Hi,
Thanks for the review.
On 06/18/2010 01:13 AM, Brian C. Lane wrote:
-----BEGIN PGP SIGNED MESSAGE-----
It looks good to me, with a couple of comments.
Some of the module references are getting ridiculously long:
pyanaconda.storage.fcoe.has_fcoe()
should ideally be fcoe.has_fcoe(), or at most storage.fcoe.has_fcoe()
(that's just one example).
I see that we have some pre-existing dotted notation relative imports in
the code:
from ..errors import *
It would make sense to me to clean these up to match importing from
pyanaconda.whatever instead so that things are consistent.
Yeah, good point. The goal of my change is just to make things work
again, but I'd encourage the guys who feel most responsible for such
modules to make the imports consistent the next time they're touching
the files. But let's just get this working now before setting the
conventions (if we do that I vote for the absolute imports, they're
quicker to be deciphered).
(and as an unrelated nit, I really dislike import *, it makes it really
hard to trace something back to its source, but I don't see that
changing anytime soon)
Love them too. My favorite one is 'from time import *' in timezone_text.py.
3) fix the problem of updates.img having to include everything
Is having it all really a big problem? It isn't that big, and it is alot
less complication to just drop it all in instead of trying to mix with
existing modules.
Possibly. When we do hotspot fixes just to get over a particular problem
in some build you want to give them just one file so it's guaranteed
nothing else is changed and you know your fix works. This also allows
you to combine two fixes in one updates.img just by merging their contents.
Anyway---the remedy is easy and I have a patch ready, will post it the
next week.
Ales
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
