You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In 2.6 a deprecation warning was added if object.__new__ was called
with arguments. Per se this is fine, but the detection seems to be faulty.
The following code shows the problem:
>>> classA(object):
... def__new__(self):
... raiseTypeError('i do not exist')
...
>>> classB(A):
... __new__=object.__new__
... def__init__(self, x):
... self.x = x
...
>>> B(1)
__main__:1: DeprecationWarning: object.__new__() takes no parameters
<__main__.B object at 0x88dd0>
In the B case __new__ is not overridden (in the sense that it
differs from object.new) but __init__ is. Which is the default
behaviour. Nonetheless a warning is raised.
I used the pattern with the "__new__ switch" to achieve a
cStringIO.StringIO behavior that supports typechecks: IterIO() returns
either a IterI or IterO object, both instances of IterIO so that
typechecks can be performed.
The problem seems to be caused by tp_new being slot_tp_new which then
invokes whatever __new__ in the class dict is.
I'm not so sure what would be the solution to this. One could of course
check if tp_new is either object_new or slot_tp_new and in the latter
case check if the class dict's __new__ item is object_new...
Trundle commented on Mar 24, 2009
Trundlemannequin
Mannequin
More actions
I think the real problem here is update_one_slot and not object_new. It
is impossible to set "new" to a PyCFunction inside Python code, which
may be a feature, but is in fact very irritating.
I would rather expect this behaviour (or at least that Dict.__new__ is not
object.__new__):
>>> Dict.__new__isobject.__new__
True
>>> Dict()
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: object.__new__(Dict) is not safe, use dict.__new__()
The attached patch leads to that behaviour, which also fixes the argument
calling autodetection of object.__new__.
The problem is that type_setattro() sets the new "new" attribute
in the type's dict (through PyObject_GenericSetAttr()), but the
corresponding slot will never be updated if the new "new" is a
PyCFunction.
The affected code in update_one_slot() was added by Guido van Rossum
in r28090, so maybe he would like to comment on that.
Yes, it was what the patch did by setting tp_new to slot_tp_new. The problem is that the same code is used for inherited __new__ and assigned in class body. It is hard to distinguish between these cases.
In any case I think that Cython shouldn't generate trivial __new__. This will help to change the order of __new__ resolution at least in 3.7.
It worries me that nothing in the Python docs nor in any PEP describes how tp_new is inherited. In my opinion, this patch makes a significant change which should be subject to a PEP. However, neither the old nor new behaviour is described anywhere. This also makes it harder to argue which behaviour is correct.
The bug is still there, just that it's now not just a warning but an error. The auto detection is incorrect here. It should allow the instantiation of the object with arguments.
Activity
mitsuhiko commentedon Feb 19, 2009
In 2.6 a deprecation warning was added if
object.__new__was calledwith arguments. Per se this is fine, but the detection seems to be faulty.
The following code shows the problem:
In the
Bcase__new__is not overridden (in the sense that itdiffers from object.new) but
__init__is. Which is the defaultbehaviour. Nonetheless a warning is raised.
I used the pattern with the "__new__ switch" to achieve a
cStringIO.StringIO behavior that supports typechecks: IterIO() returns
either a IterI or IterO object, both instances of IterIO so that
typechecks can be performed.
Real-world use case here:
http://dev.pocoo.org/projects/werkzeug/browser/werkzeug/contrib/iterio.py
mitsuhiko commentedon Feb 19, 2009
The problem seems to be caused by tp_new being slot_tp_new which then
invokes whatever __new__ in the class dict is.
I'm not so sure what would be the solution to this. One could of course
check if tp_new is either object_new or slot_tp_new and in the latter
case check if the class dict's __new__ item is object_new...
Trundle commentedon Mar 24, 2009
I think the real problem here is
update_one_slotand notobject_new. Itis impossible to set "new" to a PyCFunction inside Python code, which
may be a feature, but is in fact very irritating.
For example the following snippet:
I would rather expect this behaviour (or at least that Dict.__new__ is not
object.__new__):
The attached patch leads to that behaviour, which also fixes the argument
calling autodetection of
object.__new__.pitrou commentedon Apr 9, 2009
I'm sorry, I don't have any opinion on this.
Trundle commentedon Apr 18, 2009
The problem is that
type_setattro()sets the new "new" attributein the type's dict (through
PyObject_GenericSetAttr()), but thecorresponding slot will never be updated if the new "new" is a
PyCFunction.
The affected code in
update_one_slot()was added by Guido van Rossumin r28090, so maybe he would like to comment on that.
Trundle commentedon Apr 27, 2009
See also issue bpo-1694663.
benjaminp commentedon Oct 10, 2011
I think it needs tests.
serhiy-storchaka commentedon Nov 17, 2016
Here are updated patches with tests for 3.x and 2.7.
python-dev commentedon Dec 2, 2016
New changeset a37cc3d926ec by Serhiy Storchaka in branch '2.7':
Issue bpo-5322: Fixed setting __new__ to a PyCFunction inside Python code.
https://hg.python.org/cpython/rev/a37cc3d926ec
15 remaining items
python-dev commentedon Dec 14, 2016
New changeset f89ef18f9824 by Serhiy Storchaka in branch '2.7':
Issue bpo-5322: Restored tests for __new__.
https://hg.python.org/cpython/rev/f89ef18f9824
New changeset 06e4b9f2e4b0 by Serhiy Storchaka in branch '3.5':
Revert changeset 1f31bf3f76f5 (bpo-5322) except tests.
https://hg.python.org/cpython/rev/06e4b9f2e4b0
benjaminp commentedon Dec 15, 2016
BTW, at least for bpo-25731, I think the right approach in the MI case is to synthesize a __new__ on the subclass that calls the solid base __new__.
serhiy-storchaka commentedon Dec 15, 2016
Yes, it was what the patch did by setting tp_new to slot_tp_new. The problem is that the same code is used for inherited __new__ and assigned in class body. It is hard to distinguish between these cases.
In any case I think that Cython shouldn't generate trivial __new__. This will help to change the order of __new__ resolution at least in 3.7.
jdemeyer commentedon Jan 4, 2017
It worries me that nothing in the Python docs nor in any PEP describes how tp_new is inherited. In my opinion, this patch makes a significant change which should be subject to a PEP. However, neither the old nor new behaviour is described anywhere. This also makes it harder to argue which behaviour is correct.
[-]Python 2.6 object.__new__ argument calling autodetection faulty[/-][+]object.__new__ argument calling autodetection faulty[/+]iritkatriel commentedon Jan 24, 2022
Can we close this now?
mitsuhiko commentedon Jan 24, 2022
The bug is still there, just that it's now not just a warning but an error. The auto detection is incorrect here. It should allow the instantiation of the object with arguments.
iritkatriel commentedon Jan 24, 2022
Right, I see now.
vstinner commentedon Jan 25, 2022
Python has the same behavior since Python 2.6. While it annoys a few persons, the majority doesn't care. I suggest to close the issue.
It's easy to workaround limitation that object.__new__() which doesn't accept arguments.