Is dmp a reasonable prefix for global variables?

I once used my-* for my global variables but abandoned it
when other people started to use my code.

It is rather unlikely that there is a dmp out there in the world today.

Lisp namespaces should be allocated on a first in first served basis,
which is how internet domain names are allocated.



On Thu, 14 Jul 2022 at 13:48, Stefan Monnier <monnier@iro.umontreal.ca>
wrote:

> > I use debugging checkpoints every second line
> > so the following code cannot be written:
> >
> > (defun foo ()
> >    (progn
> >       (progn
> >          (zip)
> >          (message "&apple:1")
> >          (zap)
> >          (message "&apple:2")
> >          (boo)
> >          (message "&apple:3")
> >          (bum)
> >          (message "&apple:4"))))
>
> I do use separate lines for closing parens every once in a while, don't
> get me wrong.  It's just an exception rather than a rule.
> [ Note also that inserting/removing such `message` needs to be different
>   for the last element in any case because it determines the return
>   value.  ]
>
> >> importantly:
> >> - should use `with-current-buffer`.
> >
> > is that instead of (set-buffer buf)?
>
> If you compile the code, the byte-compiler will give you a warning
> suggesting to replace with-current-buffer for save-excursion + set-buffer.
>
> >> - should not `setq` on a global variable that doesn't have an
> appropriate
> >>   namespace prefix.
> >>
> > Do you mean the following is not allowed:
> >
> > (setq apple 'banana)
> >
> > but the following code is allowed:
> >
> > (setq dmp-apple 'banana)
>
> Neither are recommended without first declaring that global variable
> (since that's what it is: a global variable).
> And global declarations need to use a proper namespace prefix.
> Otherwise you're at risk of stepping on someone else's toes/variables.
>
>
>         Stefan
>
>