scoder

> [python/cpython#124292](https://github.com/python/cpython/pull/124292) That's a change in `ctypes`, though. We don't use that anywhere. I don't see how that change could be related.

Yeah, the symptoms hint at something like `__hash__`. Also that the error appears unstable – the hash variation across Python runs may well induce that. I'll try if setting `PYTHONHASHSEED`...

I've reported the issue in the original CPython PR for now.

> require the `cdef` at the beginning of the function. It's not limited to the beginning of the function. You can put the declaration anywhere in the function except for...

It's definitely one hack replacing another, and it's not obvious that it's improving on the hackishness.

We need to disable at least some of the tests in old Python versions (2.7).

I'm fine with merging this, but yes, please document it as experimental. It's not obvious that there will be changes later, but it's going to take a while until we...

The new test is failing in the freethreading CPython build: ``` Errors from shard 6: ====================================================================== FAIL: runTest (__main__.EndToEndTest.runTest) [6] End-to-end freethreading_compatible ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/runner/work/cython/cython/runtests.py",...

> I _think_ the original intent was to throw an exception when NAN is read I don't think so. I think it's meant to propagate the `NoDataLeft` exception if the...

When I read the PEP, my immediate response was: "but I'm using this all over the place to swallow exceptions!". It's actually quite handy that you can just return from...