bugGNU Octave - Bugs: bug #46034, Undo some changes from Changeset...

 
 

bug #46034: Undo some changes from Changeset 564e59f75ef8

Submitted by:  Kai Torben Ohlhus <siko1056>
Submitted on:  Wed 23 Sep 2015 09:05:13 AM UTC  
 
Category: InterpreterSeverity: 2 - Minor
Priority: 5 - NormalItem Group: Matlab Compatibility
Status: DuplicateAssigned to: Kai Torben Ohlhus <siko1056>
Originator Name: Open/Closed: Closed
Release: devOperating 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 Jump to the original submission)

Thu 12 Jan 2017 10:19:28 PM UTC, comment #12:

For the record, Symbolic now uses the new "[format, formatspacing] = format()" call [1] and it works great---thanks.

[1] https://github.com/cbm755/octsympy/commit/deeca1c883bf471a6bde922a8710b7eb6ccdfd06

Colin Macdonald <cbm>
Thu 29 Dec 2016 11:34:26 PM UTC, comment #11:

Since Matlab no longer "officially" supports this syntax I think it is okay to be different. Early on TMW made some bad architectural decisions. They couldn't figure out where to locate various user preferences so they put them in to the root figure, into the userdata property of graphics objects, and eventualy into the addpref mechanism. I've changed format() and diary() to be sensible and return the current state if called with no arguments. It is on my list to transform any other properties stored in the root figure in the same manner.

Rik <rik5>
Project Administrator
Thu 29 Dec 2016 09:50:39 PM UTC, comment #10:

Thanks, I just missed comment #9 when I made up a new bug report.
This does work.

We're also deliberately Matlab-incompatible. But as I stated elsewhere TMW doesn't seem to have made up their mind as to really deprecating & removing the relevant root properties query mechanisms.

Philip Nienhuis <philipnienhuis>
Project Member
Thu 29 Dec 2016 04:56:30 PM UTC, comment #9:

In Octave, these properties are now available from the format() function. Try 'help format' on the dev branch.

Usage:

This won't help for stable unless the change I made in this cset from a few weeks ago is backported.

Rik <rik5>
Project Administrator
Thu 29 Dec 2016 01:45:05 PM UTC, comment #8:

In Matlab 2017a prerelease:

I hit this when trying the symbolic package (sym-py-bundle 2.4.0) that complains about missing _compactformat_.
The fix in symbolic/@syms/display.m still depends on these root properties.

Philip Nienhuis <philipnienhuis>
Project Member
Sat 26 Sep 2015 11:42:52 PM UTC, comment #7:

It does seem that Matlab is keeping them around. I was
not as careful as I should have been when checking my
2015a system. get(0) does not list them, but

Michael Godfrey <godfrey>
Project Member
Fri 25 Sep 2015 12:20:07 PM UTC, comment #6:

Good catch Kai. Confirmed as well on 2015b prerelease.

So Matlab apparently doesn't want to expose those properties to users but perhaps needs it for some undocumented purpose - or maybe plain backward compatibility as you say.

Philip Nienhuis <philipnienhuis>
Project Member
Thu 24 Sep 2015 10:09:55 PM UTC, comment #5:

They are gone as graphic root properties, but I expect for compatibility issues, they are still supported. On my University System I get the following:

Kai Torben Ohlhus <siko1056>
Project MemberIn charge of this item.
Thu 24 Sep 2015 04:15:54 PM UTC, comment #4:

Yep, I see the same in r2015b prerelease.

However, in bug #46000 comment #5 Kai claims that he still found them in r2015b (final release?).

Philip Nienhuis <philipnienhuis>
Project Member
Wed 23 Sep 2015 11:36:41 PM UTC, comment #3:

I tried matlab R2015a and find that they are gone...

Michael Godfrey <godfrey>
Project Member
Wed 23 Sep 2015 12:15:31 PM UTC, comment #2:

In Matlab r2014a:

...so indeed they're still there.
The full output for each step is attached for reference (for if there's discussion about other properties).

At home I can try with r2015b prerelease (if not expired); r2015b is out but I haven't been able to get it installed at work.

(file #34962)

Philip Nienhuis <philipnienhuis>
Project Member
Wed 23 Sep 2015 09:29:18 AM UTC, comment #1:

This bug depends on bug #46000. The question asked on that report is which Matlab documentation is correct? The root properties are not documented to have "Format" and "FormatSpacing". If someone has access to Matlab and can verify that they still exist in a new version of Matlab then we can re-instate these two properties.

Rik <rik5>
Project Administrator
Wed 23 Sep 2015 09:05:13 AM UTC, original submission:

In changeset http://hg.savannah.gnu.org/hgweb/octave/rev/564e59f75ef8 two graphics root properties:

- format
- formatspacing

have been removed, due to depreciation in MATLAB. According to their documentation https://www.mathworks.com/help/matlab/ref/format.html they still exists without obvious deprecation.

Any obligations in reintroducing these properties again, as I use them via get(0,'formatspacing'). A dirty hack around it for me currently is checking the size of PI

length (int32 (disp (pi)))

Or are there different ways to query for 'compact' or 'loose'?

Regards,
Kai

Kai Torben Ohlhus <siko1056>
Project MemberIn charge of this item.

 

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

Attach File(s):
   
   
Comment:
   

Attached Files
file #34962:  getParentML.txt added by philipnienhuis (6KiB - text/plain)

 

Digest:
   bug dependencies.

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by cbm (Posted a comment)
  • -unavailable- added by godfrey (Posted a comment)
  • -unavailable- added by rik5 (Posted a comment)
  • -unavailable- added by siko1056 (Submitted the item)
  •  

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

    Only project members can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 5 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Wed 23 Sep 2015 12:15:31 PM UTCphilipnienhuisAttached File-=>Added getParentML.txt, #34962
    Wed 23 Sep 2015 10:03:59 AM UTCsiko1056StatusNeed Info=>Duplicate
      Open/ClosedOpen=>Closed
    Wed 23 Sep 2015 09:24:55 AM UTCrik5Dependencies-=>Depends on bugs #46000
    Wed 23 Sep 2015 09:05:29 AM UTCsiko1056Assigned torik5=>siko1056

    Back to the top


    Powered by Savane 3.1-cleanup1