[Stackless] Preparing Stackless Python 3.1a2

Richard Tew richard.m.tew at gmail.com
Tue Apr 7 03:41:59 CEST 2009

On Sun, Apr 5, 2009 at 4:08 PM, Richard Tew <richard.m.tew at gmail.com> wrote:
> On Linux:
>  test_watchdog.py:test_pickle() fails.
> It looks like range(1000) is unpickled as something which causes an
> infinite loop.

I ignored this precise description and spent ages comparing pickle
outputs, then gave up and just looked at the source code, saw a
disparity in the range iterator pickling code and fixed it.

> There are seven Python unit test failures.  The Windows unit tests all
> pass now.  Some of these failures are because of the configuration of
> the Linux machine, but which ones, I haven't determined yet.

I exported a copy of the mainline r3.1a2 branch and compiled it.
Running the unit tests gave the same failures as the Stackless branch,
so these are not Stackless problems.

Can people please get the Stackless r3.1a2 branch, compile it, run the
Python unit tests and the Stackless unit tests?  And then post to the
list with your results.


The script to run the Python unit tests with the executable you build is:

The script to run the Stackless unit tests with the executable you build is:

If you get test failures in the Python suite, if you can download and
compile the mainline branch and run the Python unit tests there and
compare the results, it would be appreciated.



More information about the Stackless mailing list