matlab-emacs-discuss Mailing List for MATLAB in Emacs (Page 20)

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Hi all,

We have a new pull request on github, does anyone have time to look at
it?
https://github.com/matlab-mode/mirror/pull/2

Simon

I put my changes on the github mirror in a proposed changes branch and requested a pull.

Thanks

________________________________
From: Eric Ludlam
Sent: Tuesday, October 11, 2016 2:26:06 PM
To: Uwe Brauer; John Ciolfi
Cc: mat...@li...
Subject: RE: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

Hi Uwe,

I added you to the Admin group for matlab-emacs.  You should now get the "Admin" tab in sourceforge.

Thanks
Eric

-----Original Message-----
From: Uwe Brauer [mailto:ou...@ma...]

I just realized, that although I have write access I don't have admin rights for the sourceforge repo, which is the primary one anyhow.
I just asked Eric to add you, could you please provide us with our sourceforge uid?

Hi Uwe,

I added you to the Admin group for matlab-emacs.  You should now get the "Admin" tab in sourceforge.

Thanks
Eric

-----Original Message-----
From: Uwe Brauer [mailto:ou...@ma...] 

I just realized, that although I have write access I don't have admin rights for the sourceforge repo, which is the primary one anyhow.
I just asked Eric to add you, could you please provide us with our sourceforge uid?

I'm somewhat familiar with git/github, so if I could get a branch to push to then you could pull that would be best. I sporadically work on emacs and while not the best at elisp, I'm good enough to produce decent quality changes and I can get Eric to look at them ahead of time.

The tab completion fixes won't hurt any compatibility problems. Neither will the debugging change. The change is simply to use the standard MATLAB editor for R2015b and onward. I'm looking to see if we can support native debugging in Emacs which requires MATLAB to give file/line number info when it stops. The warnings fixes shouldn't have altered Xemacs, but I didn't test Xemacs.

Thanks

John

________________________________
From: Uwe Brauer <ou...@ma...>
Sent: Monday, October 10, 2016 3:48:41 PM
To: John Ciolfi
Cc: mat...@li...
Subject: Re: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > Hi

   > I have the following updates, but can send zip files or the diff -r
   > -u output (zip's aren't allowed and the diff -r -u is too long).
   > Any thoughts on how I could get these into the source forge repo?

Thanks for your work. If you want to continue working on matlab.el would
you like to have write access to the repo? Either for sourceforge or the
mirror in github.

There was a discussion which VC system to use and git won. So if you are
familiar with git, things should be easy.

If you don't want all that hassle send me your patch privately and I
check, run some elementary tests and push it.

   > 1. Fixed several problems with TAB completion, and added direct support for
   >    company (complete anything) mode tab completion. If company is installed TAB
   >    key is bound to use company completion and C-TAB is classic emacs TAB
   >    completion.

   > 2. Enabled debugging of matlab code (dbstop, etc.) in R2015b MATLAB and
   >    later. This requires using the MATLAB editor instead of emacs because recent
   >    MATLAB's are not giving enough info to debug within emacs.

Are you sure that does not break backwards compatibility?

   > 3. Fixed/suppressed all warnings as reported by emacs 24.4 (and removed support
   >    for emacs 22 and earlier).

What's about Xemacs, did you removed that support as well?

   > File changes:

   > - company-matlab-shell.el: made completion using company (complete anything) work
   >   with new TAB handling paradigm

   > - linemark.el: fixed compiler warnings

   > - matlab-publish.el: fixed compiler warnings

   > - matlab.el: fixed TAB completion handling. Prior to fix, hitting tab would
   >   generate noise (new unnecessary ">>" prompts) and it didn't handle certain
   >   cases, e.g. ">> ! mv file.<TAB>", ">> set(h,'<TAB>')", etc. would fail to
   >   complete.

I did not notice.

   > - mlint.el: fixed compiler warnings

   > - cedet-matlab.el, semantic-matlab.el, and semanticdb-matlab.el: fixed warnings, though I'm not
   >   sure if these files are still usable because some functions they reference
   >   no longer exist.

   > - tlc.el: fixed compiler warnings, including autoload issue that was preventing use.

   > - toolbox/emacsdocomplete.m: fixed handling of tab completion for commands with strings

   > - toolbox/emacsinit.m: don't activate emacs-based debugging (dbstop handling) in R2015b and
   >   later because the 'hotlinks' emacs needs to debug are no longer available.

Ok

Uwe

Got it. Thanks. I'll try pushing the changes tonight.

________________________________
From: Uwe Brauer <ou...@ma...>
Sent: Monday, October 10, 2016 4:58:26 PM
To: John Ciolfi
Cc: Uwe Brauer; mat...@li...
Subject: Re: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > JohnC32

I just send you an invitation it seems that you have write access.
Please check and report back.

Uwe

>>> "John" == John Ciolfi <Joh...@ma...> writes:

    > Got it. Thanks. I'll try pushing the changes tonight.

Good! But on a separate branch for the moment, please not to master.

