unofficial mirror of bug-gnu-emacs@gnu.org 
 help / color / mirror / code / Atom feed
From: Jambunathan K <kjambunathan@gmail.com>
To: Stefan Monnier <monnier@iro.umontreal.ca>
Cc: 12638@debbugs.gnu.org
Subject: bug#12638: 24.2.50; FR: Some suggestions for icomplete-mode
Date: Thu, 13 Dec 2012 19:21:06 +0530	[thread overview]
Message-ID: <87bodykqlx.fsf@gmail.com> (raw)
In-Reply-To: <87391ieck9.fsf@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 1147 bytes --]

Jambunathan K <kjambunathan@gmail.com> writes:

> Stefan Monnier <monnier@iro.umontreal.ca> writes:
>
>>> 2. M-x debbugs-gnu
>>>    There is no way I can C-s C-j to the second or third candidates.
>>>    I normally use debbbugs-gnu-search which is the second candidate.
>>
>> Better ask this elsewhere, I'm not up to speed on debbugs-gnu.
>
> I am attaching screenshot of minibuffer. 
>
>     1. M-x debbugs-gnu
>     2. Immediately cycle with C-s
>
> Cycling may happen internally but the UI presents the "most-is-exact"
> always at the head.  This is a definite bug.

The case to be handled is "Complete, but not unique".  

The attached patch fixes the issue by sneaking a Ctrl-M in to the try
try completion field.

Attachments: 
M-x debbugs-gnu RET with icomplete-mode on.

icomplete-1.png:  
        Note the ^M and the empty first entry

icomplete-2.png:  
        After a C-s.  Note the ^M indicator and empty entry pushed to
        the last.

icomplete-3.png:
        This is the second invocation of M-x debbugs-gnu RET.  Note that
        the entry from the recent history is at the head.  Also note ^M
        and the empty entries.


[-- Attachment #2: icomplete-bug12638.diff --]
[-- Type: text/x-diff, Size: 5337 bytes --]

=== modified file 'lisp/ChangeLog'
--- lisp/ChangeLog	2012-12-13 05:29:15 +0000
+++ lisp/ChangeLog	2012-12-13 13:32:58 +0000
@@ -1,3 +1,10 @@
+2012-12-13  Jambunathan K  <kjambunathan@gmail.com>
+
+	* icomplete.el (icomplete-hide-common-prefix): New user variable.
+	(icomplete-first-match): New user face.
+	(icomplete-completions): Honor above settings.  Correctly handle
+	the case when NAME is complete but not unique (Bug#12638).
+
 2012-12-13  Alan Mackenzie  <acm@muc.de>
 
 	Make CC Mode not hang when _some_ lines end in CRLF.  Bug #11841.

=== modified file 'lisp/icomplete.el'
--- lisp/icomplete.el	2012-11-30 17:09:55 +0000
+++ lisp/icomplete.el	2012-12-13 13:16:31 +0000
@@ -76,6 +76,18 @@
   :type 'string
   :version "24.4")
 
+(defcustom icomplete-hide-common-prefix t
+  "When non-nil, hide common prefix from completion candidates.
+When nil, show candidates in full."
+  :type 'boolean
+  :version "24.4"
+  :group 'icomplete)
+
+(defface icomplete-first-match  '((t :weight bold))
+  "Face used by icomplete for highlighting first match."
+  :version "24.4"
+  :group 'icomplete)
+
 ;;;_* User Customization variables
 (defcustom icomplete-prospects-height
   ;; 20 is an estimated common size for the prompt + minibuffer content, to
@@ -332,14 +344,12 @@ are exhibited within the square braces.)
 	     (determ (unless (or (eq t compare) (eq t most-try)
 				 (= (setq compare (1- (abs compare)))
 				    (length most)))
-		       (concat open-bracket
 			       (cond
 				((= compare (length name))
                                  ;; Typical case: name is a prefix.
 				 (substring most compare))
 				((< compare 5) most)
-				(t (concat "..." (substring most compare))))
-			       close-bracket)))
+			(t (concat "..." (substring most compare))))))
 	     ;;"-prospects" - more than one candidate
 	     (prospects-len (+ (length determ) 6 ;; take {,...} into account
                                (string-width (buffer-string))))
@@ -350,6 +360,8 @@ are exhibited within the square braces.)
                     ;; one line, increase the allowable space accordingly.
                     (/ prospects-len (window-width)))
                  (window-width)))
+	     (prefix (when icomplete-hide-common-prefix
+		       (try-completion "" comps)))
              (prefix-len
               ;; Find the common prefix among `comps'.
 	      ;; We can't use the optimization below because its assumptions
@@ -359,37 +371,58 @@ are exhibited within the square braces.)
 	      ;;     ;; Common case.
 	      ;;     (length most)
 	      ;; Else, use try-completion.
-	      (let ((comps-prefix (try-completion "" comps)))
-		(and (stringp comps-prefix)
-		     (length comps-prefix)))) ;;)
-
+	      (and (stringp prefix) (length prefix))) ;;)
 	     prospects most-is-exact comp limit)
 	(if (eq most-try t) ;; (or (null (cdr comps))
 	    (setq prospects nil)
+	  (when (and (cdr comps) (member name comps))
+	    ;; NAME is complete but not unique.  This scenario, poses
+	    ;; following UI issues:
+	    ;;
+	    ;; - When common prefix is hidden, NAME is stripped empty.
+	    ;;   This would make the entry inconspicuous.
+	    ;; - Due to sorting of completions, NAME may not be the
+	    ;;   first of the prospects and could be hidden deep in the
+	    ;;   displayed string.
+	    ;; - Because of truncation, NAME may not even be displayed
+	    ;;   to the user.
+	    ;;
+	    ;; To circumvent all the above problems, provide a visual
+	    ;; cue to the user via try completion field.
+	    (setq determ "\r"))
 	  (while (and comps (not limit))
 	    (setq comp
 		  (if prefix-len (substring (car comps) prefix-len) (car comps))
 		  comps (cdr comps))
-	    (cond ((string-equal comp "") (setq most-is-exact t))
-		  ((member comp prospects))
-		  (t (setq prospects-len
+	    (setq prospects-len
                            (+ (string-width comp) 1 prospects-len))
 		     (if (< prospects-len prospects-max)
 			 (push comp prospects)
-		       (setq limit t))))))
+	      (setq limit t))))
+	(setq prospects (nreverse prospects))
+	;; Decorate the first of prospects.
+	(when prospects
+	  (let ((first (copy-sequence (pop prospects))))
+	    (put-text-property 0 (length first)
+			       'face 'icomplete-first-match first)
+	    (push first prospects)))
         ;; Restore the base-size info, since completion-all-sorted-completions
         ;; is cached.
         (if last (setcdr last base-size))
+	(when determ
+	  (setq determ (concat open-bracket determ close-bracket)))
 	(if prospects
 	    (concat determ
+		    (if (not (cdr prospects))
+			;; NAME is not complete but will complete to
+			;; an EXACT match on next try.
+			""
+		      (concat
 		    "{"
-		    (and most-is-exact
-                         (substring icomplete-separator
-                                    (string-match "[^ ]" icomplete-separator)))
-		    (mapconcat 'identity (nreverse prospects)
-                               icomplete-separator)
+		       (mapconcat 'identity prospects icomplete-separator)
 		    (and limit (concat icomplete-separator "…"))
-		    "}")
+		       "}"
+		       )))
 	  (concat determ " [Matched]"))))))
 
