Hello PyLint Developers!
I have a suggestion for something that could be done better. It may be
the way Python works and PyLint can't do anything about it, in which
case this is just a rant from a frustrated developer (sorry). This
import error message from PyLint (I now realize Python produces a
similar one) caused me to waste a lot of time:
No name 'a' in module 'module_a'
Setup:
-module_a
-a.py
-b.py
-__init__.py
b.py:
from module_a.a import my_class
a.py contains a syntax error.
$ pylint module_a/b.py
No config file found, using default configuration
************* Module module_a.b
C: 1: Missing docstring
E: 1: No name 'a' in module 'module_a'
W: 1: Unused import my_class
Which I took to mean that there was no file a.py inside the directory
"module_a" or some problem with the filesystem or the names... The
problem was quite different (a syntax error in a.py). The Python error
message is just as misleading. I was scrambling for quite a while
trying to find out what was wrong in the filesystem or the names I had
given to the files.
I think that a Python syntax error inside a file should generate
something more appropriate and helpful than something that sounds like
the claim that there is no such file! I understand now that the syntax
error prevented the name from appearing in the namespace, so from that
oblique point of view the error message is correct. However, why on
earth not report syntax errors as such? I found no option to increase
the verbosity of PyLint.
Thanks for maintaining PyLint.
Regards,
Pascal Meunier
_______________________________________________
Python-Projects mailing list
Python-Projects@lists.logilab.org
http://lists.logilab.org/mailman/listinfo/python-projects
