[Qgis-developer] Bug priority for release

[Andreas Neumann]

Hi Paolo,

Half of the blocker tickets are in feedback state. Seems like some users 
don't care so much as to response in a reasonable amount of time. Given 
that our "dedicated" bug fixing period is rather short, it is crucial 
that the users who reported a bug also respond quickly.

regarding

http://hub.qgis.org/issues/12926

Are you aware that there are two modes for loading DB tables? There is 
the checkbox "Select at ID". If you uncheck this for large tables it 
means that the initial loading is faster, but once you loaded all 
features they are cached clientside and should allow for fast scrolling. 
Of course this also means more local memory usage.

Note that for slower db-views or slow network connections it is also 
recommended to uncheck the "Select at ID" checkbox, so that the table 
will be cached locally.

Also: Matthias Kuhn is working on the SQL expression compiler that will 
allow serverside filtering. But it won't make it to 2.10. You'd have to 
wait for 2.12.

Regarding issue 12578: the new snapping mechanism was implemented by 
Martin Dobias - so it would be best, if he could fix the issue. 
Hopefully he can do that.

Andreas

On 12.06.2015 11:35, Paolo Cavallini wrote:
> Hi all.
> With 159 tickets classified as High and Blocker, it is quite obvious
> that even with the investment we are doing, many important bugs will be
> left in 2.10.
> Obviously the appointed developers will have the best judgement as to
> what is most efficient in the limited time frame. I respectfully point
> out a couple of serious ones that make working with large (millions of
> records) PG tables almost impossible:
> * http://hub.qgis.org/issues/12578
> * http://hub.qgis.org/issues/12926
> I think we would do a good service to our enterprise users if we can fix
> them. Actually, I'm surprised none of our big users already stepped in:
> maybe I'm missing something?
> All the best, and thanks.