Bug report #13067: The difference tool gives broken results in both 2.10 and 2.11 - QGIS Application - QGIS Issue Tracking

Bug report #13067

The difference tool gives broken results in both 2.10 and 2.11

Added by Etienne Trimaille over 9 years ago. Updated over 9 years ago.

Status:	Closed
Priority:	Severe/Regression
Assignee:	Victor Olaya
Category:	Vectors
Affected QGIS version:	2.10.0	Regression?:	No
Operating System:		Easy fix?:	No
Pull Request or Patch supplied:	No	Resolution:
Crashes QGIS or corrupts data:	No	Copied to github as #:	21137

Description

Running the difference tool (vector menu and processing) in QGIS 2.10 and master gives broken results.
Instead of removing features, the layer contains invalid geometries.

No error is raised, but the layer can't be used for a further analysis.

Associated revisions

Revision fe1a0321
Added by Etienne Trimaille over 9 years ago

fix #13067 : bad geometries in the difference tool

Revision 4fba5ff4
Added by Jürgen Fischer over 9 years ago

Merge pull request #2192 from Gustry/fix-13067

fix #13067 : bad geometries in the difference tool

Revision 2e77e9cd
Added by Etienne Trimaille over 9 years ago

fix #13067 : bad geometries in the difference tool

(cherry picked from commit fe1a032181fb8f3c696918da71df777521f05ceb)

History

#1 Updated by Anita Graser over 9 years ago

I can reproduce the issue computing the difference of regions and lakes in our QGIS Alaska sample data. (Difference between regions and builtups seems fine though.)

#2 Updated by Giovanni Manghi over 9 years ago

Target version changed from Version 2.12 to Future Release - High Priority
Priority changed from High to Severe/Regression

This is the second (basic) vector geoprocessing tool we at the moment see also #11986 (not to speak about how slow they all are).

I really think that this tools should be really fixed once for all as any gis user expect them to work and having to say "don't use the intersection/difference/whatever tool is kind of embarrassing. It even could be dangerous... what about if I take for good the result of one of this operations and then give wrong results that have a real world (legal for example) impact?

Now that we have the routine to create virtual vectors I plan at some point to add the same basic tools based only ogr/sql/geos (there are already a few in Processing).