John Ciolfi <Joh...@ma...> writes:

> 2. Enabled debugging of matlab code (dbstop, etc.) in R2015b MATLAB and
>    later. This requires using the MATLAB editor instead of emacs because recent
>    MATLAB's are not giving enough info to debug within emacs.

I'm confused: I thought this was fixed already? I am using R2016b with
the latest version of matlab-mode from MELPA and debugging works great
using GUD. I believe someone wrote a patch that prints the correct
hotlinks. Could we have some clarification on this?

> - matlab.el: fixed TAB completion handling. Prior to fix, hitting tab would
>   generate noise (new unnecessary ">>" prompts) and it didn't handle certain
>   cases, e.g. ">> ! mv file.<TAB>", ">> set(h,'<TAB>')", etc. would fail to
>   complete.

This is a great fix!

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > On github, my id is JohnC32 (I don't have one on source forge) and
   > have only used the github id to comment a couple times, so I have a
   > little learning to do.

Ok I try github then, might take a day or two.

Uwe 

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > JohnC32

I just send you an invitation it seems that you have write access.
Please check and report back.

Uwe 

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > Also before I forget, I put a comments into the matlab.el near
   > matlab-shell-tab on what to test for tab completion. If you try some
   > of them, they won't work with the current matlab.el:

Ok the problem is that I don't have access to Matlab 2016b, only to 2014
on certain machines I think.

Also before I forget, I put a comments into the matlab.el near matlab-shell-tab on what to test for tab completion. If you try some of them, they won't work with the current matlab.el:

;;   Test cases (using R2016b):
;;
;;   >> h=figure;
;;   >> h.Num<TAB>                   Should show Number
;;   >> h.Number<TAB>                Should show Number and NumberTitle
;;   >>          h.Num<TAB>          Should do same. The extra white space shouldn't mess things up.
;;
;;   >> h.NumberTitle<TAB>           Should display message "No completions"
;;   >> set(h,'<TAB>                 Should display a long list
;;      type P<TAB>                  Should narrow to Parent, Position, ...
;;
;;   >> !touch file.ext              Assuming no other fil* names in current directory.
;;   >> !mv file.<TAB>               Should complete to file.ext
;;   >> !mv file.ext<TAB>            Should do nothing
;;
;;   >> !/usr/b<TAB>                 Should complete to /usr/bin
;;   >> !/usr/bin/cpp-4.<TAB>        Should complete to something like /usr/bin/cpp-4.9
;;
;;   >> ls /usr/include/byteswap.<TAB>  Some file with a '.' should complete correctly
;;
;;   >> !touch foo.ext1 foo.ext2
;;   >> ! mv foo.<TAB>               Should give options foo.ext1 foo.ext2
;;   >> ! mv foo.ext1<TAB>           Should say no completions
;;   >> ! mv foo.ext1 <TAB>          Should say no completions (note the space before the TAB)
;;
;;   >> vdp
;;   >> get_param('vdp','Pos<TAB>    Should show several completions

Thanks

John

________________________________
From: John Ciolfi
Sent: Monday, October 10, 2016 4:40:12 PM
To: Uwe Brauer
Cc: mat...@li...
Subject: Re: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

On github, my id is JohnC32 (I don't have one on source forge) and have only used the github id to comment a couple times, so I have a little learning to do.

Thanks

John

________________________________
From: Uwe Brauer <ou...@ma...>
Sent: Monday, October 10, 2016 4:36:56 PM
To: John Ciolfi
Cc: Uwe Brauer; mat...@li...
Subject: Re: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I'm somewhat familiar with git/github, so if I could get a branch to
   > push to then you could pull that would be best. I sporadically work on
   > emacs and while not the best at elisp, I'm good enough to produce
   > decent quality changes and I can get Eric to look at them ahead of
   > time.

Good, I just realized, that although I have write access I don't have
admin rights for the sourceforge repo, which is the primary one anyhow.
I just asked Eric to add you, could you please provide us with our
sourceforge uid?

If you prefer github I have to contact a different person because again
I have write but no admin access as I only discovered by now.

   > The tab completion fixes won't hurt any compatibility problems.
   > Neither will the debugging change. The change is simply to use the
   > standard MATLAB editor for R2015b and onward. I'm looking to see if we
   > can support native debugging in Emacs which requires MATLAB to give
   > file/line number info when it stops.

Oh yes I agree, although I am not too impressed by the native Matlab
editor.

   > The warnings fixes shouldn't have altered Xemacs, but I didn't test
   > Xemacs.

I just asked because you explicitly said that you removed  support for
GNU emacs < 21. So I wanted to make sure whether you did that also for
Xemacs, but I am glad to hear that you did not (not that there very many
Xemacs left).

On github, my id is JohnC32 (I don't have one on source forge) and have only used the github id to comment a couple times, so I have a little learning to do.

