incremental ports/INDEX builder

[Oliver Eikemeier]

Kris Kennaway wrote:

> On Tue, Jun 22, 2004 at 12:28:19PM +0200, Oliver Eikemeier wrote:
> 
>>Some ports include a Makefile.local that isn't there. Of course
>>they won't make it into CVS, but you can have them in local port
>>trees. I just asserted that examples can be constructed where the
>>make(1) approach fails, too.
> 
> That's OK, I'm not trying to solve the halting problem here, only deal
> with the dependencies in the CVS ports collection.  My point was that
> an implementation that only works for less than 100% of the changes
> that are actually made to the ports collection (excluding the
> pathological cases that don't and aren't going to occur) isn't a
> solution I can use.  That amounts to tracking .included files and
> updating that dependency list when it changes, since those are the
> corner cases that a straightforward implementation doesn't catch.

I hope that my implementation is correct in the changes I've seen so
far. I'm interested on integrating it in a build cluster that does
check which packages may be directly affected and generates a blame
log (done in this script, packges affected indirectly, e.g. via build
dependencies should be done in a second step), does INDEx builds,
version checks and alike and delivers the output to a queue on the
package building cluster. The generated dependency file could also be
used for portlint, to automatically check affected ports along with
the port currently checked. I guess it could be advantageous where
we able to break INDEX generation down to a modular process with
defined interfaces.

Just my 2 cents
-Oliver