richard.m.tew at gmail.com
Wed Apr 4 21:53:45 CEST 2012
On Wed, Apr 4, 2012 at 9:28 PM, Kristján Valur Jónsson
<kristjan at ccpgames.com> wrote:
> I wonder,
> Why does stackless redefine PyHeapTypeObject?
> Is there any reason to? Can we fix that so that compatibility with stuff
> such as PyQT is maintained?
Yes, this has been a long time known problem. We extend the object
structure with Stackless-specific fields, and so does SIP which is
what PyQT uses for its bindings. However, the number of complaints
about this since Christian and I last discussed it number exactly zero
If PyQT is the only problem, then I am not sure we need to fix it.
Other problems with PyQT which I am not familiar with have
precipitated the development of PySide. PySide, as I understand it,
can be interchanged with PyQT with minor changes.
It might be sufficient to encourage people to switch to PySide.
Of course, if you can see a way to easily remove the redefinition, we
should probably look at doing that anyway.
More information about the Stackless