Thanks

John

________________________________
From: Uwe Brauer <ou...@ma...>
Sent: Monday, October 10, 2016 4:36:56 PM
To: John Ciolfi
Cc: Uwe Brauer; mat...@li...
Subject: Re: [Matlab-emacs-discuss] Fixes for tab completion, debugging in with R2015b and later

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I'm somewhat familiar with git/github, so if I could get a branch to
   > push to then you could pull that would be best. I sporadically work on
   > emacs and while not the best at elisp, I'm good enough to produce
   > decent quality changes and I can get Eric to look at them ahead of
   > time.

Good, I just realized, that although I have write access I don't have
admin rights for the sourceforge repo, which is the primary one anyhow.
I just asked Eric to add you, could you please provide us with our
sourceforge uid?

If you prefer github I have to contact a different person because again
I have write but no admin access as I only discovered by now.

   > The tab completion fixes won't hurt any compatibility problems.
   > Neither will the debugging change. The change is simply to use the
   > standard MATLAB editor for R2015b and onward. I'm looking to see if we
   > can support native debugging in Emacs which requires MATLAB to give
   > file/line number info when it stops.

Oh yes I agree, although I am not too impressed by the native Matlab
editor.

   > The warnings fixes shouldn't have altered Xemacs, but I didn't test
   > Xemacs.

I just asked because you explicitly said that you removed  support for
GNU emacs < 21. So I wanted to make sure whether you did that also for
Xemacs, but I am glad to hear that you did not (not that there very many
Xemacs left).

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I'm somewhat familiar with git/github, so if I could get a branch to
   > push to then you could pull that would be best. I sporadically work on
   > emacs and while not the best at elisp, I'm good enough to produce
   > decent quality changes and I can get Eric to look at them ahead of
   > time.

Good, I just realized, that although I have write access I don't have
admin rights for the sourceforge repo, which is the primary one anyhow.
I just asked Eric to add you, could you please provide us with our
sourceforge uid?

If you prefer github I have to contact a different person because again
I have write but no admin access as I only discovered by now.

   > The tab completion fixes won't hurt any compatibility problems.
   > Neither will the debugging change. The change is simply to use the
   > standard MATLAB editor for R2015b and onward. I'm looking to see if we
   > can support native debugging in Emacs which requires MATLAB to give
   > file/line number info when it stops.

Oh yes I agree, although I am not too impressed by the native Matlab
editor.

   > The warnings fixes shouldn't have altered Xemacs, but I didn't test
   > Xemacs.

I just asked because you explicitly said that you removed  support for
GNU emacs < 21. So I wanted to make sure whether you did that also for
Xemacs, but I am glad to hear that you did not (not that there very many
Xemacs left).

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > Hi

   > I have the following updates, but can send zip files or the diff -r
   > -u output (zip's aren't allowed and the diff -r -u is too long).
   > Any thoughts on how I could get these into the source forge repo?

Thanks for your work. If you want to continue working on matlab.el would
you like to have write access to the repo? Either for sourceforge or the
mirror in github.

There was a discussion which VC system to use and git won. So if you are
familiar with git, things should be easy.

If you don't want all that hassle send me your patch privately and I
check, run some elementary tests and push it.

   > 1. Fixed several problems with TAB completion, and added direct support for
   >    company (complete anything) mode tab completion. If company is installed TAB
   >    key is bound to use company completion and C-TAB is classic emacs TAB
   >    completion.

   > 2. Enabled debugging of matlab code (dbstop, etc.) in R2015b MATLAB and
   >    later. This requires using the MATLAB editor instead of emacs because recent
   >    MATLAB's are not giving enough info to debug within emacs.

Are you sure that does not break backwards compatibility?

   > 3. Fixed/suppressed all warnings as reported by emacs 24.4 (and removed support
   >    for emacs 22 and earlier).

What's about Xemacs, did you removed that support as well?

   > File changes:

   > - company-matlab-shell.el: made completion using company (complete anything) work
   >   with new TAB handling paradigm

   > - linemark.el: fixed compiler warnings

   > - matlab-publish.el: fixed compiler warnings

   > - matlab.el: fixed TAB completion handling. Prior to fix, hitting tab would
   >   generate noise (new unnecessary ">>" prompts) and it didn't handle certain
   >   cases, e.g. ">> ! mv file.<TAB>", ">> set(h,'<TAB>')", etc. would fail to
   >   complete.

I did not notice.

   > - mlint.el: fixed compiler warnings

   > - cedet-matlab.el, semantic-matlab.el, and semanticdb-matlab.el: fixed warnings, though I'm not
   >   sure if these files are still usable because some functions they reference
   >   no longer exist.

   > - tlc.el: fixed compiler warnings, including autoload issue that was preventing use.

   > - toolbox/emacsdocomplete.m: fixed handling of tab completion for commands with strings

   > - toolbox/emacsinit.m: don't activate emacs-based debugging (dbstop handling) in R2015b and
   >   later because the 'hotlinks' emacs needs to debug are no longer available.

Ok

Uwe 

Hi

I have the following updates, but can send zip files or the diff -r -u output (zip's aren't allowed and the diff -r -u is too long). Any thoughts on how I could get these into the source forge repo?

1. Fixed several problems with TAB completion, and added direct support for
   company (complete anything) mode tab completion. If company is installed TAB
   key is bound to use company completion and C-TAB is classic emacs TAB
   completion.

