← Back to team overview

nova team mailing list archive

Re: pylint and methods that could be functions


On 08/10/2010 04:26 PM, Chris Behrens wrote:
> All,
> What's the preference on methods that could be functions?   pylint is
> currently complaining on any class method that doesn't access 'self',
> meaning that the method doesn't necessarily need to be a part of a
> class.   I'm going back and forth on it.  I tend to think it makes
> sense to keep some things as a part of a class for organizational
> purposes... but I don't know.

I like inclusion for org purposes... but:

> Some examples are in nova/compute/model.py's InstanceDirectory.  Most
> of the methods don't need to be a part of the class, but I can see
> where it's nice to organize it all that way...   If we want to keep
> it, I think we may want to make pylint ignore this code?

Or perhaps just remove the self arg and make it a static method:

class Foo:

  def bar(a, b):
    # isn't really part of the class, but is organizationally.


Follow ups