Phil Elson

I appreciate all of the effort that has gone into this issue so far. Unfortunately I think it has stalled a bit, possibly as a result of a switch of...

> At the moment, the only evidence we have is from one user, the OP I _currently_ represent 400+ scientific and engineering Python users, and have quite extensive experience of...

> Not sure how we can support exception reporting without a replacement API for PyFrameObject. Is this something we need to raise to the CPython mailing list perhaps? It would...

> It was pointed out that we need a formal method to construct a stacktrace and register it as part of a cause chain in the past. Unfortunately, I have...

Sorry to bother you @vstinner, but I suspect it may be interesting for you to see the current approach and motivation for how certain things are done in JPype. The...

> I took my best shot at the GC allocator kludge. Please give it some tests to see if it is close to working. I don't see a new commit...

> If this is successful then we don't need a cap, we just have to disable that frame test or find a way to make it work when I cycle...

I wasn't entirely sure how to exercise the part that you've had to tweak, but based on some of the tests I went for: ``` import jpype as jp jp.startJVM()...

We still have compile errors: ``` ative/common/jp_exception.cpp: In function ‘PyObject* tb_create(PyObject*, PyObject*, const char*, const char*, int)’: native/common/jp_exception.cpp:527:9: error: ‘PyThreadState’ {aka ‘struct _ts’} has no member named ‘frame’; did you...

> see https://docs.python.org/3.11/whatsnew/3.11.html#id6 Thanks @rapgro, but the problem is that we are *setting* the frame, not *getting* the frame in this case. I appreciate that this hasn't been public API...