unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Dmitry Gutov <dmitry@gutov.dev>
To: Spencer Baugh <sbaugh@janestreet.com>
Cc: 65137@debbugs.gnu.org, Stefan Monnier <monnier@IRO.UMontreal.CA>
Subject: bug#65137: 29.1; completion-substring-try-completion doesn't return the longest common substring
Date: Wed, 30 Aug 2023 02:25:41 +0300	[thread overview]
Message-ID: <263f9281-67f4-2eea-ef9b-8dfa67740854@gutov.dev> (raw)
In-Reply-To: <ierv8cx7sym.fsf@janestreet.com>

On 29/08/2023 18:45, Spencer Baugh wrote:

>> Could you describe the usage scenario a little more?
>>
>>  From my brief testing, the current behavior seems okay most of the
>> time: you still get the short input which matches a bunch of strings
>> (e.g. filenames), you can type a little more chars and narrow down.
>>
>> With your change, TAB will insert the most common prefix for all those
>> completions, which in case of project-file can be a pretty long
>> string. Not a huge problem, but on the face of it that doesn't seem
>> like an improvement. So which scenario would that make better?
> 
> As a very concrete example that I ran into frequently, for
> project-find-file if you have two files:
> 
> dir/foo.ml
> dir/foo.mli
> 
> and you input "foo" and press Tab, you get "foo.ml".  But it is then
> impossible to expand that to dir/foo.ml using completion because of this
> bug.  So you have to manually select dir/foo.ml if you want to visit
> that file, either by switching to *Completions* and selecting it or by
> using minibuffer-next-completion.

Now I understand, thank you. I guess this is something that certain 
environments (such as OCaml) are more prone to than others.

Also this problem seems somewhat unique to the default completion 
mechanism. If I use Ivy, or Helm, or Vertico, or even the "bare" 
icomplete-mode, they all have the notion of the currently selected 
completion, with a short key sequence to choose it.

E.g. with icomplete-mode on I would type until the needed completion is 
highlighted in the minibuffer and then press C-j (maybe press C-, or C-. 
to select it).

The default completion, as you say, has the means to do a similar thing 
with M-<up> and M-<down>, but it's less obvious and requires more 
keypresses.

> After this bugfix, inputting "foo" and pressing Tab will expand to
> "dir/foo.ml".
> 
> In general, this bug makes it impossible to input a file name with
> completion (in project-file) if that file name is a prefix of another
> file name.  Like dir/foo and dir/foo.tar, or dir/foo.log and
> dir/foo.log.bak.

Your patch fixes that by expanding inputs into longer lines after TAB 
(meaning, the user will see their input text shift, sometimes 
considerably, to the right, and perhaps feel a little disoriented). It 
might be a minor thing, but a downside nevertheless.

I do wonder what Stefan thinks what would be the right behavior here.





      reply	other threads:[~2023-08-29 23:25 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-07 23:24 bug#65137: 29.1; completion-substring-try-completion doesn't return the longest common substring Spencer Baugh
2023-08-07 23:50 ` Spencer Baugh
2023-08-08  0:41   ` Spencer Baugh
2023-08-08 11:04 ` Eli Zaretskii
2023-08-08 12:40   ` sbaugh
2023-08-30  1:31     ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
2023-09-05 19:51       ` Spencer Baugh
2023-09-05 21:26         ` Stefan Monnier via Bug reports for GNU Emacs, the Swiss army knife of text editors
2023-09-06 11:00         ` Eli Zaretskii
2023-08-25  0:40 ` Dmitry Gutov
2023-08-25  2:30   ` Drew Adams
2023-08-29 15:45   ` Spencer Baugh
2023-08-29 23:25     ` Dmitry Gutov [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=263f9281-67f4-2eea-ef9b-8dfa67740854@gutov.dev \
    --to=dmitry@gutov.dev \
    --cc=65137@debbugs.gnu.org \
    --cc=monnier@IRO.UMontreal.CA \
    --cc=sbaugh@janestreet.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).