter
> to the mp1 Apache-Filter package ...). However, when mod_perl 2
> gets officially released and is indexed, what might happen is
> that PAUSE will refuse to index mp2's Apache::Filter, as that's
> owned by Ken Williams, while the others (who are owned by Doug)
> will get indexed, and those with the higher version will become
> the recommended package ...

Indeed, this is a big mess. Though I have met big walls of refusal to do 
anything about making CPAN/PAUSE more flexible while discussing those issues 
on p5p, because it requires lots of changes. Now you raise yet another problem 
with indexing :(

__________________________________________________________________
Stas Bekman            JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/     mod_perl Guide ---> http://perl.apache.org
mailto:stas@stason.org http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org   http://ticketmaster.com


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@perl.apache.org
For additional commands, e-mail: dev-help@perl.apache.org


Mime
View raw message