You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@perl.apache.org by "Philippe M. Chiasson" <go...@cpan.org> on 2003/06/20 10:46:49 UTC

Re: [mp2 Patch] Simple changes to current Apache::Reload for mp1 & mp2

On Thu, 2003-06-19 at 12:52, Stas Bekman wrote:
> Philippe M. Chiasson wrote:
> > On Thu, 2003-06-19 at 12:22, Stas Bekman wrote:
> > 
> >>Philippe M. Chiasson wrote:
> >>
> >>>I needed this simple patch to get Apache::Reload working in mp1.
> >>>
> >>>Any objections to at least applying this so it is ready for possible
> >>>inclusion in mp1's upcoming release?
> >>>
> >>>Of course, if Steve Hay's suggestions make it through, keeping the
> >>>backwards compatibility would become important as well.
> >>>
> >>>Gozer out.
> >>
> >>you obviously haven't tested it with perl 5.005. what about the 'warnings' pragma?
> > 
> > 
> > Oupss, my bad ;-( But that's quite trivially fixable, isn't it?
> 
> either if (MP2) require warnings; warnings->import/unimport. else local $^W...

I know it's a bit messy, but it has more to do with supporting different
versions of Perl, not mod_perl, isn't it ? I wouldn't minding seeing
this kinda cruft at the top of it.
BEGIN {
if ($] > 5.006) {
	use warnings et all	
}
else {
	$^W ...	
}
> or do similar to Apache::Test5005compat, implementing a dummy warnings module. 
> However this won't quite work since we do want to  local $^W = 0 on reload 
> where the warnings pragma isn't available.
> 
> >>I think Apache::Reload is simple enough to have two copies of it, adjusted for 
> >>each mod_perl generation.
> > 
> > 
> > It's also quite simple enough to have one copy handle mp1 and mp2 with
> > only a few lines of code. Or is that a "bad thing" ?
> 
> Not a bad thing, just a bit of a mess.

Really minimal mess IMHO. Worthy mess even, for the sake of laziness ;-)

> >>If we "hijack" Matt's distro and keep it as a separate module on CPAN, only 
> >>syncing modperl and modperl-2.0 trees, then it's a whole different story. but 
> >>again we hit this problem with PAUSE still not quite supporting bundling :(
> >>(i.e. we don't want CPAN.pm to try to install mod_perl, when someone is trying 
> >>to install Apache::Reload)
> > 
> > 
> > That's indeed a problem with CPAN ;-( Then why not simply make sure
> > Apache::Reload works with both mod_perl generations and maintain it on
> > CPAN? As long as there is only one version of it and that it works with
> > mp1 and mp2, I don't see what the problem could be.
> > 
> > perl -MCPAN -e 'install Apache::Reload'
> > 
> > Another problem I can see is someone having a dual setup with both mp1
> > and mp2. Shoulb be able to install Apache::Reload once in site_perl, not
> > twice in mp specific PREFIXes ...
> 
> Currently this problem doesn't exist. If we make it live on CPAN we have to 
> make sure that it doesn't go into Apache2/, that's it. It doesn't matter if it 
> goes to site_perl or else otherwise.

That's right.

> >>Proposal: keep the two separate for now, just make sure that the versions are 
> >>different. For example 1.00 for mp1 and 2.00 for mp2. For the next release(s) 
> >>hopefully PAUSE will support bundles and then we can decide how to act.
> > 
> > 
> > Proposal: Just take Apache::Reload off mp2 and released on CPAN.
> 
> <inertia mode>but Apache::StatINC was always in mp1. Apache::Reload is just a 
> better replacement</inertia mode>
> 

Guess so. In that case, I'd say keep Apache::Reload in mp2, add it to
mp1 as well, but with a lower version number. But keep them identical
where practical.

> 
> __________________________________________________________________
> 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
-- 
--------------------------------------------------------------------------------
Philippe M. Chiasson /gozer\@(cpan|ectoplasm)\.org/ 88C3A5A5 (122FF51B/C634E37B)
http://gozer.ectoplasm.org/    F9BF E0C2 480E 7680 1AE5 3631 CB32 A107 88C3 A5A5
Q: It is impossible to make anything foolproof because fools are so ingenious.
perl -e'$$=\${gozer};{$_=unpack(P7,pack(L,$$));/^JAm_pH\n$/&&print||$$++&&redo}'