[Octave-patch-tracker] [patch #8897] Fix regression for coordinate transforms on 3D arrays

[Rik]

Update of patch #8897 (project octave):

                  Status:                    None => Done                   
             Open/Closed:                    Open => Closed                 

    _______________________________________________________

Follow-up Comment #1:

I checked in your patch on the stable branch here
(http://hg.savannah.gnu.org/hgweb/octave/rev/5f62b5dae8b1).

I did a few extra things as well.  We like to include the bug number in the
commit message so you can rapidly search Mercurial to find which patches were
related to a bug.  I did that.  The commit message also has the form of

one-line summary

general description

per-file change notes

You had the first two items, but I added the per-file change notes.

Who knew such simple functions could hide such complexity?  I found, for
example, that in the 1-input case we weren't checking whether the variable was
even numeric.  I added a check for that as well as BIST tests to verify that
the code was correct.

I also changed the error messages to reference by name the input variables
that were incorrect.  For example, 'X, Y, Z must be numeric...'.  We have been
trying to change all such instances in the Octave code base, but there is a
lot of code and it is a work in progress.

Marking as done.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/patch/?8897>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/