gh-90562: Support zero argument super with dataclasses when slots=True #124455
Conversation
… showing that the original class, if saved, will have in incorrect __class__ value.
There was a problem hiding this comment.
Thanks for working on this!
There is the possible case where all methods of a class are decorated with opaque decorators that don't use functools.wraps, and thus we have no route to find the cell to update, and thus this approach simply fails. Should this case be mentioned in the tests? In the docs? Would it be worth explicitly introducing a __classcell__ member on classes through which the cell can reliably be accessed, so as to avoid this problem?
| B = dataclass(slots=True)(A) | ||
| self.assertIs(B().cls(), B) | ||
|
|
||
| # This is probably undesirable behavior, but is a function of |
There was a problem hiding this comment.
I think we can safely remove this "probably" :)
Lib/dataclasses.py
Outdated
| _update_func_cell_for__class__(member, cls, newcls) | ||
| elif isinstance(member, property): | ||
| _update_func_cell_for__class__(member.fget, cls, newcls) | ||
| _update_func_cell_for__class__(member.fset, cls, newcls) | ||
| _update_func_cell_for__class__(member.fdel, cls, newcls) |
There was a problem hiding this comment.
I think technically we could break out of the loop as soon as we've successfully updated one cell, because all the closures for a given class will always share a single __class__ cell. This may improve performance noticeably for larger classes?
There was a problem hiding this comment.
Good point. Will change.
Lib/dataclasses.py
Outdated
| _update_func_cell_for__class__(member, cls, newcls) | ||
| elif isinstance(member, property): | ||
| _update_func_cell_for__class__(member.fget, cls, newcls) | ||
| _update_func_cell_for__class__(member.fset, cls, newcls) | ||
| _update_func_cell_for__class__(member.fdel, cls, newcls) |
There was a problem hiding this comment.
Good point. Will change.
| B = dataclass(slots=True)(A) | ||
| self.assertIs(B().cls(), B) | ||
|
|
||
| # This is probably undesirable behavior, but is a function of |
Apply review suggestions. Co-authored-by: Carl Meyer <[email protected]>
…loop; minor change to comments; fix a test.
📚 Documentation preview 📚: https://cpython-previews--124455.org.readthedocs.build/