2011/2/16 Jürgen Hermann <[email protected]>
> > My task is basically to find all python scripts in the source repository
> and run pylint on each.
>
> The best way to do that is to make your scripts (almost) look like any
> other module in your package, and then either start them via "python -m
> yourpkg.scripts.foo", or use "console_scripts" entry points (which
> conveniently causes setuptools/distribute to auto-generate wrapper scripts
> for the target platform).
>
I don't quite get what you mean here. Currently in the source tree there are
a few places where there are dirs that are python modules with submodules
and classes. In some other places, there are python scripts that lie on its
own as a one-off script. Do you mean I make those one-off scripts like any
other module? How can I do that. Ideally I would like to run pylint without
making any edits. Meanwhile, let me research further on these topics.
Currently, I do this and seems to have eliminated some of the import errors:
targetfile = args[0]
targetfiledir = os.path.dirname(targetfile)
if dir not in sys.path
sys.path.insert(1, dir)
try:
lint.Run(args)
except OSError, e:
It can also be, perhaps, some python codes are obsolete and import modules
that are not there so I am talking to the code owners to find out. Thanks
for the help
> For pylint the scripts then ARE like any other module. Just make sure you
> follow the __name__ == "__main__" idiom, i.e. have your scripts importable
> and w/o any top-level logic.
> _______________________________________________
> Python-Projects mailing list
> [email protected]
> http://lists.logilab.org/mailman/listinfo/python-projects
>
_______________________________________________
Python-Projects mailing list
[email protected]
http://lists.logilab.org/mailman/listinfo/python-projects
