On 09/06/2017 08:26 AM, Guido van Rossum wrote:
So we've seen a real use case for __class__ assignment: deprecating
things on access. That use case could also be solved if modules
natively supported defining __getattr__ (with the same "only used if
attribute not found otherwise" semantics as it has on classes), but it
couldn't be solved using @property (or at least it would be quite hacky).
I guess it's a matter of perspective. I solved this problem using
@property, and I don't think it's particularly hacky. (See my
implementation at the bottom of this email). The worst thing it does is
look up the current module via sys.modules[__name__]... which
Nathaniel's code also must do.
My example is a lot simpler than Nathaniel's code but it's just a
proof-of-concept. Nathaniel's code does much more. In particular, I
didn't override __dir__ to hide the deprecated attributes; doing that
would mean assigning to __class__ just as Nathaniel does. (If you're
actually curious to see it I could add that to the proof-of-concept.)
IMO my PEP strikes the right balance. @property is far more popular
than __getattr__ or __getattribute__; 19 times out of 20, people use
@property. Meanwhile, people for whom @property is insufficient (like
Nathaniel) can already assign to module.__class__ and override
__getattr__, __getattribute__, __del__, or any other existing magic
method. In other words: the commonplace is easy, and the unusual is
possible. Perfect!
//arry
-----
/test_deprecated.py:
import depmod
print(depmod.depr1) # throws an exception
depmod.py:
# module with deprecated properties
import sys
_deprecated_properies = (
("depr1", 33),
("depr2", 44),
)
__module__ = sys.modules[__name__]
def set_deprecated_property(name, value):
@property
def prop(self):
raise RuntimeError(f"property '{name}' is deprecated")
return value
setattr(__module__, name, prop)
for name, value in _deprecated_properies:
set_deprecated_property(name, value)
_______________________________________________
Python-Dev mailing list
Python-Dev@python.org
https://mail.python.org/mailman/listinfo/python-dev
Unsubscribe:
https://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com
