On Fri, Mar 29, 2013 at 6:30 PM, Sylvain Thénault <
sylvain.thena...@logilab.fr> wrote:
> Hi Torsten,
>
> On 19 mars 22:15, Torsten Marek wrote:
> > A collection of bugfixes and new warnings. No large changesets, the most
> > interesting one is definitely warning_scopes.diff.
>
> nice series anyway. I've taken a look at it, do some minor modification.
> See
> comments below. All have my blessing, though I've pushed them all to our
> review process anyway so others have the opportunity to do some remark,
> catch
> typos and all. I'll probably handle remarks if any by myself.
>
> > >> bad_str_strip.diff
> > Warn about suspicious arguments in {bytes,str,unicode}.{l,r,}strip calls.
> >
> > Closes #74013
>
> pushed after a slight modification: merged your checker with former
> string_format,
> renamed for the occasion into 'string' so the message is now E1310 instead
> of W1400,
> 14 being occupied by the string_constant checker and casted to an Error as
> even if
> it won't cause a crash, it's definitly something that should be fixed.
>
> I've also moved this string_constant from format.py to the renamed as well
> strings.py
> module. IMO we could also unify checker number and message here, but I
> have not
> bothered with that yet as it may break configurations.
>
Sounds good.
> >> loop_break.diff
> > Emit a warning for loops that have an else clause but no break or
> return.
> >
> > Closes #81378
>
> I would vote to kill the special case for return which is IMO highly
> arguable.
> I feel like the rule should be the same for all, and if one feel like it's
> more readable with the else clause anyway, he should add the pragma to
> allow
> it explicitly .
>
Ok. It looks like this is already in the repository (though I'm a bit
confused wrt hg.logilab.org/review vs hg.logilab.org/master), should I
modify the old patch or create a new one to change the warning and the code?
>
> Also, I would vote for using a Warning rather than a Convention.
>
I don't mind either way.
>
> Side note for the future: I've renamed newly introduced test files to use
> the
> message name rather than the message id, has it's more readable and less
> subject
> to change.
>
Good idea.
// Torsten
_______________________________________________
Python-Projects mailing list
Python-Projects@lists.logilab.org
http://lists.logilab.org/mailman/listinfo/python-projects
