Andrea,

are these rules based on coverage fraction or absolute lines covered? Is 
coverage relative to cyclomatic complexity?

-0 for rules that we never reduce code coverage. I can think of several 
cases where it is quite acceptable to reduce coverage fraction, such as 
fixing a leak by adding extra lines in a hard-to-cover catch block, or 
disabling a failing test class or method to fix the build and preserve 
the rest of the coverage. Given infinite time we can always add more 
tests ...

It might be quite useful to have a coverage change report for every pull 
request. There is no harm trying Coverall to see what it looks like. I 
am only against the rule that we never allow coverage to be reduced. It 
might be good to know if a module's coverage falls below 40%.

In my view, coverage metrics fall into the same category as formatting 
convention compliance and lack of compiler warnings: good to have but 
often a pain in the arse when enforced by automated tools.

Kind regards,
Ben.

On 19/09/15 19:13, Andrea Aime wrote:
> Hi,
> speaking with the OpenLayers guys this week they showed me how
> their pull requests check do not only include a travis build verification,
> but also a code coverage level check via Coveralls.
>
> Basically, their rule is that a pull request must never reduce the code
> coverage, and this can be automated via Travis and Coveralls.
>
> As far as I can see Coveralls is free for open source projects, what do you
> think of adding it to our pull requests checks?
>
> Cheers
> Andrea
>
>
>
> ------------------------------------------------------------------------------
>
>
>
> _______________________________________________
> GeoTools-Devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/geotools-devel
>

-- 
Ben Caradoc-Davies <[email protected]>
Director
Transient Software Limited <http://transient.nz/>
New Zealand

------------------------------------------------------------------------------
_______________________________________________
GeoTools-Devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to