File for minor mode not found

[fatiparty--- via Users list for the GNU Emacs text editor <help-gnu-emacs@gnu.org>]

Jan 10, 2022, 18:09 by tomas@tuxteam.de:

> On Mon, Jan 10, 2022 at 01:19:55PM +0100, fatiparty@tutanota.com wrote:
>
>> Jan 9, 2022, 22:10 by tomas@tuxteam.de:
>>
>
> [...]
>
>> > Well, typically some prefix of the path element has a significance. Say,
>> > for example "~/Admin/bin/gadi-1.0" be gadi-files or so. That's the way I
>> > usually do it.
>> >
>> > Bonus points: you only get to change one when that changes :-)
>> >
>> That is correct.  Would one use setq or something buffer local.  How would.  
>> for instance, gadi-files be constructed?
>>
>
> This is a matter of style and taste -- and can vary a lot.
>
> If there is just one place where you do it (quite possible) then I
> wouldn't even bother (someone else might, though). And actually, I would
> try to keep it that way.
>
> But if I found myself doing `(add-to-list 'load-path "~/Admin/bin/gadi-1.0/...")',
> I'd be tempted to do something along these lines:
>
>  (defvar my-gadi-basedir "~/Admin/bin/gadi-1.0"
>  "the base dir for gadi programs")
>
>  (defun my-gadi-add-to-load-path (dir)
>  (add-to-list 'load-path
>  (file-name-concat my-gadi-basedir dir)
>

With file-name-concat, I get 

Symbol's function definition is void: file-name-concat

I am using GNU Emacs 27.2 of 2021-09-28


> and then:
>
>  (my-gadi-add-to-load-path "typex/ziggurat-protowrit/daph-ideograms")
>
> I'm assuming that your "package prefix" is `my-gadi'. It is, of course,
> a judgement call whether or when this extra complication is worth the
> hassle. And there are (at least) 43 other ways to skin that cat.
> Besides, my choice of the split point for -basedir is just based on a
> very inexact feeling: you have more details.
>
> So take this just as a rough sketch for inspiration.
>
> Cheers
> -- 
> t
>