On 18 avril 13:03, Claudiu Popa wrote:
> On Fri, Apr 18, 2014 at 12:09 PM, Sylvain Thénault
> <sylvain.thena...@logilab.fr> wrote:
> >> * 
> >> https://bitbucket.org/logilab/pylint/issue/75/wrong-default-for-dummy-variables
> >> * 
> >> https://bitbucket.org/logilab/pylint/issue/99/request-that-r-flag-default-to-n
> >> * 
> >> https://bitbucket.org/logilab/pylint/issue/171/pylint-does-not-install-by-easy_install
> >
> > No one on those? Here are my points:
> >
> > Regarding install (#171), the most important pb there, I would like to find 
> > a
> > solution that don't compile test data files so it don't scare people.
> >
> > For #99, I really don't known. I think that's true for usability reason, 
> > but I
> > fear this implies killing the whole feature. I'm tempted to keep them by 
> > default
> > and to advertise they can be easily disabled.
> >
> > #75, I propose to fix the doc rather than the regexp.
> 
> For #99 I think we should leave it as it is now. I know lots of people at work
> who look at first at the rating that Pylint give for their code and afterwards
> at the warnings and messages. We try to appeal to both experienced
> developers and beginners as well and having `-r` defaulting to `y` works 
> better
> for the latter group. There's no problem for experienced ones to
> modify their pylint rc file
> to disable the flag completely.
 
That's my point. Well stated. Do you mind rejecting the issue with the above
reason?

Regarding #171, the pb doesn't arise with bare distutils, seems related to
distutils/pip. I propose the following patch:

diff --git a/setup.py b/setup.py
--- a/setup.py
+++ b/setup.py
 
+    # override this since pip/easy_install attempt to byte compile test data
+    # files, some of them being syntactically wrong by design, and this scares
+    # the end-user
+    def byte_compile(self, files):
+        testdir = join('pylint', 'test')
+        files = [f for f in files if testdir not in f]
+        install_lib.install_lib.byte_compile(self, files)

Though setuptools break with a strange error on my box, so I can't really test
it works :( 

-- 
Sylvain Thénault, LOGILAB, Paris (01.45.32.03.12) - Toulouse (05.62.17.16.42)
Formations Python, Debian, Méth. Agiles: http://www.logilab.fr/formations
Développement logiciel sur mesure:       http://www.logilab.fr/services
CubicWeb, the semantic web framework:    http://www.cubicweb.org
_______________________________________________
Python-Projects mailing list
Python-Projects@lists.logilab.org
http://lists.logilab.org/mailman/listinfo/python-projects

Reply via email to