Message158444
| Author |
pitrou |
| Recipients |
benjamin.peterson, kristjan.jonsson, meador.inge, pitrou |
| Date |
2012年04月16日.13:31:12 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<1334583073.39.0.473040779285.issue8212@psf.upfronthosting.co.za> |
| In-reply-to |
| Content |
Urg, that's a horrible hack.
How about instead having an API function to resurrect an object from a tp_dealloc?
That way the iobase_dealloc code would be written:
if (_PyIOBase_finalize((PyObject *) self) < 0) {
_PyObject_ResurrectFromDealloc(self);
return;
}
That API function could also perhaps take care of the _Py_NewReference stuff (see the end of _PyIOBase_finalize). |
|