[cheesecake-users] pylint curiousness

Grig Gheorghiu grig.gheorghiu at gmail.com
Mon Jul 10 20:56:53 PDT 2006


On 7/10/06, will guaraldi <willg at bluesock.org> wrote:
>
>
> PyBlosxom has a package hierarchy like this:
>
>     Pyblosxom/
>        - cache/
>        - entries/
>        - renderers/
>
> Modules in the "entries" directory import modules in the "Pyblosxom"
> directory.  So, for example, there's a module "entryshelve" in entries and
> a module "tools" in Pyblosxom and entryshelve imports tools like this:
>
>     from Pyblosxom import tools
>
> This works fine when running pylint all by itself on the entire Pyblosxom
> package but I see massive penalties on individual files when run in
> cheesecake because cheesecake runs pylint on individual files.  This is
> hidden by the fact that cheesecake ignores negative pylint scores.
>
>
> I did some poking around in pylint to see if I could add to the sys.path
> or something like that through the command-line arguments, but nothing
> jumped out at me.
>
> For PyBlosxom, the pylint average changes from 8.21 (running pylint
> standalone) to 7.16 (pylint from cheesecake).  My issue is that the files
> that have this sort of importing don't get reflected in the KWALITEE
> score, but they should be.
>
> Any thoughts on how that behavior could get changed/fixed?
>
> /will


Will,

Thanks for reporting this issue. It does seem like the sane approach for
cheesecake should be to run pylint on the package as a whole, as opposed to
the current way of running it on individual files. We have a pylint-related
story coming up soon, so we'll add this task to the story.

Thanks again,

Grig
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists2.idyll.org/pipermail/cheesecake-users/attachments/20060710/6454746b/attachment.html 


More information about the cheesecake-users mailing list