httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rasmus Lerdorf <ras...@raleigh.ibm.com>
Subject Re: Apache 2000, err Apache 2.0 gets real
Date Sun, 01 Aug 1999 14:12:56 GMT
> That's why I appreciate that we don't try a quick "translate all to Autoconf"
> for Apache 2.0.  Instead especially for the module config and build process
> and DSO situations we should create a well-considered environment.  And I'm
> convinced Autoconf is not all we need here. I'm even convinced that the
> "configure" script(s) should be not directly the ones generated by Autoconf!
> Instead the Autoconf scripts should be run in a special environment in the
> background of our own "configure" scripts.  But ok, it's too early for those
> details...

This sounds overly complicated to me.  One of the nice things about a
straight autoconf setup is that when you realize you are missing a test
for something, chances are someone out there has already written it.  It
should not be made unneccesarily complex for people to tweak the autoconf
stuff and contribute the changes back.  The current
./configure-src/Configure-src/support/apxs.pl-apxs-make install chain
effectively produces 4-levels of indirection and is nearly impossible to
grok unless you are really stubborn and determined to figure it out.

Obviously an automake/autoconf setup has many levels as well, but at least
it is a known sequence.  People familiar with the environment can dive
right in and be productive right away.  I would really like to avoid
adding extrananeous levels here.  Like you asked Jim, can you give an
example of something that can't be done within the normal confines of an
autoconf/automake setup that would require this additional step?

-Rasmus


Mime
View raw message