On Fri, Aug 31, 2012 at 10:04 AM, Simon King <[email protected]> wrote:
> Hi!

Hi. Yeah, I've been making a lot of improvements over the last couple
of days. Glad to see someone noticed :).

> At #13370, I just noticed that the patchbot isn't even running the
> doctest, because plugins.coverage complains. I think it is a bad idea to
> not run the tests just because plugins.commit_messages or
> plugins.coverage or plugins.trailing_whitespace fails.

Yes, very true. This particular patchbot is running in a "plugin only"
mode for testing and quick coverage; the normal behavior has not
changed.

> In addition, I don't understand why the plugin is complaining: All
> functions introduced by my patch have a test, and I even add a test that
> refers to the ticket (showin that some problem is fixed). So, what
> criteria are to meet for making the plugin happy?

Looking at 
http://patchbot.sagemath.org/log/13370/debian/lenny/sid/x86_64/2.6.24-28-server/sage.math.washington.edu/2012-08-31%2008:05:43%20-0700?plugin=plugins.coverage&diff=/log/0/debian/lenny/sid/x86_64/2.6.24-28-server/sage.math.washington.edu/2012-08-31%2003%3A35%3A10%20-0700&ticket=13370&base=5.3.rc0
, I see

Decreased doctests misc/cachefunc.pyx from 56 / 63 = 88% to 60 / 68 = 88%

which indicates that there were formerly 7 undoctested functions in
this file and now there are 8. Now it's quite possible that sage
-coverageall has a bug in it, which is beyond the scope of the
patchbot.

Also, unhappy plugins shouldn't be a blocker, rather they should be
something the reviewer should at least be aware of (which may of
course indicate a blocker, if there are for example new undoctested
functions).

- Robert

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
Visit this group at http://groups.google.com/group/sage-devel?hl=en.


Reply via email to