[QGIS-Developer] Processing 3.0: Possible change to the Convex Hull algorithm

[Paolo Cavallini]

Il 28/06/2017 00:54, Nyall Dawson ha scritto:

> Ok - perhaps requiring only multipart feature inputs is too extreme. How about:
> 
> 1. We drop the field group option, for the same reasons initially
> described (there's better algorithms to do this)
> 2. Add a new boolean option for which creates as single hull for all
> features in the layer
> 
> Would that work?
> 
> Alternatively we could have two algorithms:
> 
> 1. Convex hull - operates per feature on just the geometry for each feature
> 2. Layer convex hull - generates a convex hull covering a whole input layer

In my experience, CH is mostly used for points, and multipoints are
little used, as they are problematic in several respects. So the case of
CH over multiparts is IMHO the exception rather than the role.
What happens for shp, that does not really have a multi- concept?
If we use the boolean as you suggested, I think the default should
probably be all features.
Thanks.

-- 
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
https://www.google.com/trends/explore?date=all&geo=IT&q=qgis,arcgis