[Biopython-dev] buildbot issue on Python 3.1 - stdout?

[Wibowo Arindrarto]

Hi everyone,

>> It's reproducible in my machine: Arch Linux 64 bit running
>> Python3.1.5. Haven't figured out a fix yet, but trying to see if I
>> can.
>
> Great. We haven't really proved this is down to a change in
> either Python 3.1.4 or 3.1.5 but it does look likely.

It's reproduced in my local 3.1.4 installation. Seems like an unfixed
bug that went through to 3.1.5.

>> By the way, I was wondering, what's our deprecation policy for
>> Python3.x? I saw that 3.1.5 was released in 2009, and there doesn't
>> seem to be any major updates coming soon. How long should we keep
>> supporting Python <3.2?
>
> As long as it doesn't cost us much effort? If we can't solve this
> issue easily that might be enough to drop Python 3.1?

Fixing this seems difficult (has anyone else tried a fix?). The _io
module is built-in and compiled when Python is installed, so fixing it
(I imagine) may require tweaking the C-code (which requires fiddling
with the actual Python installation).

> My impression is that Python 3.0 is dead, and the only sizeable
> group stuck with Python 3.1 will those on Ubuntu lucid (LTS is
> supported through 2013 on desktops and 2015 on servers),
> but as with life under Python 2.x it is fairly straightforward
> to have a local/additional Python without disturbing the system
> installation.
>
>
> Since we don't yet officially support Python 3 (but we probably
> should soon) we have the flexibility to recommend
> either Python 3.2 or 3.3 as a baseline.

Yes. I think it may be easier and better for us to officially start
supporting from Python3.2 or 3.3 onwards.

regards,
Bow