i have just tested latest release of qgis with a simple dataset. it is polygon data - type geodatabase with some categorial columns. the one i've tested characterises land-use types in 15 categories. colum-type is integer - value range is from 1 to 15. layer properties / style / categorized / classify returns the following classification: enter image description here

these values originate from the first numeric column within the corresponding attribute table - instead of the selected column.

are there any known workarounds yet ?

  • Cannot duplicate your issue on a postgis table. The selected column (and no other column) is used for classification?
    – Jakob
    Mar 2, 2016 at 12:03
  • 1
    Can you maybe post your attribute table?
    – Jason
    Mar 2, 2016 at 12:24
  • tried to reproduce with a shape-file copy of my gdb-dataset. classification worked fine with that shp ! according to my additional tests this seems to be gdb-related, becuase graduated style returns faulty classification too. has anyone tested gdbs right now and similar results ? ... i can send a section of the data including geometry (dbd)
    – robert tuw
    Mar 2, 2016 at 16:50
  • ... maybe a bug tracker issue ?? right now i am trying to find an interim-workaround.
    – robert tuw
    Mar 2, 2016 at 16:54

2 Answers 2


Looks like it's a bug in the GDAL geodatabase driver. Can you file it over at http://hub.qgis.org/issues along with a copy of your dataset and I'll follow it up with the GDAL crew?

  • just added a bug-report to hub.qgis.org
    – robert tuw
    Mar 3, 2016 at 11:10

finally found a solution (i.e. a workaround) for (ESRI)geodatabases:
(.. which potentially doesn't support the GDAL-driver hypothesis (?))

uese expressions ! instead of selecting a field form the properties/style/categorized/column-dropdown i use the following expression: "lbtyp" * 1 where "lbtyp" is my categorial field
the resulting classification works fine.

the same procedure works with graduated-type as well ! enter image description here

  • Good workaround! There's still a bug in gdal itself though (QGIS falls back to a slower, non GDAL categorization when an expression is used)
    – ndawson
    Mar 5, 2016 at 3:56
  • your comment draws my attention back on one of my questions from last year (gis.stackexchange.com/questions/95166/…) ... where i argued that not beeing able to have selects on objects from joined tables is like having databasesystems without SELECT on VIEWS and use CREATE TABLE instead ... which would be a no-go there. ... are those two things somehow related in GDAL ?
    – robert tuw
    Mar 13, 2016 at 17:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.