RetroSearch Browse

Mon Jul 21 11:59:42 CEST 2008 · https://mail.python.org/pipermail/python-dev/2008-July/081457.html

On Mon, Jul 21, 2008 at 11:48, Jesus Cea <jcea at jcea.es> wrote:
> I can comment about some issues I had this weekend.

I don't do C development myself, so comments aren't that useful for
me, but code examples are, so we can stick them into
python-incompatibility.

> Remember that my intention is to keep a single C codebase for 2.6 and 3.0.

Cool.

> * Int/Long integration. In Python 3.0 "PyInt_*" has vanished. But using
> "PyLong_*" in Python 2.x surfaces so many issues that I have decided to
> define "NUMBER_*" macros to be conditionally expanded to
> "PyInt"/"PyLong" when compiling to 2.x/3.0. Not nice, but I can't see a
> better way.

Seems resonable.

> PS: I'm learning the hard way, doing "diff" between 2.6 and 3.0 module
> sourcecode. It must be a better way!.

Yeah, these changes should be properly documented in the CHANGES.txt.
I've seen some C-API chnges mentiones at least.

-- 
Lennart Regebro: Zope and Plone consulting.
http://www.colliberty.com/
+33 661 58 14 64

Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://mail.python.org/pipermail/python-dev/2008-July/081457.html below:

[Python-Dev] Any PEP about 2.6 -> 3000 code transition?