Re: [Matlab-emacs-discuss] matlab-mode, electric-pair-mode and the single quote (')

[Eric L. <eri...@gm...>]

Hi Peter,

Sorry, I completely lost track of this thread when it came in.

Thanks for trying out the patch.  I'm glad it started working for you!

Regarding your questions:
1) That defvar tells the byte-compiler that variable exists so it 
doesn't throw a warning when compiling in batch-mode.  You are correct 
that it has no other functional value.
2) \\s_ matches symbol characters. This if you type MyVar_'  it will 
correctly match.

4) Yes, nconc modifies the list.  You would need to make a deep copy of 
electric-pairs-pairs first before modifying it.  Thus, your soln of just 
typing in something is good. :)
5) Use 'toggle-debug-on-error' for your first file and see what the 
stack says.  Hard to speculate otherwise.  Run the toggle-debug command 
a 2nd time to turn off debugging.

Good Luck
Eric

On 4/28/22 11:24 PM, Peter Mao wrote:
> Hi Eric,
>
> Thank you -- that inspired me to make electric-pair-pairs a buffer 
> local variable.  Surprisingly (to me) electric still pairs 
> double-quote (") as I would expect and single-quote (') works in all 
> the contexts I can think of right now.
>
> questions/comments:
> 1.  Your `(defvar electric-pair-inhibit-predicate)` did not seem 
> necessary, so I removed it in my patch
> 2.  I don't understand the `\\s_` option in your `looking-at` 
> argument, but kept it in, as it seems to do no harm to any of my 
> matlab code.
> 3.  I added the `\\.` to `looking-at` to handle non-conjugate transpose.
> 4.  I tried using `nconc` to add single quote to the local 
> `electric-pair-pairs`, but it would always change the global value as 
> well, so I fell back to `setq`.  This is some bit of elisp that I just 
> don't understand.
> 5.  I'm using emacs 28.1 now with the native-compilation on.  I find 
> that the changes to matlab.el don't work for the first m-file opened, 
> but on the second one (or kill and open the first) all the electric 
> mods work as desired.
>
> Peter
>
> On Tue, Apr 26, 2022 at 4:10 PM Eric Ludlam <eri...@gm...> wrote:
>
>     Hi, yes - I meant `show-paren-mode'.  Not sure what I was thinking
>     at the time.
>     In the code for matlab.el, it is all about show-paren-* features
>     it is extending.
>
>     I poked around some more to see how to extend electric-pair-mode,
>     and found the defcustom `electric-pair-inhibit-predicate'.  I
>     hacked together a little thing that prevents the mode from pairing
>     up a single quote after a symbol, ie - when used as transpose. The
>     downside is it also disables the user's selection, so not sure
>     what to do about that.
>
>     Anyway, I attached it as a way to brainstorm what the right
>     patterns are to fix the original problem.
>
>     Eric
>
>     On Tue, Apr 26, 2022 at 2:30 AM Uwe Brauer <ou...@ma...> wrote:
>
>         >>> "PM" == Peter Mao <pet...@gm...> writes:
>
>         Hi all
>         > hi Eric and John,
>         > Thanks for your responses.  Eric -- did you mean
>         `show-paren-mode`?  I
>         > can't find any references to a `paren-mode` in melpa, elpa
>         or emacs
>         > (28.1).  I'll try it out if I can find it!
>
>         I was wondering myself. I tend to believe that years ago there
>         was some
>         paren-mode package, but I cannot find it in my archives.
>
>         Uwe Brauer
>         _______________________________________________
>         Matlab-emacs-discuss mailing list
>         Mat...@li...
>         https://lists.sourceforge.net/lists/listinfo/matlab-emacs-discuss
>