GNU bug report logs - #79265
[PATCH] Treat point more consistently in PCM completion

Previous Next

Full log

View this message in rfc822 format

From: help-debbugs <at> gnu.org (GNU bug Tracking System)
To: Dmitry Gutov <dmitry <at> gutov.dev>
Cc: tracker <at> debbugs.gnu.org
Subject: bug#79265: closed ([PATCH] Treat point more consistently in PCM
 completion)
Date: Thu, 21 Aug 2025 01:06:02 +0000

[Message part 1 (text/plain, inline)]

Your message dated Thu, 21 Aug 2025 04:05:38 +0300
with message-id <0a5e4c88-2b50-4a69-b1db-845a41f234cb <at> gutov.dev>
and subject line Re: bug#79265: [PATCH] Treat point more consistently in PCM completion
has caused the debbugs.gnu.org bug report #79265,
regarding [PATCH] Treat point more consistently in PCM completion
to be marked as done.

(If you believe you have received this mail in error, please contact
help-debbugs <at> gnu.org.)


-- 
79265: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=79265
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems

[Message part 2 (message/rfc822, inline)]

From: Spencer Baugh <sbaugh <at> janestreet.com>
To: bug-gnu-emacs <at> gnu.org
Subject: [PATCH] Treat point more consistently in PCM completion
Date: Mon, 18 Aug 2025 14:53:24 -0400

[Message part 3 (text/plain, inline)]

Tags: patch


Properly fix bug#38458, which is fundamentally an issue with
completion-ignore-case, by checking if the completions are
unique ignoring case.  When the completions are unique, the
normal code to delete a wildcard naturally causes point to be
moved to the end of the minibuffer, which is the correct
behavior.

Now that the bug is fixed properly, remove a hack which
previously was used to "fix" it, which made point behave
inconsistently if it was in the middle of the minibuffer versus
at the end of the minibuffer.

* lisp/minibuffer.el (completion-pcm--merge-completions):
Respect completion-ignore-case when checking for completion
uniqueness.
(completion-pcm--string->pattern)
(completion-pcm--optimize-pattern): Allow point at the end of
the pattern.
* test/lisp/minibuffer-tests.el (completion-table-test-quoting)
(completion-test--pcm-bug38458, completion-pcm-test-8): Update
tests for more correct behavior.

In GNU Emacs 30.1.90 (build 9, x86_64-pc-linux-gnu, X toolkit, cairo
 version 1.15.12, Xaw scroll bars) of 2025-08-14 built on
 igm-qws-u22796a
Repository revision: 6adc26ffa74aedbd1cfa9a1ee72073ebccea2b96
Repository branch: emacs-30
Windowing system distributor 'The X.Org Foundation', version 11.0.12011000
System Description: Rocky Linux 8.10 (Green Obsidian)

Configured using:
 'configure --with-x-toolkit=lucid --without-gpm --without-gconf
 --without-selinux --without-imagemagick --with-modules --with-gif=no
 --with-cairo --with-rsvg --without-compress-install --with-tree-sitter
 --with-native-compilation=aot
 PKG_CONFIG_PATH=/usr/local/home/garnish/libtree-sitter/0.22.6-1/lib/pkgconfig/'

[0001-Treat-point-more-consistently-in-PCM-completion.patch (text/patch, attachment)]

[Message part 5 (message/rfc822, inline)]

From: Dmitry Gutov <dmitry <at> gutov.dev>
To: Spencer Baugh <sbaugh <at> janestreet.com>
Cc: Stefan Monnier <monnier <at> iro.umontreal.ca>, 79265-done <at> debbugs.gnu.org
Subject: Re: bug#79265: [PATCH] Treat point more consistently in PCM completion
Date: Thu, 21 Aug 2025 04:05:38 +0300

On 20/08/2025 17:35, Spencer Baugh wrote:

>> So IDK - maybe it's better to use the alien cl-loop, or maybe to go
>> back to seq-some, if no measurements show a cause for worry.
> 
> Eh, I went back to seq-every-p for now.  This shouldn't really cause a
> performance problem, since it will usually just return nil quickly.
> Also it's only run when the user hits TAB.

Sounds good. Its overhead it two dynamic dispatches, so probably fine 
for most uses.

> (I'd rather not use cl-loop, this code is already hard enough to
> understand :) )

All right then. (I like cl-loop ;-( )

>> Just to check: in our limited test examples all elements of COMP match
>> PREFIX already. And the beginning of
>> 'completion-pcm--merge-completions' even 'error'-s if string-match
>> comparison fails.
>>
>> So it's possible that we could omit this comparison from the lambda:
>>
>>    (and (string-prefix-p prefix comp completion-ignore-case)
>>
>> But I suppose that would make it a bigger change, a bit riskier.
> 
> Ah, you're right.  I missed that we already have this property
> guaranteed, but indeed we do.
> 
> So thanks to that, the change can be further simplified.  Removing my
> cond refactoring, the patch is now just replacing
> 
>    (eq t (try-completion prefix comps))
> 
> with
> 
>    (seq-every-p
>      (lambda (comp) (= (length prefix) (length comp)))
>      comps)
> 
> which is a nice simple change.  I added some comments to clarify why
> this is correct, too.

Thanks, pushed to master, this and the the following patch (nice 
optimization!).

And with that, time to close.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.