[Qgis-user] Bug copying features

[Reginald Carlier]

Hi Richard,

Yes my second shape has a subset of attributes of the first. In fact only one fieldname matches with the first shape.
In fact the first shape is a modelshape from the digirup project (Belgium) and the second is a modelshape from the DSI project.
If I find the time I will test it on linux. Maybe it is a bug in windows only.

Regards,

Reginald Carlier
Deskundige GIS


-----Oorspronkelijk bericht-----
Van: Richard Duivenvoorde [mailto:rdmailings at duif.net] 
Verzonden: donderdag 20 april 2017 14:42
Aan: Reginald Carlier; 'qgis-user at lists.osgeo.org'
Onderwerp: Re: [Qgis-user] Bug copying features

On 20-04-17 09:26, Reginald Carlier wrote:
> Hello,
> 
>  
> 
> I am using QGIS 2.18.6 on Windows 7 (OSGEO).
> 
> I am encountering a bug everytime I want to copy features from one 
> layer to another. Some fields of layer 1 are the same in layer 2.
> 
> As I add the layers to the project, I then set the two layers editable.
> Then I open the attribute table of layer 1 and select all the features.
> Then I copy all the features. Then I select layer2 and I paste all the 
> features. As a result the features are copied to layer 2 but in the 
> attribute table I can only see null values.
> 
> The solution for this problem is to save the project and reopen the 
> project and then do the steps mentioned here above again. Then the 
> features are copied in a correct way.

Hi Reginald,

I tried to reproduce your problem here with 2.18.6 (on Debian though).
And I cannot reproduce it. I:

- opened a dataset and saved it as a 2nd shape
- removed all features from the 2nd shape (to have a shape with exact the same attributes)
- openend both shapes in QGIS
- copied one feature from 1 to 2
- stopped editing of both layers
- note: still not a saved project
- and see the right features in 2...

So question: is your second dataset maybe ANOTHER shape with a subset of the attributes of the first one?

Anyway, this looks like an issue, feel free to create an issue for it, BUT make it as reproducable as possible: add small datasets to it.
Because as you can see it's not easy to reproduce apparently :-(

Regards,

Richard Duivenvoorde