bugGNU Octave - Bugs: bug #47536, "test imread" produces a...

 
 

bug #47536: "test imread" produces a warning from randi

Submitter:  Hartmut <hardy>
Submitted:  Fri 25 Mar 2016 07:10:19 PM UTC
   
 
Category:  Octave Function Severity:  3 - Normal
Priority:  5 - Normal Item Group:  None
Status:  Invalid / Not an Octave Bug Assigned to:  None
Originator Name:  Open/Closed:  * Closed
Release:  * dev Operating System:  * Microsoft Windows
Fixed Release:  None Planned Release:  None
* Mandatory Fields

Add a New Comment Rich Markup
   

Jump to the original submission

Sun 27 Mar 2016 06:22:36 PM UTC, comment #9: 

No problem.

Mike Miller <mtmiller>
Group Member
Sun 27 Mar 2016 06:01:39 PM UTC, comment #8: 

I've now done an mxe-octave build of the dev branch with a freshly cloned mxe-repository. When I run this under Windows, the original strange behavior of this bug report is fully ONE.

Please go ahead and close this report. And a big sorry for the confusion on my side. (I just now learned how to correctly build the dev branch with mxe-octave for a non-first time.)

Hartmut <hardy>
Sat 26 Mar 2016 07:24:50 PM UTC, comment #7: 

I haven't built mxe-octave in a long time, but I think it caches the source packages it has downloaded. When you build with --enable-octave=default, it downloads source tarballs from the hydra autobuilder. If you have a pkg/octave-4.1.0+.tar.gz, you should probably delete it and try building again.

Mike Miller <mtmiller>
Group Member
Sat 26 Mar 2016 07:05:34 PM UTC, comment #6: 

Here is the file randi.m attached. This is the one my Octave on Windows (supposedly freshly build from dev branch) used. In line 109 this file has the issued warning string "randi: maximum integer IMAX or range exceeds requested ".

Otherwise I fully agree with your observations, that this string is currently nowhere in today's version of the Octave repository.

Before building this weired mxe-octave version (with the option ./configure --enable-octave=default to use the dev branch of Octave) I properly cleaned my local mxe-octave repository by deleting the whole dist folder and running "make clean". Afterwards tortoise-hg told my that there wheren't any additional local files in my local repository any more.

Tonight I will try to once again build mxe-octave from the dev branch. This time I will delete my whole local repository and start with a fresh clone of the repo. I'll let you know if that helps.

(file #36772)

Hartmut <hardy>
Sat 26 Mar 2016 03:30:23 PM UTC, comment #5: 

When I built the dev system with the new textscan was
not removed when doing the make install. I had to manually
remove it
rm /usr/local/share/octave/4.1.0+/m/io/textscan.m

Michael Godfrey <godfrey>
Group Member
Sat 26 Mar 2016 03:23:35 PM UTC, comment #4: 

Ok, reopening. This still doesn't make sense to me because the warning message you are seeing simply does not exist in the default branch anymore. There should be no way for that message to be shown unless you have some old code being built.

Can you attach the complete randi.m function file that is being used in your Windows environment? Use "which randi" to ensure you're grabbing the right one.

Mike Miller <mtmiller>
Group Member
Sat 26 Mar 2016 01:29:44 PM UTC, comment #3: 

@Mike: Please REOPEN this issue.

I have now also done a clean rebuild of the current dev branch for Windows, via mxe-octave. And to my own suprise, the original warning message is still there! (So it did not have anything to do with bug #47372.)

The origin of the warning message in "test imread" is the following line:


w = randi (255, 100, 100, "uint8");


This gives me the originally quoted warning message under Windows! The same line of code, run in the linux program of the (nominally) same Octave source code does NOT give any warnings.

Note: If I change the first number to 254 then the warning message is not issued any more.

So could you please REOPEN this issue, unless you can confirm that this warning does NOT appear on a current WINDOWS build of the dev branch. You might also want to change the OS tag of this issue to "Windows".


Hartmut <hardy>
Fri 25 Mar 2016 09:13:38 PM UTC, comment #2: 

You were right: I could NOT reproduce this behavior with a fresh Linux built of the dev version.

I will now also re-compile the mxe-version for Windows with a clean dev branch. This bug report was origionally based on a compiled version that contained an additional tiny patch (from bug #47372, comment #12). Maybe this patch provoked the behavior of this bug report as a side effect. In this case I will mention it in bug #47372, otherwise I will also consider this to be irrelevant.

Hartmut <hardy>
Fri 25 Mar 2016 07:52:22 PM UTC, comment #1: 

Thanks for the bug report. No, there is no warning in my testing on Debian.

In fact that warning message is not in either the current stable or default branches, so I'm not sure that you are building a current up-to-date development version of Octave.

I'm going to close as invalid because I don't think this message is currently possible with the version that you say you are using. Please ensure that you are up to date and try building again and follow up if you can reproduce this warning message.

Mike Miller <mtmiller>
Group Member
Fri 25 Mar 2016 07:10:19 PM UTC, original submission:  

In the current dev version (default branch in the repository), I get a warning from randi when I run the tests of imread:


>> test imread
warning: randi: maximum integer IMAX or range exceeds requested type.  Values might be truncated t
o requested type.
PASSES 5 out of 5 tests


I have NOT had a deeper look at all the other tests.

I observed this behavior with a current mxe-octave version of the current Octave "default" branch under Windows 7. But I assume it will be the same under Linux as well.

Hartmut <hardy>

 

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

Attach Files:
   
   
Comment:
   

Attached Files
file #36772:  randi.m added by hardy (6KiB - 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 godfrey (Posted a comment)
  • -email is unavailable- added by mtmiller (Posted a comment)
  • -email is unavailable- added by hardy (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

    Only group members can vote.

     

    Follow 8 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2016-03-27 mtmiller StatusWorks For Me Invalid / Not an Octave Bug
        Open/ClosedOpen Closed
    2016-03-26 hardy Attached File- Added randi.m, #36772
    2016-03-26 mtmiller StatusInvalid / Not an Octave Bug Works For Me
        Open/ClosedClosed Open
        Operating SystemAny Microsoft Windows
    2016-03-25 mtmiller StatusNone Invalid / Not an Octave Bug
        Open/ClosedOpen Closed

    Back to the top

    Powered by Savane 3.13-caa5.
    Corresponding source code