Bug report #2051
Rendering issues in 1.3.0
|Affected QGIS version:||Regression?:||No|
|Operating System:||All||Easy fix?:||No|
|Pull Request or Patch supplied:||Resolution:||worksforme|
|Crashes QGIS or corrupts data:||Copied to github as #:||12111|
I've been trying to get me head around this for a while now but seem to keep coming back to this as a potential bug but I'm afraid it seems a little odd to me, so please excuse the explanation.
- When I load a single layer - no problems.
- When I load multiple layers, from any source, with different CRS and on-the-fly ticked, the additional layers don't appear on the main area.
- Zooming to the extent of the invisible layers confirms they are in the right place, and adding them to the overlay also confirms this.
Sometimes, the additional layers appear in the overlay box correctly but are still invisible in the main area, sometimes they are invisible but the red bounding box in overlay doesn't move when zoom to the extent of a hidden layer.
Changing something on a hidden layer to force QGis to re-render hasn't worked so far.
These layers - generally a mixture of shape files and postgis - worked perfectly under earlier releases of QGis.
The thing which confuses me about this is that when I save an invisible layer to a shapefile and load that, its loads in the right place with the polygons properly placed. However, the attribute table has been distorted.The column headings are correct, but the data after the second column has been shifted by one column, with the final colm of data missing:
- columns 1 & 2 are correct
- the 3rd column now contains only nulls
- the 4th contains column 3 data,
- the fifth column contains column 4 data,
In case its relevant the reference systems being used are the Mapnik/Google (900913) projection as a custom CRS generally used for raster images generated by Mapnik, and 27700 OSGB PostGIS and shapefile layers.
I have QGis 1.3 running under 64-bit Ubuntu 9.04 (and now 9.10) on an AMD quad-core with 8Gb RAM - installed from both the QGis unstable repository and the UbuntuGIS source. I was able to replicate the issue on a Windows XP machine from a copy installed from OSGeo4W.
I'd be very happy to find it was a mess-up on my behalf but no responses on the forum except to report a bug suggests otherwise.