[Qgis-developer] QGIS 2.0 and PyQT v2 update

[Matthias Kuhn]

It was not my intent to encourage the postponement of these features to 
a later release.

I wanted to get a decision about inclusion or non-inclusion which is 
not based on time-constraints of the developers, but instead is 
strategic decision.

In my opinion, it can shed a bad light onto the project and discourage 
developers from maintaining their plugins if they've got to rework the 
whole plugin with several subsequent major releases.

If this is a top-priority goal for 2.0 I don't mind having it excepted 
from the API freeze (and therefore rendering the API freeze pretty much 
useless) for a week or two.

Concerning Python 3: This is a mail from 2010 which states, that there 
is the possibility to support Python 2 and Python 3 at the same time:
http://lists.osgeo.org/pipermail/qgis-developer/2010-January/008803.html
If this is really possible, I don't see any problems with postponing 
the Python 3 update, as we'll be able to have a painless grace period.

Thank you for your feedback
Matthias

On Son 28 Apr 2013 11:13:51 CEST, Régis Haubourg wrote:
> Matthias Kuhn wrote
>> We have already rendered many plugins inoperable by changing the vector
>> api (and some other API's) so lots of plugins will need some love to
>> get ready for 2.0.
>> Then there is the PyQt API V2 change which will render some more
>> plugins useless.
>> And I was just remembering, that I've heard, that the upgrade to PyQt
>> API V2 is done to be ready for Python 3. So it's probably also worth
>> considering, if the PyQt version changes and python upgrade should
>> happen at the same point in time. I have never upgraded code from
>> Python 2 to Python 3 myself, but there seem to be some differences
>> which require work to migrate.
>>
>> In order to be able to plan the plugin development. Is the upgrade to
>> Python 3 something which is really considered to happen in the next
>> couple of years? Would it not scare too many plugin developers and
>> users away, because it will require once more a considerable effort to
>> update?
>> If Python 3 is not an option for the next couple of years, is it worth
>> upgrading the PyQt API now?
>> Would it be better to upgrade to Python 3 right now! Because this will
>> require plugin developers to go over all their code now. Once only.
>>
>> I know it's quite late (with respect to the 2.0 roadmap) to bring this
>> up, but I haven't really thought of it earlier and I think it's a very
>> important decision.
>>
>> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
> +1 with Matthias, we should plan pyqt2 and python 3 for a future release. I
> see major risks in migrating now when we are in feature freeze with so many
> blockers to solve in 2 months. We have very few users and developpers that
> did real test to migrate to pyqt2 and python 3 , so we might have major
> issues hidden in there.
> We users really need 2.0 to happen this year in a stable state and I guess
> this is already a great challenge.
> My two cents.
> Régis
>
>
>
> --
> View this message in context: http://osgeo-org.1560.x6.nabble.com/QGIS-2-0-and-PyQT-v2-update-tp5049723p5050042.html
> Sent from the Quantum GIS - Developer mailing list archive at Nabble.com.
> _______________________________________________
> Qgis-developer mailing list
> Qgis-developer at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-developer