[Python-3000] threading, part 2

[Tim Peters]

[back and forth on PyThreadState_SetAsyncExc(), and
 the 2-year old discussion in

 http://www.python.org/sf/1069160
]

[Tim]
>> [still-current deadlock & refcount issues not fixed at the time]

[Guido]
> So why didn't we check that in?

The shallow answer is that you closed the report without checking it
in, so ask a mirror ;-)  The real answer seems to be that nobody
(including me) really cared about this function, since it's both
unused and untested in the core, and there were no known uses from
anyone's C extensions either.

[on adding it to the language]

>>>>> +0

>>>> Me too, although ... I'm clear as mud on how implementations other
>>>> than CPython could implement this.

>>> Now I'm -0 on adding it. I suggest that if someone really wants this
>>> accessible from Python, they should research how Jython, IronPython,
>>> PyPy and Stackless could handle this, and report their research in a
>>> PEP.

>> As a full-blown language feature, I'm -1 unless that work is done
>> first.  I'm still +0 on adding it to CPython if it's given a
>> leading-underscore name and docs to make clear that it's a
>> CPython-specific hack that may never work under any other
>> implementation.

> Fine with me then. In 2.5? 2.6? Or py3k? (This is the py3k list.)

Since the 2.5 beta series is supposedly done with, I strongly doubt
Anthony wants to see a new feature snuck into 2.5c1.  Someone who
wants it enough could target 2.6.  I'm only +0, so I'd do that only if
someone wants it enough to pay for it.  For 2.5, I'll check in the
anal correctness changes, add a ctypes-based test case, and reword the
docs to stop warning about a return value > 1 (all those are just
fixing what's going to be in 2.5 anyway).