09.03.2011 23:59, Tomer Filiba kirjoitti:
yeah, i had the feeling someone would sneak in redhat and
their nonexistent releases...
you know, being stuck with software from 2004 in 2011... how come
people PAY money for that "support"?
okay, we'll keep python 2.4 support (even though it's ugly as shit :))
========================================
alex, fruch, and the everyone -- i would value your opinion on this:
do you think having the python3 code living side-by-side with the
python2 code is better than two branches?
i'm aware of the possible merge-nightmares, but i wonder what's worst...
Why side by side? You can provide py3k support in the very same code.
-tomer
An NCO and a Gentleman
On Wed, Mar 9, 2011 at 23:26, Jorge Maroto <[email protected]
<mailto:[email protected]>> wrote:
+1 to python 2.4
I agree RH is guilty for that, but there is a good amount of users you
should not just ignore (me included :)).
On Wed, Mar 9, 2011 at 8:41 PM, David Moss <[email protected]
<mailto:[email protected]>> wrote:
> Mostly due to Redhat's glacial support cycles ...
>
> On 9 Mar 2011, at 19:11, yairchu <[email protected]
<mailto:[email protected]>> wrote:
>
>> +1 to Python 2.4
>>
>> I get the impression that Python 2.4 is widely used in
well-known tech
>> giants.
>>
>> On Mar 9, 8:10 pm, Tal Einat <[email protected]
<mailto:[email protected]>> wrote:
>>> Indeed Python 2.5 has been around for many years and is a
reasonable minimum
>>> version to support IMO.
>>>
>>> Python 2.5 added with blocks, conditional expressions and unified
>>> try/except/finally, among other things. It would be a shame to
deprive
>>> ourselves of using these features in the RPyC codebase.
Versions 2.6 and
>>> 2.7, on the other hand, didn't add as many empowering features.
>>>
>>> +1 on Python 2.5 as the oldest supported 2.x version.
>>>
>>> - Tal
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Wed, Mar 9, 2011 at 1:07 AM, Tomer Filiba
<[email protected] <mailto:[email protected]>> wrote:
>>>> do you think it's feasible to have a single codebase for the
two versions?
>>>
>>>> at first i thought the changes would be minimal -- but they
turned out to
>>>> be quite numerous.
>>>> and it's not only syntax -- brine and vinegar had to be partially
>>>> rewritten, as well as lots of small changes,
>>>> like items() instead of iteritems() and byte strings instead
of strings.
>>>
>>>> i think maintaining the two versions in the same codebase
would be a
>>>> nightmare...
>>>> everything would be so cumbersome and tedious.
>>>
>>>> another idea i had was to have a xxx3.py version of every
xxx.py module, so
>>>> they all live side by side,
>>>> but i don't think it makes much sense.
>>>
>>>> ==========
>>>
>>>> on a side note, what is the minimum version of python do you
people think
>>>> RPyC should support?
>>>> i think 2.5 is early enough (been released in 2006)... should
we retain 2.4
>>>> compatibility?
>>>
>>>> -tomer
>>>
>>>> An NCO and a Gentleman
>>>
>>>> 2011/3/8 Alex Grönholm <[email protected]
<mailto:[email protected]>>
>>>
>>>> I see that you've put some py3k only code in the master3
branch. What's
>>>>> your plan? Are you abandoning the idea of compatibility code
and a shared
>>>>> codebase?
>