2. Enabled debugging of matlab code (dbstop, etc.) in R2015b MATLAB and
   later. This requires using the MATLAB editor instead of emacs because recent
   MATLAB's are not giving enough info to debug within emacs.

3. Fixed/suppressed all warnings as reported by emacs 24.4 (and removed support
   for emacs 22 and earlier).

File changes:

- company-matlab-shell.el: made completion using company (complete anything) work
  with new TAB handling paradigm

- linemark.el: fixed compiler warnings

- matlab-publish.el: fixed compiler warnings

- matlab.el: fixed TAB completion handling. Prior to fix, hitting tab would
  generate noise (new unnecessary ">>" prompts) and it didn't handle certain
  cases, e.g. ">> ! mv file.<TAB>", ">> set(h,'<TAB>')", etc. would fail to
  complete.

- mlint.el: fixed compiler warnings

- cedet-matlab.el, semantic-matlab.el, and semanticdb-matlab.el: fixed warnings, though I'm not
  sure if these files are still usable because some functions they reference
  no longer exist.

- tlc.el: fixed compiler warnings, including autoload issue that was preventing use.

- toolbox/emacsdocomplete.m: fixed handling of tab completion for commands with strings

- toolbox/emacsinit.m: don't activate emacs-based debugging (dbstop handling) in R2015b and
  later because the 'hotlinks' emacs needs to debug are no longer available.

Thanks
John

Hi, thanks for the great matlab emacs, I have been using since R2011a all the way to R2016a.

I’ve recently updated to Matlab R2016b, and I get a segfault, sometimes accompanied with the message below (and sometimes not).

objc[761]: Class JavaLaunchHelper is implemented in both /Applications/MATLAB_R2016b.app/sys/java/jre/maci64/jre/bin/java and /Applications/MATLAB_R2016b.app/sys/java/jre/maci64/jre/lib/jli/libjli.dylib. One of the two will be used. Which one is undefined.

M-Shell segmentation fault: 11

I can launch Matlab R2016b fine from the command line. 

Does anybody else get this? Any hints to avoid? ATM, I am keeping both R2016a (works with emacs matlab) and (works as a stand alone). 

Best wishes — Cumhur

Hi Dick,

Thank you for you reply. I will take a look at it.

Later I realized that if I manually turn on Semantic from Emacs' Tool
menu, I can get some autocomplete. Not very convenience to use... What
do you use for autocomplete?

Thanks,
Jeff

On Mon, Sep 19, 2016 at 9:10 AM, Dick <rc...@cs...> wrote:
> je...@gm...,
>
> I cannot speak for others, but my experience has been that matlab-emacs
> worked fairly well until matlab version r2015a.  You can see some of Eric's
> commentary in Dec2015 on the matlab-emacs-discuss regarding changes to
> Matlab's output.
>
> I continue to use emacs for matlab because the thought of bringing up a
> separate matlab gui unnerves me, but I don't expect all the bells and
> whistles such as CEDET to work.
>
>
> On Sat, Sep 17, 2016 at 3:53 PM, Jeff <je...@gm...> wrote:
>>
>> Hi, all,
>>
>> I never used emacs before, but decided to try emacs+matlab-emacs+cedet
>> after reading Eric Ludlam's post "MATLAB-Emacs integration is back"
>> posted back in 2009. I cannot get CEDET working and wonder whether
>> someone can help me. Below are details.
>>
>> I installed the emacs 24.5 which came with cedet builtin, then I
>> downloaded matlab-emacs. I set up my emacs init file following the
>> instructions in Eric's post, that is:
>>
>> (add-to-list 'load-path "~/path/to/matlab_dot_el")
>>     (load-library "matlab-load")
>>     ;; Enable CEDET feature support for MATLAB code. (Optional)
>>     (matlab-cedet-setup)
>>
>> When I opened a dot m file, the MATLAB menu was there, and some
>> features listed in Eric's post (syntax highlighting, mlint, etc.)
>> seemed to work, but I don't think the completion engine was working.
>> TAGS and SRecode menus were not there, which might be a sign.
>>
>> I wonder how can I turn CEDET on?
>>
>> Thanks,
>> Jeff
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> Matlab-emacs-discuss mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matlab-emacs-discuss
>
>

Hi, all,

I never used emacs before, but decided to try emacs+matlab-emacs+cedet
after reading Eric Ludlam's post "MATLAB-Emacs integration is back"
posted back in 2009. I cannot get CEDET working and wonder whether
someone can help me. Below are details.

I installed the emacs 24.5 which came with cedet builtin, then I
downloaded matlab-emacs. I set up my emacs init file following the
instructions in Eric's post, that is:

(add-to-list 'load-path "~/path/to/matlab_dot_el")
    (load-library "matlab-load")
    ;; Enable CEDET feature support for MATLAB code. (Optional)
    (matlab-cedet-setup)

When I opened a dot m file, the MATLAB menu was there, and some
features listed in Eric's post (syntax highlighting, mlint, etc.)
seemed to work, but I don't think the completion engine was working.
TAGS and SRecode menus were not there, which might be a sign.

