Re: What does python 3000 mean for zope?

Tres Seaver wrote:
>> In fact Python 2.5 porting was not as much difficult as predicted in an 
>> old thread [1].
> 
> It isn't done yet, so I'm not sure what you are talking about.

Especially the difficult part, the untrusted code stuff in Zope 2, 
hasn't been tackled at all.

>> Nikhil has completed porting to Python 2.5 as part of Google Summer of 
>> Code project [2].
> 
> He ported *ZODB*, not Zope.

He ported the Zope 3 libraries (zope.*). He also worked on the ZODB ina 
a branch, but I see no sign of its merging. When compiling the ZODB with 
Python 2.5, I still get loads of compiler warnings.

> The ExtensionClass changes are not done,
> and I think there are other C-level changes which have not

It was never said whether Zope 2 would be part of the GSoC project or 
not. That said, we can't drop Python 2.4 support until

>> But we cannot officially support Python 2.5 until Zope 2 is also ported.
>> (This is a policy of Zope Foundation, I guess)
>> But we can give support for individual packages, is it ?
>>
>> May be we can try Python 3.0 porting in next GSoC ? :)
> 
> Frankly, I'm uninterested in spending *any* effort on Py3K support:
> we'd be more likely to get traction out of Jython / IronPython (which
> are alreday stable, and run on platforms we don't yet support).

... or PyPy, which would allow us to keep single-source implementations 
of things that currently have to be coded in Python (as a reference 
implementation) and C (for speedup). Furthermore it also runs on top of 
the CLI and JVM. To be honest, I find *that* much more exciting that Py3k.


-- 
http://worldcookery.com -- Professional Zope documentation and training