[Python-Dev] New thread death in test_bsddb3

[Thomas Heller]

Tim Peters <tim.one@comcast.net> writes:

> [Mark Hammond]
> > Actually, some guidance would be nice here.
> 
> It's easy this time.  BTW, I agree your new check is the right thing to do!
> If another case like this pops up, though, we/you should probably add a
> section to the PEP explaining what to do about it.
> 
ctypes ;-) is another case (and more cases will pop up as soon as the beta is
released, and people try their extensions under it).

I agree it is easy to fix, but usually when Python crashes with an
invalid thread state I'm very anxious at first.

So is the policy now that it is no longer *allowed* to create another
thread state, while in previous versions there wasn't any choice,
because there existed no way to get the existing one?

IMO a fatal error is very harsh, especially there's no problem to
continue execution - excactly what happens in a release build.

Not that I am misunderstood: I very much appreciate the work Mark has
done, and look forward to use it to it's fullest extent.

Thomas