GNU bug report logs - #1062
23.0.60; minibuffer-completion-help incomplete

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 1062 in the body.
You can then email your comments to 1062 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> emacsbugs.donarmstrong.com (full text, mbox):

From: Stephen Berman <stephen.berman <at> gmx.net>
To: emacs-pretest-bug <at> gnu.org
Subject: 23.0.60; minibuffer-completion-help incomplete
Date: Wed, 01 Oct 2008 12:00:52 +0200

GNU Emacs 23.0.60.5 (i686-pc-linux-gnu, GTK+ Version 2.12.9) of
2008-10-01 on escher

1. emacs -Q
2. Let some directory contain the files "tested" and "Testing" and no other
   files beginning with "test" or "Test".
3. M-x set-variable RET read-file-name-completion-ignore-case RET t RET
4. C-x C-f <directory/in/step/2>/test
5. Now type TAB or "?" and a *Completions* buffer pops that shows only
   "tested" as a possible completion.  However, typing "i" next and then
   TAB does complete to "Testing".

If before doing step 4 completion-styles is set to (emacs22), then after
the first TAB or ? in step 5 the *Completions* buffer shows both
"tested" and "Testing".

The following patch makes the *Completions* buffer show all possible
completions in step 5 also with the default completion-style basic, but
I don't know if it has undesirable consequences elsewhere.

Steve Berman

*** emacs/lisp/minibuffer.el.~1.58.~	2008-09-04 16:16:19.000000000 +0200
--- emacs/lisp/minibuffer.el	2008-10-01 11:42:40.000000000 +0200
***************
*** 1316,1322 ****
                     "" (list (substring beforepoint (car bounds))
                              'point
                              (substring afterpoint 0 (cdr bounds)))))
!          (all (completion-pcm--all-completions prefix pattern table pred)))
      (completion-hilit-commonality
       (if (consp all) (nconc all (car bounds)) all)
       point)))
--- 1316,1324 ----
                     "" (list (substring beforepoint (car bounds))
                              'point
                              (substring afterpoint 0 (cdr bounds)))))
!          (all (if completion-ignore-case
! 		  (all-completions (substring string 0 point) table pred)
! 		(completion-pcm--all-completions prefix pattern table pred))))
      (completion-hilit-commonality
       (if (consp all) (nconc all (car bounds)) all)
       point)))

Message #10 received at 1062-done <at> emacsbugs.donarmstrong.com (full text, mbox):

From: Chong Yidong <cyd <at> stupidchicken.com>
To: Stephen Berman <stephen.berman <at> gmx.net>
Cc: 1062-done <at> debbugs.gnu.org
Subject: Re: 23.0.60; minibuffer-completion-help incomplete
Date: Sat, 08 Nov 2008 11:57:14 -0500

> 1. emacs -Q
> 2. Let some directory contain the files "tested" and "Testing" and no
> other
>    files beginning with "test" or "Test".
> 3. M-x set-variable RET read-file-name-completion-ignore-case RET t RET
> 4. C-x C-f <directory/in/step/2>/test
> 5. Now type TAB or "?" and a *Completions* buffer pops that shows only
>    "tested" as a possible completion.  However, typing "i" next and then
>    TAB does complete to "Testing".
>
> The following patch makes the *Completions* buffer show all possible
> completions in step 5 also with the default completion-style basic, but
> I don't know if it has undesirable consequences elsewhere.

The bug was actually in the C code: file-name-all-completions did not
handle completion-regexp-list properly.  I've checked in a fix.  Thanks
for the bug report.

Previous Next

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