I wonder how can I turn CEDET on?

Thanks,
Jeff

Hi Uwe,

The overlay abstraction should avoid the warnings, but perhaps the defalias isn't working with the byte compiler correctly.  I don't know why those warnings exist.  It may be an eval-and-compile type trick is needed around that block.

There were some other warnings in the list where I suspect we just need to move a definition sooner in the file.  I think they are harmless otherwise.

There are some changes from Emacsen that are pretty old, perhaps we can just claim to not support Emacs 22 or earlier.  That was quite a while ago.

Eric

-----Original Message-----
From: Uwe Brauer [mailto:ou...@ma...] 
Sent: Friday, September 02, 2016 8:09 AM
To: John Ciolfi <Joh...@ma...>
Cc: mat...@li...
Subject: Re: [Matlab-emacs-discuss] compiler warnings

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I just updated to the latest matlab emacs and see a few warnings. I
   > realize some of these are to support older emacs's but some look like
   > they could be fixed (and maybe suppress the others for older
   > releases)? I'm building on Debian 8 using emacs 24.4.1

I am on GNU emacs 25 (git version, after being a die hard Xemacs user for ages).

IMHO the GNU emacs team changes to much the definition of variables (which could break backward compatibility), but I don't want to change the naming of certain variable because GNU emacs 23 and Xemacs users could run into problems, not that there are many left, but...

The most harmful warnings I can detect are those starting with

   > matlab.el:5867:1:Warning: the following functions are not known to be defined:
   >     extent-list, start-itimer, matlab-overlays-at, matlab-overlay-get,

Some of the stuff looks Xemacs specific so maybe we should wrap
a (if    (featurep 'xemacs) around it, Eric any ideas?

Uwe Brauer 

------------------------------------------------------------------------------
_______________________________________________
Matlab-emacs-discuss mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matlab-emacs-discuss

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > It would be good to fix the warnings that are real. For example, if
   > you could you fix the lambda quoting that would help. On emacs 24,
   > one of the lambda issues shows up when you load an m-file. In
   > mlint.el add the #' as in:

   >                           (mapconcat '(lambda (x) x) (cdddr entry) "\\|")
   > =>
   >                           (mapconcat #'(lambda (x) x) (cdddr entry) "\\|")

   > In semanticdb-matlab.el, two occurrences of 

   > 	    (mapcar '(lambda (x)
   > =>
   > 	    (mapcar #'(lambda (x)

Hm ok this change does not break anything in Xemacs, what's about GNU
emacs 23 or inferior??
Any opinions?

BTW this is still just a warning or did you face
any problems with
(mapconcat '(lambda (x) x) (cdddr entry) "\\|")

??

I say I wait a couple of days and if nobody objects I change and push.

Uwe 

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I just updated to the latest matlab emacs and see a few warnings. I
   > realize some of these are to support older emacs's but some look like
   > they could be fixed (and maybe suppress the others for older
   > releases)? I'm building on Debian 8 using emacs 24.4.1

I am on GNU emacs 25 (git version, after being a die hard Xemacs user
for ages).

IMHO the GNU emacs team changes to much the definition of variables
(which could break backward compatibility), but I don't want to change
the naming of certain variable because GNU emacs 23 and Xemacs users
could run into problems, not that there are many left, but...

The most harmful warnings I can detect are those starting with

   > matlab.el:5867:1:Warning: the following functions are not known to be defined:
   >     extent-list, start-itimer, matlab-overlays-at, matlab-overlay-get,

Some of the stuff looks Xemacs specific so maybe we should wrap
a (if    (featurep 'xemacs) around it, Eric any ideas?

Uwe Brauer 

It would be good to fix the warnings that are real. For example, if you could you fix the lambda quoting that would help. On emacs 24, one of the lambda issues shows up when you load an m-file.

In mlint.el add the #' as in:

                          (mapconcat '(lambda (x) x) (cdddr entry) "\\|")
=>
                          (mapconcat #'(lambda (x) x) (cdddr entry) "\\|")

In semanticdb-matlab.el, two occurrences of 

	    (mapcar '(lambda (x)
=>
	    (mapcar #'(lambda (x)

Thanks
John

________________________________________
From: Uwe Brauer <ou...@ma...>
Sent: Friday, September 2, 2016 8:09 AM
To: John Ciolfi
Cc: mat...@li...
Subject: Re: [Matlab-emacs-discuss] compiler warnings

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > I just updated to the latest matlab emacs and see a few warnings. I
   > realize some of these are to support older emacs's but some look like
   > they could be fixed (and maybe suppress the others for older
   > releases)? I'm building on Debian 8 using emacs 24.4.1

I am on GNU emacs 25 (git version, after being a die hard Xemacs user
for ages).

IMHO the GNU emacs team changes to much the definition of variables
(which could break backward compatibility), but I don't want to change
the naming of certain variable because GNU emacs 23 and Xemacs users
could run into problems, not that there are many left, but...

