On Fri, Jan 21, 2011 at 9:50 AM, Matti Airas <[email protected]> wrote:
> On 21.01.2011 16:11, ext b k wrote:
>>
>> Hello all,
>>
>> First off, I want to emphasize that I understand the nature of open
>> source development, and that I am not entitled to any free work from any
>> of the developers and it's impossible to give precise estimates for when
>> anything might get done.
>>
>> All that said, I'm trying to assess whether it is worth investing the
>> money in a commercial PyQt license. My codebase is approx 10k LOC, all
>> Python 3.x. I've read the wiki article on Python 3 Issues. Can anybody
>> give me insight on a) if this work has been started yet; b) a very rough
>> ballpark figure on when there might be even early-beta-level Python 3
>> support?
>
> Hi,
>
> for the very reasons you mentioned (as well as the general difficulty of
> giving reliable schedule estimates for complex software), I'm grown wary of
> giving any exact estimates. However, I can shed some light on the short-term
> directions for the core dev team:
>
> First, we're a few weeks off (depending on how many more betas we're doing)
> from doing the 1.0 release. The time until that is spent in stabilization:
> we're only fixing incoming bugs to improve the final release quality.
>
> Then, there're still some reference documentation issues which need
> addressing (porting the remaining C++ code snippets in the ref doc to Python
> - this would be a perfect volunteer job, btw!).
>
> Also, optimizing memory usage is one of the first things to do after the
> PySide 1.0 release. The bindings are fast (except for some specific issues
> that have popped up) but the memory footprint could be smaller, and I hope
> we can address that. The core dev team members have had ideas about how to
> do that, but the consensus was to improve the stability first.
>
> I think Python 3 support would be topical after the above issues have been
> addressed. However, since I expect PySide's use to increase once we have
> done the 1.0 release, the inflow of bugs will probably remain high, so it's
> quite difficult to even guess at which point we could get there.
>
> Of course, the description above is just my guesstimates how the work
> prioritization for the core dev team would go: if there would be sufficient
> interest within the rest of the community to provide the Python 3 porting, I
> see no reason why such work couldn't be integrated right after the 1.0
> release.
>
> Maybe a more realistic approach for you would be to port the code back to
> Python 2 for the time being? Given the rather modest differences between
> Python 2 and 3, it doesn't sound to me that this would be an unrealistic
> amount of work.
>
> Cheers,
>
> ma.

With python 2 support ending and distributions apparently preparing to
switch over, this may become a more pressing issue in the near future.

-Todd
_______________________________________________
PySide mailing list
[email protected]
http://lists.openbossa.org/listinfo/pyside

Reply via email to