[Octave-bug-tracker] [bug #58325] NaN color with imagesc

[Guillaume]

URL:
  <https://savannah.gnu.org/bugs/?58325>

                 Summary: NaN color with imagesc
                 Project: GNU Octave
            Submitted by: gyom
            Submitted on: Thu 07 May 2020 04:47:41 PM UTC
                Category: Plotting with OpenGL
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Matlab Compatibility
                  Status: None
             Assigned to: None
         Originator Name: Guillaume
        Originator Email: 
             Open/Closed: Open
                 Release: dev
         Discussion Lock: Any
        Operating System: Any

    _______________________________________________________

Details:

When displaying an image containing NaNs with imagesc, Octave displays them as
black while Matlab uses the first color of the colormap.

I could find a related comment on the Octave Help mailing list:
https://lists.gnu.org/archive/html/help-octave/2018-10/msg00059.html
and it seems to be frequently mentioned by Matlab users:
https://stackoverflow.com/questions/14933039/matlab-imagesc-plotting-nan-values
https://stackoverflow.com/questions/8481324/contrasting-color-for-nans-in-imagesc
https://www.mathworks.com/matlabcentral/answers/373508-imagesc-how-to-set-nan-as-white-color
https://www.mathworks.com/matlabcentral/answers/81938-set-nan-as-another-color-than-default-using-imagesc

Now, the Matlab documentation says: "The behavior of NaN elements is not
defined."
https://www.mathworks.com/help/matlab/ref/imagesc.html#buxkjup-2l
so there is no need to try and reproduce an undefined behavior. But could the
documentation be updated to either say that the behavior is unpredictable or
that a black color will be used?




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?58325>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/