Regarding the bug in octave bug #55438: [octave forge] (image) Binary input image for edge

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Regarding the bug in octave bug #55438: [octave forge] (image) Binary input image for edge

Ajay Tripuramallu
hello there,
               I am Ajay Tripuramallu pursuing Under graduation in  KL University i am interested in working with octave community. In this process i came across the bug " #55438[octave forge] (image) Binary input image for edge "  That pulled my interest.so i request you  to  acknowledge me with the requirements of the  bug so that i could try to fix the bug.
Ajay Tripuramallu
Reply | Threaded
Open this post in threaded view
|

Re: Regarding the bug in octave bug #55438: [octave forge] (image) Binary input image for edge

nrjank


On Fri, Aug 9, 2019 at 08:38 Ajay Tripuramallu <[hidden email]> wrote:
hello there,
               I am Ajay Tripuramallu pursuing Under graduation in  KL University i am interested in working with octave community. In this process i came across the bug " #55438[octave forge] (image) Binary input image for edge "  That pulled my interest.so i request you  to  acknowledge me with the requirements of the  bug so that i could try to fix the bug.


It may help if you  provide us  with your background. Are you familiar with octave?  With the image package?  Are you familiar with the octave building and patching process?  What parts of the bug report or unclear as to the requirements?
Reply | Threaded
Open this post in threaded view
|

Re: Regarding the bug in octave bug #55438: [octave forge] (image) Binary input image for edge

nrjank

Thank you for the reply,
I am computer science student graduating in kl university . I have good knowledge in python and familiar with matlab and octave as well. I have been working from 4 months on image package. I am aware of octave building as well but i need to know about patching process in octave.
the part where i am unclear is
1.The code part in which the error is in im2bw part or else where ?
2.like there is no proper explanation where the bug has to be fixed and in description i don't find any requirements in the description can you please list out actual requirements .
3.I want a clear picture of bug and details about which has to be fixed so as I can work my knowledge on it
Ajay Tripuramallu



Thank you. That helps. 

First, some minor things,  please always "Reply All" to keep the mailing list copied on emails. I'm not personally familiar with this particular bug, but likely someone else on this list is.  I've copied it again on this reply.

Second, the convention of Octave mailing lists is to "bottom post" like this, entering your reply after the message, trimming unneeded quotes. See 

regarding patching: we create patches as Mercurial changesets. If you're familiar with building Octave, can I assume you're able to clone the image package repository, and then create an hg diff patch? see 

Image package page:

Image pkg repository:

Octave code and patching conventions:
See the section "For More Information" at the bottom of the developers page. It links to several important pages on how Octave coding style, patching conventions, etc, that should be followed when contributing to octave:

I've directly cc'd the current package maintainer who may be able to give more insight on bug 55438

At first glance it appears to be a straightforward bug. It claims Octave's current edge function is not fully MATLAB compatible, as it cannot take a 'binary image' only a grayscale one.  I would suggest confirming the bug (try to use edge on a binary image). If it fails, look into the code of the edge function and see why/how it fails. Then see if that can be corrected. I assume edge is an m-code file, so can just be examined and debugged using the octave editor ("edit edge")

Thanks for your interest in helping improve octave.