seth vidal wrote:
Why did you check this in? I don't think we had finished discussing it.
I don't like this 'email a patch on day 1, if no comments say NOOOOO
then commit it' policy you seem to have taken up. I agree with what Tim
said yesterday that there's no compelling reason to put this in yummain
at all when we can, instead just create a separate yumprof.py which
duplicates yummain.py which we can use for running the profiling
routines instead.
Primary because that change doesn't affect the things still in question. It
just replaces a more or less useless, commented out profiling code by a more
useful, commented out profiling code.
So from my point of view it patch leaves just two questions:
1. Do we had enough discussion about that sys.exit -> return errcode
transition.
2. Where do we go from here?
To 1.: Yes, it got commited a bit quickly. But as more or less everybody was
online yesterday I had expected some loud shouting if this part of the patch
is a problem. If someone speaks up on this topic, please revert or remove
the patch.
To 2.: I don't have a strong opinion where to go from here. Either providing
a yumprof.py program or staying at "edit that damn file" is fine with me if
adding a hidden cli parameter is out of question.
My point is that it doesn't make sense to delay the patch because of
decisions that are not yet made but also not needed for applying that patch.
Florian
_______________________________________________
Yum-devel mailing list
[email protected]
https://lists.dulug.duke.edu/mailman/listinfo/yum-devel
