[SciPy-Dev] Required Python Version

Charles R Harris charlesr.harris@gmail....
Sat Jul 17 14:55:52 CDT 2010


On Sat, Jul 17, 2010 at 1:42 PM, Pauli Virtanen <pav@iki.fi> wrote:

> Sat, 17 Jul 2010 14:10:26 -0500, Benjamin Root wrote:
> [clip]
> > And I have seen a few independently administered government machines
> > running Fedora, that doesn't necessarily reflect the IT policy of
> > government servers.  What would probably be important to find out is if
> > the features/enhancements that would be made available by dropping
> > support for 2.4 outweigh leaving behind a group of (dedicated) users.
> > In addition, are there features/enhancements that are planned that don't
> > require dropping 2.4, but that these users would also want/need?
>
> I do not think supporting 2.4 has a large impact on maintenance burden.
> The differences to Python > 2.4 are small, at least if compared to Python
> 3.
>
> I would not consider the "with" statement or absolute imports very
> important at this point. We already went through the trouble to get
> things working on Python 3, and deciding to drop Python 2.4 to make the
> transition easier is no additional help at this point.
>
>
How to parse Py_ssize_t values? I think the 'n' format code is quite useful
for supporting 64 bits in a portable way and it isn't available in 2.4.

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/scipy-dev/attachments/20100717/63f873aa/attachment.html 


More information about the SciPy-Dev mailing list