On Tue, 09 Jan 2018 09:55:49 +0200, Frank Millman wrote:
> Hi all
>
> I have read that one should not call dunder methods in application code.
"Should not" rather than "must not", but that's correct. In general,
calling a dunder method directly is *not* the same as the operation you
probably want. For example, `x + y` may not be the same as `x.__add__(y)`
or `y.__radd__(x)`. Better to use `operator.add(x, y)` instead.
> Does the same apply to dunder variables? I am thinking of the instance
> attribute __dict__, which allows access to the contents of the instance.
>
> I only want to read from __dict__, not update it. Is this frowned upon?
If you are trying to look up obj.attribute where you don't know the name
of the attribute until runtime, the right way is to use getattr:
name = 'spam' # this is unknown until runtime
value = getattr(obj, name) # like obj.spam
If you want to avoid inheritance and avoid looking up attributes on the
class or any superclasses, but only look it up on the instance, then the
nice way is to use vars(obj):
vars(obj)['attribute']
But be aware that not all objects even have a __dict__ -- not even all
those with data attributes.
--
Steve
--
https://mail.python.org/mailman/listinfo/python-list