The most harmful warnings I can detect are those starting with

   > matlab.el:5867:1:Warning: the following functions are not known to be defined:
   >     extent-list, start-itimer, matlab-overlays-at, matlab-overlay-get,

Some of the stuff looks Xemacs specific so maybe we should wrap
a (if    (featurep 'xemacs) around it, Eric any ideas?

Uwe Brauer

>>> "John" == John Ciolfi <Joh...@ma...> writes:

   > ?
   > In tlc.el, there's a missing space which results in an error:

   > ;;;###autoload(add-to-list 'auto-mode-alist '("\\.tlc$" .tlc-mode))

   > a space is needed between the '.' and 'tlc', i.e.

   > ;;;###autoload(add-to-list 'auto-mode-alist '("\\.tlc$" . tlc-mode))

   > Could someone fix?

Thanks, fixed and pushed, should be in github and in sf.

Uwe Brauer 

?

In tlc.el, there's a missing space which results in an error:

;;;###autoload(add-to-list 'auto-mode-alist '("\\.tlc$" .tlc-mode))

a space is needed between the '.' and 'tlc', i.e.

;;;###autoload(add-to-list 'auto-mode-alist '("\\.tlc$" . tlc-mode))

Could someone fix?

Thanks

I just updated to the latest matlab emacs and see a few warnings. I realize some of these are to support older emacs's but some look like they could be fixed (and maybe suppress the others for older releases)? I'm building on Debian 8 using emacs 24.4.1

Thanks

John

In matlab-with-emacs-link:
matlab.el:119:8:Warning: reference to free variable `matlab-use-eei'
matlab.el:120:8:Warning: reference to free variable `matlab-eei-process'

In matlab-do-functions-have-end-p:
matlab.el:213:28:Warning: reference to free variable `matlab-defun-regex'

In matlab-font-lock-adjustments:
matlab.el:522:28:Warning: `set-face-underline-p' is an obsolete function (as
    of 24.3); use `set-face-underline' instead.
matlab.el:535:13:Warning: `set-face-underline-p' is an obsolete function (as
    of 24.3); use `set-face-underline' instead.
matlab.el:548:13:Warning: `set-face-bold-p' is an obsolete function (as of
    24.4); use `set-face-bold' instead.
matlab.el:548:24:Warning: `set-face-bold-p' is an obsolete function (as of
    24.4); use `set-face-bold' instead.

In matlab-mode:
matlab.el:1180:21:Warning: `default-fill-column' is an obsolete variable (as
    of 23.2); use `fill-column' instead.
matlab.el:1238:31:Warning: reference to free variable `matlab-defun-regex'

In matlab-semicolon-on-return:
matlab.el:2654:35:Warning: reference to free variable
    `matlab-quiesce-nosemi-regexp'

In matlab-mode-verify-fix-file:
matlab.el:3872:24:Warning: `interactive-p' is an obsolete function (as of
    23.2); use `called-interactively-p' instead.

In matlab-shell-active-p:
matlab.el:4271:31:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer

In matlab-shell-mode:
matlab.el:4476:9:Warning: assignment to free variable
    `matlab-shell-last-error-anchor'

In matlab-shell-previous-matching-input-from-input:
matlab.el:4869:51:Warning: `previous-line' is for interactive use only; use
    `forward-line' with negative argument instead.

In matlab-shell-tab:
matlab.el:5036:24:Warning: display-completion-list called with 2 arguments,
    but accepts only 1

In matlab-output-to-temp-buffer:
matlab.el:5277:56:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer

In matlab-shell-collect-command-output:
matlab.el:5357:19:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer

In matlab-find-other-window-file-line-column:
matlab.el:5445:27:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In matlab-shell-help-mode:
matlab.el:5611:18:Warning: reference to free variable
    `matlab-shell-help-mode-menu'

In matlab-shell-topic-mode:
matlab.el:5650:13:Warning: assignment to free variable `mode-motion-hook'
matlab.el:5651:18:Warning: reference to free variable
    `matlab-shell-topic-mode-menu'

In matlab-shell-topic-browser-create-contents:
matlab.el:5663:52:Warning: `toggle-read-only' is an obsolete function (as of
    24.3); use `read-only-mode' instead.
matlab.el:5673:8:Warning: `toggle-read-only' is an obsolete function (as of
    24.3); use `read-only-mode' instead.

