bugGNU Octave - Bugs: bug #41581, "axis normal" does not...


bug #41581: "axis normal" does not undo "axis equal"

Submitted by:  Hartmut <hardy>
Submitted on:  Thu 13 Feb 2014 07:27:28 PM UTC  
Category:  Plotting Severity:  3 - Normal
Priority:  5 - Normal Item Group:  Regression
Status:  Fixed Assigned to:  None
Originator Name:  Open/Closed:  Closed
Release:  3.8.0 Operating System:  Any

Add a New Comment (Rich Markup)

You are not logged in

Please log in, so followups can be emailed to you.


( Jump to the original submission)

Mon 17 Feb 2014 02:47:34 AM UTC, comment #6: 

I fixed up "square" and "normal" in this changeset (http://hg.savannah.gnu.org/hgweb/octave/rev/df8e5e3a0f67).  Closing report.

Rik <rik5>
Project Administrator
Sun 16 Feb 2014 05:01:11 AM UTC, comment #5: 

In the end this was rather easy to fix.  I pushed a changeset to the stable branch here (http://hg.savannah.gnu.org/hgweb/octave/rev/343718b2eee4).

It seems like a good idea to check for Matlab conformance on the other two options "square" and "image".  Could someone run the attached tst2.m and then upload the file "diary" which it creates?

(file #30570)

Rik <rik5>
Project Administrator
Sat 15 Feb 2014 11:56:54 PM UTC, comment #4: 

here you go

(file #30565)

Felipe G. Nievinski <fgnievinski>
Sat 15 Feb 2014 07:40:40 PM UTC, comment #3: 

I'll check with Matlab next week.

Hartmut <hardy>
Sat 15 Feb 2014 06:57:23 PM UTC, comment #2: 

Since you have access to Matlab, could you run the script tst.m which I am attaching to this bug report.  It will run through a few axis commands and produce some output and three png files.  If you can upload the output and the three image files I can debug this more.  I think the problem may be localized the file axis.m which would make this relatively easy to fix.

(file #30558)

Rik <rik5>
Project Administrator
Sat 15 Feb 2014 04:46:51 PM UTC, comment #1: 

Sigh  I can confirm this issue with either backend toolkit.

The problem seems to be quite deep in graphics.cc where the data aspect ratio and plot box aspect ratio are not calculated correctly when the plot is made initially.  I'm going to file a new bug report about that and reference this one as a dependency.

Rik <rik5>
Project Administrator
Thu 13 Feb 2014 07:27:28 PM UTC, original submission:  

The following little m-Skript demonstrates the behavior:

close all; figure;
x=[0:0.01:pi]; y=0.55.*sin(x);

title('no axis command yet');

input('go on?');
axis equal;
title('axis equal');

input('go on?');
axis normal;
title('axis normal');

What I expect from reading the octave manual is

  • I plot something
  • axis equal resizes my plot to make the x- and y-scale equal
  • axis normal resets the plot to the default scales

And this is also the behavior that you get from Matlab if you run the very same script.

But the last plot does NOT happen in the Octave 3.8.0 release! The command "axis normal" does something, but not the right thing. Please have a look at the resulting plots yourself.

Further information:

  • This problem exists in octave 3.8.0 with gnuplot as well as with fltk.
  • This problem exists on octave 3.8.0 on Ubuntu Linux as well as on Windows 7 (octave-mxe release 3.8.0-4)
  • This problem exists in octave 3.6.2 with gnuplot as well as wiwth fltk. (Only tested on Windows)
  • This problem does NOT exist in octave 3.2.4. Neither under Linux nor under Windows. (The doesn't seem to be a choice of graphics_toolkit at all in this release.)
Hartmut <hardy>


(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:

Attached Files
file #30580:  tst2.log added by fgnievinski (391B - application/octet-stream)
file #30570:  tst2.m added by rik5 (337B - d2l/unknowntype)
file #30565:  tst.zip added by fgnievinski (18KiB - application/x-zip-compressed)
file #30558:  tst.m added by rik5 (498B - d2l/unknowntype)


Depends on the following items: None found

Items that depend on this one: None found


Carbon-Copy List
  • -email is unavailable- added by fgnievinski (Updated the item)
  • -email is unavailable- added by rik5 (Posted a comment)
  • -email is unavailable- added by hardy (Submitted the item)

    Do you think this task is very important?
    If so, you can add your encouragement to it.
    This task has 0 encouragements so far.

    Only project members can vote.




    Follow 8 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2014-02-17 rik5 StatusIn Progress => Fixed
        Open/ClosedOpen => Closed
    2014-02-16 fgnievinski Attached File- => Added tst2.log, #30580
    2014-02-16 rik5 Attached File- => Added tst2.m, #30570
        StatusConfirmed => In Progress
    2014-02-15 fgnievinski Attached File- => Added tst.zip, #30565
    2014-02-15 rik5 Attached File- => Added tst.m, #30558
    2014-02-15 rik5 StatusNone => Confirmed

    Back to the top

    Powered by Savane 3.6