Trying to understand 'icomplete-in-buffer'

Trying to understand 'icomplete-in-buffer'

[Gabriel]

Hi,

I am trying to understand the variable 'icomplete-in-buffer' in
lisp/icomplete.el.

It's a defvar, but users continues to set this variable. I believe it's
due to a lack of documentation or due to a lack of double-dash to mark
it as internal. Some examples:

[1] https://www.reddit.com/r/emacs/comments/fswt7c/using_icomplete_vertically/
[2] https://www.reddit.com/r/emacs/comments/ftq4pq/making_icomplete_more_modern/
[3] https://emacs.stackexchange.com/questions/63340/disable-icomplete-with-emacs-find-file

The bug bug#41805 was opened because of this problem, and closed because
it was a defvar, but there was no clarification regarding the use of
this variable.

Looking at the code, it was introduced in commit 504a03813b (8 years
ago). There are only two references for this variable in the emacs
source tree in the original commit and today: 1) the variable
declaration with a nil value and 2) a condition on 'icomplete-mode' to
add the the function 'icomplete--in-region-setup' to hook
'completion-in-region-mode-hook'. I could not find any external package
that uses this variable.

I tried many approaches to use this variable in combination with
'completion-at-point', but in all cases I got either no results or a
hook error.

Any help is welcome.

Regards,
Gabriel

Re: Trying to understand 'icomplete-in-buffer'

[Stefan Monnier]

> It's a defvar, but users continues to set this variable. I believe it's
> due to a lack of documentation or due to a lack of double-dash to mark
> it as internal. Some examples:

I wrote it this way because I consider(ed) that code just "proof of
concept" and not working well enough to inflict it on unsuspecting users.


        Stefan

Re: Trying to understand 'icomplete-in-buffer'

[Gabriel]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> I wrote it this way because I consider(ed) that code just "proof of
> concept" and not working well enough to inflict it on unsuspecting users.
>
>
>         Stefan

Hi Stefan,

Thanks for your response. If you think there is a room for improvements
in this or other parts of the code, tell me and I will be glad to
contribute. It's hard to tell why a code it's the way it's and what
pieces can be changed.

Thanks,
Gabriel

Re: Trying to understand 'icomplete-in-buffer'

[Stefan Monnier]

Gabriel [2021-07-04 12:43:56] wrote:
> Stefan Monnier <monnier@iro.umontreal.ca> writes:
>> I wrote it this way because I consider(ed) that code just "proof of
>> concept" and not working well enough to inflict it on unsuspecting users.
> Thanks for your response. If you think there is a room for improvements
> in this or other parts of the code, tell me and I will be glad to
> contribute. It's hard to tell why a code it's the way it's and what
> pieces can be changed.

The problem is not so much in the code but in the behavior.
If you think the behavior is good enough, then by all means use it, but
I think it's a bit rough around the edges.


        Stefan