In matlab-shell-topic-highlight-line:
matlab.el:5735:37:Warning: reference to free variable `mouse-grabbed-buffer'
matlab.el:5748:41:Warning: reference to free variable `mode-motion-extent'

In end of data:
matlab.el:5867:1:Warning: the following functions might not be defined at
    runtime: comint-check-proc, comint-mode, gud-mode, gud-set-buffer,
    gud-call, comint-after-pmark-p, comint-next-matching-input-from-input,
    ring-empty-p, ring-ref, ring-insert, comint-simple-send,
    comint-send-string, gud-display-frame
matlab.el:5867:1:Warning: the following functions are not known to be defined:
    extent-list, start-itimer, matlab-overlays-at, matlab-overlay-get,
    matlab-overlay-start, matlab-overlay-end, matlab-next-overlay-change,
    mlint-minor-mode, matlab-point-at-eol, matlab-point-at-bol,
    matlab-delete-overlay, matlab-cancel-timer, matlab-run-with-idle-timer,
    matlab-make-overlay, matlab-overlay-put, mlint-buffer,
    mlint-clear-warnings, mlint-clear-cross-function-variable-highlighting,
    set-extent-begin-glyph, make-extent, make-glyph, matlab-eei-run,
    matlab-eei-eval-region, matlab-previous-overlay-change, view-major-mode,
    event-buffer, event-point, mode-motion-ensure-extent-ok, detach-extent,
    set-extent-endpoints
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/matlab.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile mlint.el

In mlint-reset-program:
mlint.el:98:20:Warning: reference to free variable `mlint-programs'

In mlint-column-output:
mlint.el:168:51:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:169:46:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:170:46:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.

In mlint-run:
mlint.el:214:19:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer
mlint.el:258:41:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:217:51:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:240:31:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:225:52:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:232:55:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:227:56:Warning: `string-to-int' is an obsolete function (as of 22.1);
    use `string-to-number' instead.
mlint.el:250:37:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.
mlint.el:262:27:Warning: (lambda (x) ...) quoted with ' rather than with #'

In mlint-linemark-create-group:
mlint.el:321:20:Warning: `object-name-string' is an obsolete function (as of
    24.4); use `eieio-object-name-string' instead.
mlint.el:388:54:Warning: reference to free variable `mlint-overlay-map'
mlint.el:410:59:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer
mlint.el:450:22:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.
mlint.el:514:22:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.
mlint.el:532:22:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In mlint-clear-warnings:
mlint.el:579:4:Warning: `mapcar' called for effect; use `mapc' or `dolist'
    instead

In mlint-next-buffer:
mlint.el:626:33:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In mlint-prev-buffer:
mlint.el:635:33:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In mlint-next-buffer-new:
mlint.el:652:33:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.
mlint.el:652:31:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In mlint-prev-buffer-new:
mlint.el:668:33:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.
mlint.el:668:31:Warning: `goto-line' is for interactive use only; use
    `forward-line' instead.

In mlint-ediff-cleanup-hook:
mlint.el:824:20:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer
mlint.el:828:26:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/mlint.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile tlc.el

In tlc-mode:
tlc.el:141:21:Warning: `make-variable-buffer-local' not called at toplevel
tlc.el:151:9:Warning: `make-variable-buffer-local' not called at toplevel

In tlc-calc-next-indentation:
tlc.el:245:17:Warning: assignment to free variable
    `indent-because-of-continuation'
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/tlc.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile matlab-publish.el

In matlab-insert-title:
matlab-publish.el:174:4:Warning: `beginning-of-buffer' is for interactive use
    only; use `(goto-char (point-min))' instead.
matlab-publish.el:175:24:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-section:
matlab-publish.el:181:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-description-text:
matlab-publish.el:188:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-bold-text:
matlab-publish.el:213:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-monospaces-text:
matlab-publish.el:220:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-preformated-text:
matlab-publish.el:228:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-equation:
matlab-publish.el:238:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.

In matlab-insert-bullet-list:
matlab-publish.el:248:4:Warning: `previous-line' is for interactive use only;
    use `forward-line' with negative argument instead.
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/matlab-publish.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile linemark.el

In linemark-new-group:
linemark.el:185:50:Warning: `object-name-string' is an obsolete function (as
    of 24.4); use `eieio-object-name-string' instead.
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/linemark.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile semantic-matlab.el
semanticdb-matlab.el: (lambda (x) ...) quoted with ' rather than with #'
semanticdb-matlab.el: (lambda (x) ...) quoted with ' rather than with #'

In toplevel form:
semantic-matlab.el:518:20:Warning: assignment to free variable
    `semantic-idle-summary-function'

In semantic-default-matlab-setup:
semantic-matlab.el:612:9:Warning: assignment to free variable
    `semantic-imenu-bucketize-type-members'

In end of data:
semantic-matlab.el:621:1:Warning: the function `fame-message-nolog' is not
    known to be defined.
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/semantic-matlab.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile semanticdb-matlab.el

In toplevel form:
semanticdb-matlab.el:114:82:Warning: Use `with-current-buffer' rather than
    save-excursion+set-buffer

In semanticdb-matlab-cache-files:
semanticdb-matlab.el:245:18:Warning: reference to free variable
    `semantic-matlab-dependency-system-include-path'
semanticdb-matlab.el:329:8:Warning: (lambda (x) ...) quoted with ' rather than
    with #'
semanticdb-matlab.el:369:8:Warning: (lambda (x) ...) quoted with ' rather than
    with #'

In end of data:
semanticdb-matlab.el:378:1:Warning: the following functions are not known to
    be defined: semanticdb-find-translate-path-default,
    semantic-matlab-find-oldstyle-classes, matlab-shell-active-p,
    matlab-shell-which-fcn, matlab-shell-completion-list
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/semanticdb-matlab.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile srecode-matlab.el
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/srecode-matlab.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile cedet-matlab.el

