all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Stephen Berman <Stephen.Berman@rub.de>
To: emacs-pretest-bug@gnu.org
Subject: bug#1488: 23.0.60; dired-pop-to-buffer: use fit-window-to-buffer
Date: Thu, 04 Dec 2008 10:44:12 +0100	[thread overview]
Message-ID: <87fxl446mr.fsf@escher.local.home> (raw)

[-- Attachment #1: Type: text/plain, Size: 1896 bytes --]

In GNU Emacs 23.0.60.21 (i686-pc-linux-gnu, GTK+ Version 2.12.9)
 of 2008-12-03 on escher

1. emacs -Q

2. Eval this:
   (define-minor-mode my-mm ()
     "My minor mode."
     :global t
     (if my-mm
         (setq default-header-line-format
   	       '(:eval (propertize "test" 'face '(:overline t))))
       (setq default-header-line-format nil)))

3. Enable my-mm.

4. C-x d foo (where `foo' is some directory with at least three files,
   e.g. emacs/lisp).

5. Mark three files in foo.

6. Type `C'
   ==> The popped up buffer *Marked Files* shows only two of the three
   marked files.

Likewise if the value of the 'face property in my-mm has a non-nil :box
attribute.  In contrast, if the value of the 'face property is
'(:underline t) (but not '(:underline t :overline t)), then in step 6
all three marked files are shown, though with no room to spare.

I guess the basic problem here is that Emacs calculates window-height in
integral equally sized line numbers.  But less fundamentally and more
directly, this problem arises, AFAICT, because dired-pop-to-buffer's
shrink-to-fit code fails to account for header lines.  To fix this it
would suffice to add this sexp above the last sexp in
dired-pop-to-buffer: 
    (with-current-buffer buf 
      (and header-line-format
           (progn (select-window w2) (enlarge-window 1))))

However, unless I'm missing something, fit-window-to-buffer does what
dired-pop-to-buffer's shrink-to-fit code does, and moreover also handles
header lines.  From perusing the change logs, I found several instances
where fit-window-to-buffer replaces library-specific code, and this
appears to be another candidate (dired-pop-to-buffer was introduced many
years before fit-window-to-buffer), in addition to fixing the above bug.
I'm not able to fix the display code to let window-height return
fractional line numbers, so I propose the following patch:


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: dired-pop-to-buffer patch --]
[-- Type: text/x-patch, Size: 2250 bytes --]

*** emacs/lisp/dired.el.~1.414.~	2008-12-03 10:17:51.000000000 +0100
--- emacs/lisp/dired.el			2008-12-04 10:11:05.000000000 +0100
***************
*** 2678,2721 ****
  
  (defun dired-pop-to-buffer (buf)
    ;; Pop up buffer BUF.
    ;; If dired-shrink-to-fit is t, make its window fit its contents.
!   (if (not dired-shrink-to-fit)
!       (pop-to-buffer (get-buffer-create buf))
!     ;; let window shrink to fit:
!     (let ((window (selected-window))
! 	  target-lines w2)
!       (cond ;; if split-height-threshold is enabled, use the largest window
!             ((and (> (window-height (setq w2 (get-largest-window)))
! 		     split-height-threshold)
! 		  (window-full-width-p w2))
! 	     (setq window w2))
! 	    ;; if the least-recently-used window is big enough, use it
! 	    ((and (> (window-height (setq w2 (get-lru-window)))
! 		     (* 2 window-min-height))
! 		  (window-full-width-p w2))
! 	     (setq window w2)))
!       (save-excursion
! 	(set-buffer buf)
! 	(goto-char (point-max))
! 	(skip-chars-backward "\n\r\t ")
! 	(setq target-lines (count-lines (point-min) (point)))
! 	;; Don't forget to count the last line.
! 	(if (not (bolp))
! 	    (setq target-lines (1+ target-lines))))
!       (if (<= (window-height window) (* 2 window-min-height))
! 	  ;; At this point, every window on the frame is too small to split.
! 	  (setq w2 (display-buffer buf))
! 	(setq w2 (split-window window
! 		  (max window-min-height
! 		       (- (window-height window)
! 			  (1+ (max window-min-height target-lines)))))))
!       (set-window-buffer w2 buf)
!       (if (< (1- (window-height w2)) target-lines)
! 	  (progn
! 	    (select-window w2)
! 	    (enlarge-window (- target-lines (1- (window-height w2))))))
!       (set-window-start w2 1)
!       )))
  
  (defcustom dired-no-confirm nil
    "A list of symbols for commands Dired should not confirm.
--- 2678,2687 ----
  
  (defun dired-pop-to-buffer (buf)
    ;; Pop up buffer BUF.
+   (pop-to-buffer (get-buffer-create buf))
    ;; If dired-shrink-to-fit is t, make its window fit its contents.
!   (when dired-shrink-to-fit
!     (fit-window-to-buffer (get-buffer-window buf))))
  
  (defcustom dired-no-confirm nil
    "A list of symbols for commands Dired should not confirm.

[-- Attachment #3: Type: text/plain, Size: 452 bytes --]


(Note that if the frame is shrunk so that not all marked file names fit
in the *Marked Files* window, then without an oversized header line as
is my-mm, dired-pop-to-buffer fails with the error "Attempt to delete
minibuffer or sole ordinary window"; whereas with an oversized header
line, this error is raised only if not even one file name can be shown.
But I see this difference in behavior whether or not the above patch is
applied.)

Steve Berman

             reply	other threads:[~2008-12-04  9:44 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <4940E3E8.8050501@gmx.at>
2008-12-04  9:44 ` Stephen Berman [this message]
2008-12-04 17:23   ` bug#1488: 23.0.60; dired-pop-to-buffer: use fit-window-to-buffer Drew Adams
2008-12-04 17:58   ` martin rudalics
2008-12-04 18:33     ` martin rudalics
2008-12-05 14:25   ` martin rudalics
2008-12-05 17:27     ` Stephen Berman
2008-12-05 17:37       ` martin rudalics
2008-12-05 18:17         ` Stephen Berman
2008-12-06 19:25           ` martin rudalics
2008-12-09 20:15             ` Stephen Berman
2008-12-11 17:29               ` martin rudalics
2008-12-11 10:05   ` bug#1488: marked as done (23.0.60; dired-pop-to-buffer: use fit-window-to-buffer) Emacs bug Tracking System

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

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

  git send-email \
    --in-reply-to=87fxl446mr.fsf@escher.local.home \
    --to=stephen.berman@rub.de \
    --cc=1488@emacsbugs.donarmstrong.com \
    --cc=emacs-pretest-bug@gnu.org \
    /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 external index

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

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.