[Qgis-developer] geo-processing tools: fixes, plans and fundings

[Paolo Cavallini]

Il 12/04/2016 15:43, Thomas Schüttenberg ha scritto:

> Within the German user community there is a discussion going on about how to effectively support the fixing of several issues with some essential vector geo-processing tools (e.g. intersection, clip, join attributes) that we think have to work without flaws. [0]
> But because of the modular structure of processing and tools in QGIS in general, this topic can also be seen in a broader context. The ftools for example have just been removed in favour of the processing tool box (QEP 36) [1]. 
> So in order to make a good investment decision we would like to know, if there are already further plans on your side to consolidate, rewrite or replace (some) geo-processing tools in/for QGIS, which eventually also solve these issues. Would it then make (more) sense to start a crowd funding campaign for this - instead of funding short term fixes on maybe soon deprecated code?

Hi Thomas,
thanks for this. May I suggest to invest, first and foremost, in a
complete set of tests for Processing? Having this will make it possible
to constantly rely on Processing across versions; without it we always
risk to get wrong or no results from our analyses.
A testing infrastructure is already in place for some algs, what we need
is a better coverage. The nice thing is that part of this job can be
daone by power users, without have to rely uniquely on core developers.
Sounds attractive?
All the best.
-- 
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html