In matlab-cedet-setup:
cedet-matlab.el:52:19:Warning: reference to free variable
    `srecode-map-load-path'
cedet-matlab.el:52:19:Warning: assignment to free variable
    `srecode-map-load-path'

In end of data:
cedet-matlab.el:62:1:Warning: the function `srecode-map-update-map' is not
    known to be defined.
Wrote /local-ssd/ciolfi/sbtools/apps/emacs-add-ons/src/matlab-emacs/matlab-emacs/cedet-matlab.elc
emacs -Q -batch --no-site-file --eval '(setq debug-on-error t)' -L ./ -L ~/cedet/trunk/lisp/cedet/ --eval '(progn )' -f batch-byte-compile company-matlab-shell.el

In company-matlab-shell-tab:
company-matlab-shell.el:30:23:Warning: reference to free variable
    `comint-prompt-regexp'

In company-matlab-shell-grab-symbol:
company-matlab-shell.el:58:26:Warning: reference to free variable
    `comint-prompt-regexp'

In end of data:
company-matlab-shell.el:84:1:Warning: the function `company-begin-backend' is
    not known to be defined.

2005	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec (3)
2006	Jan	Feb (5)	Mar	Apr	May	Jun (1)	Jul	Aug	Sep	Oct (1)	Nov	Dec
2007	Jan	Feb	Mar (1)	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2008	Jan	Feb	Mar	Apr	May (9)	Jun (1)	Jul (1)	Aug (3)	Sep (1)	Oct	Nov (1)	Dec
2009	Jan	Feb (4)	Mar	Apr	May (2)	Jun (1)	Jul (7)	Aug (1)	Sep (5)	Oct	Nov (6)	Dec (3)
2010	Jan	Feb (10)	Mar (12)	Apr (13)	May (2)	Jun (4)	Jul (4)	Aug (4)	Sep	Oct (4)	Nov (2)	Dec (4)
2011	Jan (11)	Feb	Mar (18)	Apr	May (1)	Jun (12)	Jul (10)	Aug (4)	Sep (4)	Oct (5)	Nov	Dec (10)
2012	Jan (4)	Feb (26)	Mar	Apr (1)	May	Jun (8)	Jul (3)	Aug (1)	Sep	Oct (14)	Nov (1)	Dec (2)
2013	Jan (5)	Feb (2)	Mar (2)	Apr (5)	May (3)	Jun	Jul (8)	Aug (4)	Sep	Oct (7)	Nov (2)	Dec (7)
2014	Jan (14)	Feb	Mar (6)	Apr	May (3)	Jun (2)	Jul (4)	Aug (3)	Sep (7)	Oct (9)	Nov (9)	Dec (5)
2015	Jan (2)	Feb (1)	Mar	Apr (2)	May (1)	Jun (10)	Jul (3)	Aug (4)	Sep (8)	Oct (1)	Nov (3)	Dec (3)
2016	Jan (12)	Feb (59)	Mar (23)	Apr (11)	May (4)	Jun (15)	Jul	Aug	Sep (9)	Oct (19)	Nov (12)	Dec (5)
2017	Jan (1)	Feb (5)	Mar (5)	Apr	May (2)	Jun	Jul (5)	Aug	Sep (3)	Oct (12)	Nov (15)	Dec
2018	Jan (7)	Feb (6)	Mar	Apr	May (1)	Jun	Jul (3)	Aug (2)	Sep (2)	Oct (4)	Nov	Dec
2019	Jan (2)	Feb (9)	Mar (4)	Apr (9)	May (1)	Jun (1)	Jul (1)	Aug (3)	Sep	Oct (2)	Nov (6)	Dec (5)
2020	Jan (9)	Feb	Mar (2)	Apr	May (1)	Jun (2)	Jul	Aug	Sep	Oct (3)	Nov (28)	Dec (5)
2021	Jan (11)	Feb (2)	Mar (2)	Apr (2)	May (15)	Jun (9)	Jul (11)	Aug (1)	Sep	Oct (2)	Nov	Dec (3)
2022	Jan (1)	Feb	Mar	Apr (9)	May (1)	Jun	Jul (1)	Aug (1)	Sep	Oct	Nov	Dec
2023	Jan	Feb	Mar	Apr (12)	May (4)	Jun	Jul (22)	Aug (3)	Sep	Oct (1)	Nov	Dec (14)
2024	Jan	Feb	Mar	Apr (17)	May (35)	Jun (1)	Jul (18)	Aug (31)	Sep (5)	Oct (18)	Nov (20)	Dec (9)
2025	Jan	Feb	Mar	Apr (1)	May (1)	Jun (1)	Jul (5)	Aug	Sep	Oct	Nov	Dec

matlab-emacs-discuss Mailing List for MATLAB in Emacs (Page 20)

matlab-emacs-discuss — Discuss Emacs/MATLAB integration