Re: [Matlab-emacs-discuss] [Matlab+org (use the python eninge or not)] (was: new branches, plans, rename development branch to default, mirror to github)

[martinoidar <mar...@gm...>]

On 23.04.2024 15:33, Uwe Brauer wrote:
>>>> "m" == martinoidar  <mar...@gm...> writes:
> Hi
>
> I decided to start a new thread, since I made some progress and it seems
> there are at least two ways to execute matlab from an org buffer and
> insert the result back into that buffer.
>
>      1. Using the python engine matlab ships (I am not describing the
>         setup now)
>         Disadvantage
>         a. Not all matlab releases support all ptyhon versions. For
>            example ptyhon 3.5 is only supported till 2019a
>
>         b. It takes some time for the python engine to start
>
>         Advantage
>
>         a. The format of the result is very nice.
>
>      2. Using the matlab-shell itself (thanks @Martin for pointing this
>         out)
>         Test Configuration:
>
>         a. Start emacs -Q
>
>         b. Load the following file
>
> --8<---------------cut here---------------start------------->8---
> (setq load-path (cons (expand-file-name "~/emacs/site-lisp/packages/src/Matlab/HG-Git/matlab-sf-hg")  load-path))
> (load-file "/home/oub/emacs/init/custom-init-faces.el")
> (require 'company-matlab-shell)
> (require 'org)
> (require 'gdb-mi)
> (autoload 'matlab-mode "matlab" "Enter Matlab mode." t)
> (setq auto-mode-alist (cons '("\\.m\\'" . matlab-mode) auto-mode-alist))
> (autoload 'matlab-shell "matlab" "Interactive Matlab mode." t)
> (org-babel-do-load-languages
>   'org-babel-load-languages '(
>                               (C . t) ; enable processing C, C++, and D source blocks
>                               (matlab . t)
>                               ;;(perl . t)
>                               (octave . t)
>                               (org . t)
>                               (python . t)
>                               (plantuml . t)
>                               (shell . t)
>                               ))
>
> (setq org-babel-default-header-args:matlab
>     '((:results . "output") (:session . "*MATLAB*")))
>
> --8<---------------cut here---------------end--------------->8---
>
> And load ob-octave-fix.el
>   https://github.com/karthink/.emacs.d/blob/master/plugins/ob-octave-fix.el
>
> Then the following code can be executed in a org buffer
>
> #+begin_src matlab :results output latex :exports results  :eval never-export :wrap latex
> clear all
>    x = [1, 2, 3, 4, 5];
>    disp(sprintf('|%d', x))
> #+end_src
>
> #+RESULTS:
> #+begin_latex
> x = [1, 2, 3, 4, 5];
> disp(sprintf('|%d', x))
> |1|2|3|4|5
> #+end_latex
>
>
> #+begin_src matlab :results output latex :exports results  :eval never-export :wrap latex
>    clear all
>    x = [1, 2, 3, 4, 5];
>    disp(sprintf('|%d', x))
> #+end_src
>
> #+RESULTS:
> #+begin_latex
> |1|2|3|4|5
> #+end_latex
>
> That already indicates that the formatting needs care, moreover
> =fprintf= is not supported
> #+begin_src matlab :results output latex :exports results  :eval never-export :wrap latex
>    clear all
>    x = [1, 2, 3, 4, 5] ;
>    fprintf('|%d', x)
> #+end_src
>
> #+RESULTS:
> #+begin_latex
> #+end_latex
>
> So I need to investigate this a bit more.
>
> In summary: the second solution works surprisingly well and is faster
> than the python solution, but the formatting seems to me more restricted
> as in the python case.
>
> Any comments?
>
> Has  anybody tried  this out?
>
> @John you talked about such  feature in the past.
>
> Can you confirm this behavior?
>
> Regards
>
> Uwe


Hi Uwe,

My comments on the topic have grown to such a size that I decided it's 
better to make a post from them.

Please have a look here:

https://cissic.github.io/posts/latex-output-from-matlab-octave-source-blocks/

Regards,

M