On 31Oct2018 21:23, Marko Rauhamaa <ma...@pacujo.net> wrote:
Peter Otten <__pete...@web.de>:
Do not make changes to your code only to apeace a linter.
Precisely. Don't let a tool drive a show.
While this is sound for semantic changes (and a function versus a
@staticmethod is verging on that), I would point out that cleaning lint
is often useful in order to make the linter useful.
If the useful lint warnings are buried in a sea of dross, and you want
to lint things, then it can be well worth making some changes to clean
up noise.
Ideally of course the linter should recognise some hooks (typically
special comments) to mark particular pieces of code as deliberately not
matching the linter's opinions. And certainly I run some linters with
various checks tuned or disabled.
I'm doing this on an ongoing basis in my libraries as I work on other
changes: make the change, debug, then lint, with the linting in distinct
commits unless it is tiny. This is gardually bringing consistency to the
code and catching the odd bug or rough edge such as "variable set but
not used".
The end game here is that your lint output should be empty so that when
it isn't empty the messages are useful.
Cheers,
Cameron Simpson <c...@cskk.id.au>
--
https://mail.python.org/mailman/listinfo/python-list
