[Qgis-psc] QGIS 2.18 bug fixing effort
Giovanni Manghi
giovanni.manghi at gmail.com
Tue May 16 01:57:45 PDT 2017
Hi All,
I would like to congratulate and thank all the persons that have been
involved so far in the effort for the 2.18.* bug fixing.
The regressions that have slip into 2.18 since 2.14 are almost all
squashed or awaiting a PR to be merged.
Other bugs have been fixed along the way.
There a few "minors" regressions (in 2.18, since 2.14) left:
15463 Layer "Scale dependent visibility" doesn't work anymore since 2.16
15316 symbology offset based on expressions does not work anymore
16517 Locked raster layers do not refresh in Composer
and a others "less minor" like
16296 Cannot use anymore d&d to import a layer from Spatialite in
PostGIS (just on Windows)
A remaining serious issue now is the fact that in 2.18/osgeo4w the
suport for SAGA LTR (that is the only one we want to support on all
platforms) is broken on Windows. This is not a Processing issue but
rather a packaging one
16423 Broken io_gdal raster import in SAGA LTR package shipped with OSGeo4W
16422 Processing (on Windows): external SAGA does not work anymore
probably the second could be closed as we should not allow anymore use
SAGA instalations others than the one that comes woth osgeo4w, anyway
the SAGA LTR installation in osgeo4w at the moment is not good.
I Would like also to raise some awareness on a bug that passed under
the radar for quite some time
14589 GetCapabilities gets cached in a confusing way (even when QGIS
cache is set to 0)
QGIS caches WMS requests, and that is ok. Problem is that there seems
no way to set that cache to 0 and the clear cache button does not
work. This causes a lot of headaches to the ones that are preparing
WMS services and using QGIS as client and also was source of several
reports by users because their preferred WMs service had changed and
QGIS was still using a cached (wrong) request.
Last but bot not least:
We *really* need to make something about the geoprocessing tools that
*still* are returning incomplete/wrong results in a semi-transparent
way (no, the common user do not look the log). This was last discussed
in Las Palmas but afaik not much has been done in this sense. I have
already proven that with purely SPATIAL SQL tools (via ogr2ogr) the
results are more reliable and considerably faster (tests made with
dissolve and clip, but I could add any other operation). Would each
copy of spatialite shipped with QGIS be compiled with support for
ST_Makavalid I could add a check to have the geometries fixed
beforehand and in a complete transparent way (bad geometries are the
real problem for the tools we used to call "ftools").
Cheers
-- G --
On Fri, May 5, 2017 at 10:02 AM, Andreas Neumann <andreas at qgis.org> wrote:
> Hi Alex,
>
> Thank you for your reply.
>
> I'll cc Giovanni.
>
> Giovanni - can you please provide Alex a list of processing/analysis related
> bugs/regressions - the most pressing ones?
>
> Alex: we'll be happy to pay for a couple of hours. We would be interested in
> a regular relationship with core devs who are available to fix the most
> pressing issues prior to every release and potentially also after.
>
> Can you estimate - how many hours or days you can dedicate to fixing issues?
> I mean within the next 1-2 months and for QGIS 2.18
>
> I'll add you to our list of core devs who participate in our bug fixing
> effort then -
> https://docs.google.com/spreadsheets/d/1Cdh-6eAmVCdiZIQGMhu_Aqgg09tpxsC40BGXQJtYc54/edit#gid=1281977200
>
> Thanks,
> Andreas
>
>
> On 3 May 2017 at 18:39, Alexander Bruy <alexander.bruy at gmail.com> wrote:
>>
>> Hi Andreas,
>>
>> 2017-05-03 13:10 GMT+03:00 Andreas Neumann <andreas at qgis.org>:
>> > I wonder if you would be interested / would have time - to assist us
>> > with
>> > our QGIS 2.18 bug fixing efforts - esp. with the processing issues?
>>
>> Sure I will be happy to help with this as much as I can.
>>
>> > We would pay some devs for up to 5 days (total 40h per dev) with a fixed
>> > hourly rate of 100 €. Some devs can only contribute a day or two - which
>> > is
>> > also fine.
>>
>> I would like to donate my time to project.
>>
>> --
>> Alexander Bruy
>
>
>
>
> --
>
> --
> Andreas Neumann
> QGIS.ORG board member (treasurer)
More information about the Qgis-psc
mailing list