[QGIS-Developer] QGIS 3 > Processing: some test

[Nyall Dawson]

On 20 June 2017 at 22:03, Giovanni Manghi <giovanni.manghi at gmail.com> wrote:
> Hi Nyall,
>
>
>>> and dissolving by "distrito" the ogr2ogr based tool is 2/3 seconds
>>> faster than the new native one.
>>
>> For comparison, how long did each take in total?
>
> about ~28/29 seconds ogr and ~ 31/32 native (on a testing VM with
> limited resources).

I'm happy with that :D

>> Hmm - works fine here. Can you try turning off the invalid geometry
>> checking in processing options? That's a little broken at the moment
>> (with the error you've noted below)
>
> if I turn off that option then it works, both dissolving by attribute
> and a full dissolve.
> The speed difference is not much (ogr always a bit faster in my
> testing environment).

Also that :)

I'd love to see some tests using non-ogr sources (e.g. dissolve a
postgres layer or memory layer). I think in this case QGIS native may
be able to shave off this couple of seconds lead the OGR variant has.
(Although that's not really that important - if we can get the speed
comparable, with stable invalid geometry handling, then I'll be
satisfied).

>> That should be really straightforward to add to the feature iterators
>> now. But alternatively there's a processing alg which repairs
>> geometries too (not in master yet, but in earlier 3.0 builds).
>
>
> both very good news! but don't forget that in the economy of a work
> day of a person that do all day this kind of operations, having an
> option to fix geometries on the fly while clipping/intersecting/etc is
> very important and save a lot of time. Of course having the same
> functionality as separate tool is as much as important.

Agreed.

Thanks for this testing - it's very valuable!

Nyall