[Qgis-developer] editing expressions defining virtual fields seems erratic

Janneke van Dijk janneke.qgis at gmail.com
Mon May 16 12:49:50 PDT 2016


Hi Mathias,
I've tried installing the dev version and I think that that worked (this 
is what 'about' shows).

QGIS version

	

2.15.0-Master

	

QGIS code revision

	

3d44a1e <https://github.com/qgis/QGIS/commit/3d44a1e>

Compiled against Qt

	

4.8.5

	

Running against Qt

	

4.8.5

Compiled against GDAL/OGR

	

2.0.2

	

Running against GDAL/OGR

	

2.0.2

Compiled against GEOS

	

3.5.0-CAPI-1.9.0

	

Running against GEOS

	

3.5.0-CAPI-1.9.0 r4084

PostgreSQL Client Version

	

9.2.4

	

SpatiaLite Version

	

4.3.0

QWT Version

	

5.2.3

	

PROJ.4 Version

	

492

QScintilla2 Version

	

2.7.2

	

This copy of QGIS writes debugging output.


I started with a new project, added the tst shapefile again, repeated 
the process of adding the virtual fields. The problem persists. Am I 
overlooking something?
Thanks for your help!
Janneke

On 16/05/2016 16:05, Matthias Kuhn wrote:
>
>
> On 05/16/2016 01:24 PM, Janneke van Dijk wrote:
>> Dear list,
>>
>> In a polygon shapefile I'm seeing inconstistent behaviour in virtual 
>> fields. The attached project and shapefile has reduced it to it's 
>> simplest form: the shapefile contains two virtual fields, the first 
>> one containing the expression 1 (integer) and the second the 
>> expression 2 (integer). However, when I want to edit the expression 
>> for the fields, the first field does not give a problem. For the 
>> second field I click on the 'expression' to open the expression 
>> builder, but the expression shown in the expression builder equals 1. 
>> I have not been able to find any pattern in when it happens/for what 
>> type of expressions. This example is the most simple form of the 
>> unexpected behaviour, but I have seen it with more complicated 
>> expressions as well. Sometimes after trying to edit one of these 
>> 'erratic' expressions QGIS closes with a minidump being written error 
>> message. Again, not always.
>>
>> Anyone have any idea why this might happen and how to work around it?
> Fixed in master
>>
>> Looking forward to some thoughts...
>> Janneke
>>
>>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20160516/795a0d49/attachment.html>


More information about the Qgis-developer mailing list