PR checks failing

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

PR checks failing

Tamas Papp
Hi,

I recently made a trivial PR to DataFrames.jl, #916. I thought that
since tests ran OK, it would be fine, but on Github it shows that some
checks were not successful, in particular

continuous-integration/travis-ci/pr — The Travis CI build failed
coverage/coveralls — Coverage decreased (-0.5%) to 80.902%

I have to admit that I have no clue about what these mean, and how to
start fixing them. Is there a tutorial/introduction to the process?

Best,

Tamas

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.
Reply | Threaded
Open this post in threaded view
|

Re: PR checks failing

Milan Bouchet-Valat
Le mercredi 17 février 2016 à 08:29 +0100, Tamas Papp a écrit :

> Hi,
>
> I recently made a trivial PR to DataFrames.jl, #916. I thought that
> since tests ran OK, it would be fine, but on Github it shows that
> some checks were not successful, in particular
>
> continuous-integration/travis-ci/pr — The Travis CI build failed
> coverage/coveralls — Coverage decreased (-0.5%) to 80.902%
>
> I have to admit that I have no clue about what these mean, and how to
> start fixing them. Is there a tutorial/introduction to the process?
In general, if Travis or AppVeyor fail, you need to look at the log to
see where the error comes from. If tests pass on Julia 0.4 but not on
the development branch (to-be 0.5), this might indicate that recent
changes in Julia broke the master branch of the package, in which case
you should mention this in the PR (you can also try to fix it in
another PR if you feel like it).

Coveralls is a different story: it turns red when the proportion of the
code base covered by tests goes down. But this may not necessarily
reflect a real problem: if you rewrite a long function into a shorter
one, the number of covered lines may be lower, but this is actually a
good thing. So if Coveralls complains, simply check that the lines you
introduced (if any) are covered by tests. You can click on the link and
find the relevant file to check that.


Regards

--
You received this message because you are subscribed to the Google Groups "julia-stats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/d/optout.