 ;;_* Local emacs vars.


[-- Attachment #3: icomplete-3.png --]
[-- Type: image/png, Size: 5177 bytes --]

[-- Attachment #4: icomplete-2.png --]
[-- Type: image/png, Size: 2134 bytes --]

[-- Attachment #5: icomplete-1.png --]
[-- Type: image/png, Size: 2081 bytes --]

  reply	other threads:[~2012-12-13 13:51 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-13 17:33 bug#12638: 24.2.50; FR: Some suggestions for icomplete-mode Jambunathan K
2012-10-13 22:14 ` Stefan Monnier
2012-10-14  4:18 ` Drew Adams
2012-10-23 19:43 ` Stefan Monnier
2012-10-23 20:08   ` Jambunathan K
2012-10-23 20:17     ` Jambunathan K
2012-10-24 13:09     ` Stefan Monnier
2012-11-02 11:49       ` Jambunathan K
2012-11-02 12:12         ` Jambunathan K
2012-11-09  1:53           ` Stefan Monnier
2012-11-09  2:17         ` Stefan Monnier
2012-11-09  4:25           ` Jambunathan K
2012-11-09 14:12             ` Stefan Monnier
2012-11-29 21:34 ` Stefan Monnier
2012-11-30  6:18   ` Jambunathan K
2012-11-30 19:37     ` Stefan Monnier
2012-12-04 12:54       ` Jambunathan K
2012-12-04 15:02         ` Stefan Monnier
2012-12-04 15:30           ` Jambunathan K
2012-12-04 15:45             ` Stefan Monnier
2012-12-04 16:12               ` Jambunathan K
2012-12-04 17:14                 ` Stefan Monnier
2012-12-04 17:32                   ` Jambunathan K
2012-12-12  3:18                     ` Stefan Monnier
2012-12-12  3:42                       ` Drew Adams
2012-12-12  6:34                         ` Kevin Rodgers
2012-12-12 16:15                           ` Drew Adams
2012-12-04 15:51           ` Jambunathan K
2012-12-13 13:51             ` Jambunathan K [this message]
2012-12-17 16:28               ` Stefan Monnier
2012-12-17 19:22                 ` Jambunathan K
2012-12-17 20:12                   ` Stefan Monnier
2012-12-17 20:58                     ` Jambunathan K
2012-12-18  1:26                       ` Stefan Monnier
2012-12-18  3:09                         ` Drew Adams
2012-12-18 14:40                         ` Jambunathan K
2013-01-11  5:47               ` Jambunathan K
2013-01-11 14:17                 ` Stefan Monnier
2013-02-13 13:56                   ` Jambunathan K
2013-02-13 15:15                     ` Stefan Monnier
2013-02-13 17:18                       ` Jambunathan K
2013-11-15  4:42 ` Jambunathan K

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=87bodykqlx.fsf@gmail.com \
    --to=kjambunathan@gmail.com \
    --cc=12638@debbugs.gnu.org \
    --cc=monnier@iro.umontreal.ca \
    /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).