gh-128184: Fix docstring generation in dataclasses with forward refs #128194
Conversation
| @@ -1164,6 +1164,13 @@ def _process_class(cls, init, repr, eq, order, unsafe_hash, frozen, | |||
| # In some cases fetching a signature is not possible. | |||
| # But, we surely should not fail in this case. | |||
| text_sig = str(inspect.signature(cls)).replace(' -> None', '') | |||
| except NameError: | |||
There was a problem hiding this comment.
Why not just start with SOURCE format? I don't think we need to catch NameError first.
There was a problem hiding this comment.
When we do that, a lot of tests fail with similar messages:
def test_docstring_two_fields(self):
@dataclass
class C:
x: int
y: int
self.assertDocStrEqual(C.__doc__, "C(x:int, y:int)")Produces:
======================================================================
FAIL: test_docstring_two_fields (test.test_dataclasses.TestDocString.test_docstring_two_fields)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/sobolev/Desktop/cpython2/Lib/test/test_dataclasses/__init__.py", line 2294, in test_docstring_two_fields
self.assertDocStrEqual(C.__doc__, "C(x:int, y:int)")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/sobolev/Desktop/cpython2/Lib/test/test_dataclasses/__init__.py", line 2263, in assertDocStrEqual
self.assertEqual(a.replace(' ', ''), b.replace(' ', ''))
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError: "C(x:'__dataclass_type_x__',y:'__dataclas[46 chars]e__'" != 'C(x:int,y:int)'
- C(x:'__dataclass_type_x__',y:'__dataclass_type_y__')->'__dataclass___init___return_type__'
+ C(x:int,y:int)
I am not quite sure - why.
There was a problem hiding this comment.
I tried that again and it still does not work without NameError handling.
This happens because of:
locals = {**{f'__dataclass_type_{f.name}__': f.type for f in fields},
**{'__dataclass_HAS_DEFAULT_FACTORY__': _HAS_DEFAULT_FACTORY,
'__dataclass_builtins_object__': object,
}
}
So, I am not sure that this is totally correct.
There was a problem hiding this comment.
@JelleZijlstra what do you think?
There was a problem hiding this comment.
I thought about this more and came up with a more general improvement: #130815. This also makes the docstring cleaner for dataclasses with no __init__ that contain unresolved names, and it improves other uses of inspect.signature. Please take a look.
__init__annotation eagerly evaluated #128184