Stefan Richthofer

The typechecker should be kept exchangeable as no framework (for runtime typechecking) gets everything right yet. The fact that the typing module changes heavily from Python version to Python version...

pytypes can resolve these strings/forward references. The case that such strings occur deeper within a type was supported only a while ago and no release was filed since then. See...

Calum, that's a brilliant finding. For some objects, especially type objects (``PyType``) themselves, the ``ob_type`` field is not inititalized at definition time, but by ``PyType_Ready``. IIRC there was not really...

I remember it is not easy to find the actual C-source code of ``random.rand``, it is in a cython-generated file. For easier discussion I'll simply post it here. It boils...

I checked about ``JyNI_InitJythonPyException``. That function is to be understood like the stuff in JySync.c. It is a data converter that is only to be used for initially converting a...

Regarding ``JyNI.JyNI.clearNativeHandle`` it is crucial to know in what context it was called and what type of object was passed in. Strangely, that's a Java method, so how can it...

``clearNativeHandle`` is only called by ``JyNI_CleanUp_JyObject`` in JyNI.c. ``JyNI_CleanUp_JyObject`` in turn is called from various places in JyNI-C project, also from gcmodule.c. To track down this issue it would be...

> There are a few different errors going on Let's strictly work through them by order of appearance. Feel free to open separate issues here if appropriate. > just having...

> I agree that B is probably _randint_type but I don't know what A is, _randint_type is declared outside of any classes, its needed by the RandomState class (the one...

I just found time to attempt to reproduce this myself. It seems I don't even get to the point of the segfault. I cloned the branch "Iterator_Support", using NumPy 1.13.3:...