Re: [misunderstanding] (was: for a given emacs session: insert a subject with an increasing counter)

[Emanuel Berg <moasenwood@zoho.eu>]

Uwe Brauer wrote:

> I realised that there was a misunderstanding. I thought you
> and others suggest just to use

I didn't know of the closure thing, it is actually an
excellent for this task in particular!

Only I still don't know what total is supposed to express,
nevermind, take a look:

(require 'cl-lib)
(require 'message)

(let ((subject-counter 0))
  (defun insert-subject-counter (total)
    (interactive "nTotal: ")
    (save-mark-and-excursion
      (message-carefully-insert-headers
       (list (cons 'Subject
                   (format "[%s/%s]"
                           (cl-incf subject-counter)
                           total) ))))))

(defalias 'iscr #'insert-subject-counter)

> Without (defvar my-subject-counter 0) (or something
> like this.)
>
> As a matter of fact I deleted the defvar definition from my
> code and it still worked[1] but which puzzled me.

This is a tricky case. No, you don't need the `defvar' to make
globals, you can do it like this

(defun make-global ()
  (setq global 5) )
;; (make-global)
;; global ; 5

I don't recommend it but it is possible. So that isn't where
the shoe hurts but here

> So I thought 
> (setq my-subject-counter (+ my-subject-counter 1))
> defines and sets the variable in a recursive way, but it
> does not. You need a defvar (or a surrounding let).

`setq' doesn't get to do that as (+ my-subject-counter 1) uses
an "void-value variable symbol" in my-subject-counter, to
paraphrase the error message ... (I think you realized that
but I say/write it anyway.)

Also, check out the `1+' and, which is even better here,
`cl-incf' functions!

-- 
underground experts united
https://dataswamp.org/~incal