#3221: r.cross output misses one category
-------------------------+-------------------------
Reporter: pvanbosgeo | Owner: grass-dev@…
Type: defect | Status: new
Priority: normal | Milestone: 7.0.6
Component: Default | Version: unspecified
Keywords: | CPU: Unspecified
Platform: Unspecified |
-------------------------+-------------------------
I have two layers with 2 categories (1 and 2) each. When combined using
r.cross the output is a map with three categories according to the message
on the command line. Also, the map created has three colours.
Yet, there should be four categories, and there is indeed a fourth
category with value 0 (but without category label)
This category 0 is the area where both layers have a value 1. See also the
attached screenshot.
So there are the problems that:
* only three categories are reported, while there are four in reality
* one (the first) category has no category label with the values of the
input maps
* the category 0 (without category label) is assigned the same colour as
category 3
* When running the tool to create interactively the colour table, colours
are only assigned to categories 1-3, so the category 0 should not have a
colour (see attached screenshot). Clicking the 'apply' button without
changing the categories, and the category 0 will indeed disappear from the
map.
#3221: r.cross output misses one category
--------------------------+-------------------------
Reporter: pvanbosgeo | Owner: grass-dev@…
Type: defect | Status: new
Priority: normal | Milestone: 7.0.6
Component: Default | Version: svn-trunk
Resolution: | Keywords:
CPU: Unspecified | Platform: Linux
--------------------------+-------------------------
Comment (by mlennert):
Replying to [ticket:3221 pvanbosgeo]:
> I have two layers with 2 categories (1 and 2) each. When combined using
r.cross the output is a map with three categories according to the message
on the command line. Also, the map created has three colours.
>
> Yet, there should be four categories, and there is indeed a fourth
category with value 0 (but without category label)
>
> {{{
> r.stats -l input=grevrobenvclust
> 0
> 1 category 1; category 2
> 2 category 2; category 1
> 3 category 2; category 2
> * no data
> }}}
>
I can confirm this with the NC dataset in grass73. Maybe some issue with
category value 0 ?: