[Stackless] Barriers to getting stackless into Python trunk?

[Richard Tew]

On 9/20/07, Lloyd Weehuizen <lloyd at fusion.net.nz> wrote:
> Richard Tew wrote:
> >
> > This is assuming that I am the one who does the port.  Anyone is
> > welcome to give it a shot if they want to get more involved..
> > Anyone?..  Anyone?.. Bueller?
>
> I'd be keen to help out with a Py3k port where I can. Of course my
> knowledge of the inner workings of stackless is a lot more limited than
> yours and Chris's :)

My approach to porting it is to just diff a Stackless version of the
Python source code against a non-Stackless version and then to apply
the patch to the branch I am going to make Stackless.  Then it is a
matter of resolving the conflicts, running the unit tests and fixing
the bugs.

I imagine that if people were to work together to do this, one would
do the initial merge and resolve the conflicts, then check in the
result.  From there people would step forward and fix the various
failing tests.  If you have an idea for a better approach, let me know
:-)

My knowledge of the Stackless internals was pretty lacking when I
first helped with this process at PyCon for 2.4.2, but thankfully I
had Kristjan Valur and Christian Tismer there to work with.  But I
think that the most can be learned through this by just plugging away
and making sure you understand the code and what it should be doing,
rather than relying on others more familiar with the code to lead you
through it -- which is what taking a failing test and trying to fix it
would hopefully encapsulate.  I know it sounds like handwaving to say
that, but the fact of the matter is that this code is not trivial and
the only way to really understand it is to have a reason to study it
and debug it.

Cheers,
Richard.