Re: [Condor-devel] What does Condor _really_ depend on?

[Erik Paulson <epaulson@xxxxxxxxxxx>]

On Mon, Jul 30, 2007 at 10:00:19PM -0700, Derek Wright wrote:
> 
> On Jul 30, 2007, at 7:29 PM, Erik Paulson wrote:
> 
> >I don't think it's really clear how it's supposed to work. I like  
> >the idea of
> >just using HAVE_EXT_<foo>, combined with a real --with-foo and -- 
> >without-foo
> >from autoconf, and getting rid of the HAVE && WANT tests. I think  
> >turning
> >off externals is better than turning off "features."
> 
> Right.  I've been trying to move us towards HAVE_EXT_<foo>.  Not  
> everyone plays along with that.  Vast areas of the build system are  
> using other, older systems for the same basic thing.  It's no  
> surprise that when people sit down to add something new, they look  
> around, are confused at the diversity of existing practice, shoot in  
> the dark, and miss.
> 
> http://www.cs.wisc.edu/condor/developers/GENERIC/HOWTO-add- 
> externals.html
> 
> Unfortunately, that isn't as clear about this as it should be,  
> probably because I originally wrote it before Imake was including  
> config.h, which was when I really started to get clear in my own head  
> that we should formalize on HAVE_EXT_<foo>.
> 

One very important thing that needs to be defined is what should

./configure; make

build? Currently, as a fail-safe, if ANY of the externals are missing the
build fails - this helps us make sure the nightly build/tests produce
builds with everything we want to release. However, at every other site 
than the UW, that's probably not what is really wanted. If the build
can't find the Nordugrid GAHP, for example, the build shouldn't fail. 

-Erik