Re: Org Capture Menu cannot be fully viewed

[TRS-80 <lists.trs-80@isnotmyreal.name>]

On 2020-12-13 16:02, pietru@caramail.com wrote:
> 
> Would that apply with respect to inserting long headings or
> descriptions in org file?

Yes.  If you have not used completing-read, just play around with it a
bit and you will very quickly see how it works.  It takes a list (Elisp
data type) as input, on which you can do narrowing selection as you
type.

Ivy was one of recommendations which I can second, I prefer it's more
intuitive (to me) interaction style and support for native
completing-read format.  But there are (many) others, too.

> Example:
> 
> ;;                "Site_SubType:
> ;;                   [1a] Settlement > Encampment
> ;;                   [1a] Settlement > Hamlet or Village
> ;;                   [1a] Settlement > Town or City
> [...]

However to make it even simpler to use / maintain your candidate lists,
I would just put them in a simple plain text file, aligned to left
margin.  Example:

File name: Site_SubType

[1a] Settlement > Encampment
[1a] Settlement > Hamlet or Village
[1a] Settlement > Town or City

Then you need a function to read from plain text file with your "list"
of candidates, and turn that into an (Elisp data type) list:

#+begin_src emacs-lisp

(defun my-file-to-list (file)
     "Read FILE and return it as a list of strings.

   List items will be split upon newlines."
     (with-temp-buffer
       (insert-file-contents file)
       (split-string (buffer-string) "\n" t)))

#+end_src

You then use the above function (with filename argument) for your
candidate list in completing-read.  Modifying Jean Louis' earlier
example, it now becomes:

#+begin_src emacs-lisp

   (completing-read "Choose: "
		   (my-file-to-list "/path/to/Site_SubType"))

#+end_src

You can even use this to fill in Org Properties.  Or you can use Org
Properties similar native completion, although by default that only uses
whatever values already exist in the buffer (which therefore could be
"none"), instead of your specified controlled vocabulary file as I used
above.  I (by far) prefer the controlled vocabulary method, for lots of
reasons.

Cheers,
TRS-80