Dne 17.2.2014 16:17, Chris Evich napsal(a):
On 02/17/2014 09:29 AM, Jiri Zupka wrote:
Hi,
I like this idea. Checker is not for us. It is for all autotest developers
which
tries to send patch to github. In my opinion it is good way how to inform
developers
about common problems in theirs code without long waiting. It could be more
confortable for them.
Have a nice day,
Jiří Župka
I'm confused, how isn't requiring they running check_patch.py first
enough? It seems obvious enough to me (when reviewing) whether or not
the basic checks were done or not (or ignored). Is there something
_not_ being checked in check_patch.py that could be?
Hi Chris,
this is exactly my point. Everyone who sends the patch can check his
patch by check_patch. He can (and probably should) run his test to see
the results.
The same with reviewer. It's up to reviewer to check it doesn't break
master. It can happened, but in an odd matter which wouldn't be
detectable by boot test (eg. windows 2007 installation vs 2003).
If we found ways to detect these easily, we can add these directly into
check_patch and voila the responsibility is right on the developer.
The only income I see is that you can see the possible conflict with
other pull request and you have automatically created next with all
workable pull requests applied. For me this is not enough, but maybe I'm
wrong. It was requested by Paolo and Eduardo, guys, please, would you
tell us your point or will you join our Thursday's meeting?
Regards,
Lukáš
_______________________________________________
Virt-test-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/virt-test-devel
