[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Fedora-packaging] PHP guidelines



On Wed, 2006-07-26 at 16:20 -0700, Christopher Stone wrote:
> On 7/26/06, Toshio Kuratomi <toshio tiki-lounge com> wrote:
> > On Wed, 2006-07-26 at 15:32 -0700, Christopher Stone wrote:
> > > On 7/26/06, Toshio Kuratomi <toshio tiki-lounge com> wrote:
> > > > Read again what I wrote -- it's not about a feature that's in php
> > > > (well, actually a feature of the web server) "?> ", it's about a
> > > > hypothetical bug discovery process involving "<?php ".
> > >
> > > So should we change all php source files to use "<?php" or should you
> > > fix php to accept "<?php ".  Then ask yourself, should we fix all spec
> > > files to add a %build, or should we fix redhat-rpm-config?
> >
> > Yep.  We fix redhat-rpm-config.  Then the next time we run across
> > something unexpected happens we break packages again.  Then we fix it
> > again.  Then it breaks again.....
> >
> > We're supposed to be promoting good packaging practice here.  If we know
> > that something as simple as including %build in your spec is a way to
> > isolate your package from some subset of problematic special cases, then
> > we should do that.
> 
> Okay, so you are saying it is better to hide or mask problems rather
> than fix them? 

Nope.  We should try not to purposefully stick our hand in any fires.
If we find a problem, it should be fixed, but promoting practices that
we know risk triggering bugs when there are simple, straightforward, and
clean ways to code it instead is just good sense. 

>  I'm sorry, but I disagree.  Probably the reason things
> are so hacky as they are now is because we have been hiding problems
> rather than fixing them.

Nah.  They're hacky because bugzilla allows jbj to close rpm bugs
WONTFIX :-)

-Toshio

Attachment: signature.asc
Description: This is a digitally signed message part


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]