Jeffrey C. Ollie wrote:
On Mon, 2006-02-06 at 09:12 -0800, Jesse Keating wrote:
>On Mon, 2006-02-06 at 09:03 -0800, Steve G wrote:
>
>>Out of curiosity, when the rebuild occurs...does the script use an alphabetical
>>listing of the packages from a to z. Or does the script try to build the packages
>>in a somewhat ordered fashion from no dependencies (attr, zlib, bzip, etc) to
>>more complicated packages that have many dependencies (mkinitrd, pam, slang,
>>etc.)
>
>Mostly just an a to z thing. The way our current build system works,
>every package is installed into the build root, so at this point we're
>pretty assured that all build reqs will be met. This is not optimal for
>many reasons, and a replacement build system is being developed that
>fixes this.
I know that yum/mock/plague doesn't quite fit the bill as a beehive
replacement, but it would be nice if RedHat could work with the
community to extend yum/mock/plague rather than re-inventing the wheel.
I know that one thing that plague could use is a way to easily switch
the plague client between two or more plague servers.
s/RedHat/Red Hat/
Would it make that much of a difference though, or would people still
say the same thing? I think even if Red Hat devoted 2 or more
employees working full time on mock/plague that people would still
say the same thing. Why do I think that? Well for starters, Red Hat
has contributed quite a lot to plague already from what I understand.
I don't know the level of contribution to mock, but it gets heavy
usage internally by individuals, and our next generation buildsystem
is aparently based on mock/plague. What more exactly are you looking
for? I suppose Red Hat could in theory hire every single person who
has contributed to either project... ;o) But then there would
be conspiracy theories about something else right? ;o)
--
Mike A. Harris * Open Source Advocate *
http://mharris.ca
Proud Canadian.