* bug in file-name-shadow-mode
@ 2005-03-19 20:12 Luc Teirlinck
2005-03-20 12:59 ` Richard Stallman
0 siblings, 1 reply; 7+ messages in thread
From: Luc Teirlinck @ 2005-03-19 20:12 UTC (permalink / raw)
The following would appear to be a bug `file-name-shadow-mode'.
Do emacs -Q.
M-x customize-option RET file-name-shadow-mode RET
C-x C-f
Then set file-name-shadow-mode to t for the current session using the mouse.
You are still in the minibuffer. Clearly, file-name-shadow-mode does
not work in that minibuffer. This might be somewhat expected. Worse
is that file-name-shadow-mode stays disabled in subsequent invocations
of the minibuffer, even though the variable file-name-shadow-mode is t.
Sincerely,
Luc.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-19 20:12 bug in file-name-shadow-mode Luc Teirlinck
@ 2005-03-20 12:59 ` Richard Stallman
2005-03-20 16:26 ` Luc Teirlinck
0 siblings, 1 reply; 7+ messages in thread
From: Richard Stallman @ 2005-03-20 12:59 UTC (permalink / raw)
Cc: emacs-devel
Worse
is that file-name-shadow-mode stays disabled in subsequent invocations
of the minibuffer, even though the variable file-name-shadow-mode is t.
I am not sure what you mean. Would you please provide precise instructions
for that part too?
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-20 12:59 ` Richard Stallman
@ 2005-03-20 16:26 ` Luc Teirlinck
2005-03-21 1:19 ` Richard Stallman
0 siblings, 1 reply; 7+ messages in thread
From: Luc Teirlinck @ 2005-03-20 16:26 UTC (permalink / raw)
Cc: emacs-devel
Richard Stallman wrote:
Worse
is that file-name-shadow-mode stays disabled in subsequent invocations
of the minibuffer, even though the variable file-name-shadow-mode is t.
I am not sure what you mean. Would you please provide precise instructions
for that part too?
Do C-g to quit the current minibuffer. Then do C-x C-f.
One now has ~/ in the minibuffer. If you type an extra "/", the "~/"
preceding it does not get any special highlighting, as it should if
file-name-shadow-mode is t.
Sincerely,
Luc.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-20 16:26 ` Luc Teirlinck
@ 2005-03-21 1:19 ` Richard Stallman
2005-03-21 16:20 ` Stefan Monnier
0 siblings, 1 reply; 7+ messages in thread
From: Richard Stallman @ 2005-03-21 1:19 UTC (permalink / raw)
Cc: emacs-devel
Do C-g to quit the current minibuffer. Then do C-x C-f.
One now has ~/ in the minibuffer. If you type an extra "/", the "~/"
preceding it does not get any special highlighting, as it should if
file-name-shadow-mode is t.
The cause of this is that the minibuffer binds minibuffer-setup-hook
and therefore the local binding is what gets set. Thus, exiting the
minibuffer wipes it out.
I think there is nothing to be done about this. It would be hard to
fix and it probably isn't worth the trouble.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-21 1:19 ` Richard Stallman
@ 2005-03-21 16:20 ` Stefan Monnier
2005-03-22 0:04 ` Luc Teirlinck
2005-03-22 3:34 ` Richard Stallman
0 siblings, 2 replies; 7+ messages in thread
From: Stefan Monnier @ 2005-03-21 16:20 UTC (permalink / raw)
Cc: Luc Teirlinck, emacs-devel
> Do C-g to quit the current minibuffer. Then do C-x C-f.
> One now has ~/ in the minibuffer. If you type an extra "/", the "~/"
> preceding it does not get any special highlighting, as it should if
> file-name-shadow-mode is t.
> The cause of this is that the minibuffer binds minibuffer-setup-hook
> and therefore the local binding is what gets set. Thus, exiting the
> minibuffer wipes it out.
> I think there is nothing to be done about this. It would be hard to
> fix and it probably isn't worth the trouble.
How 'bout the untested patch below?
Stefan
--- files.el 15 fév 2005 11:00:48 -0500 1.745
+++ files.el 21 mar 2005 11:19:48 -0500
@@ -928,20 +928,30 @@
(defvar find-file-default nil
"Used within `find-file-read-args'.")
+(defmacro minibuffer-with-setup-hook (fun &rest body)
+ "Add FUN to `minibuffer-setup-hook' while executing BODY.
+BODY should use the minibuffer at most once.
+Recursive uses of the minibuffer will not be affected."
+ (declare (indent 1) (debug t))
+ (let ((funname (make-symbol "setup-hook"))
+ (oldval (make-symbol "old-val")))
+ `(unwind-protect
+ (progn
+ (fset ',funname (lambda () (,fun)
+ ;; Clear out this hook so it does not interfere
+ ;; with any recursive minibuffer usage.
+ (remove-hook 'minibuffer-setup-hook ',funname)))
+ (add-hook 'minibuffer-setup-hook ',funname)
+ ,@body)
+ (remove-hook 'minibuffer-setup-hook ',funname))))
+
(defun find-file-read-args (prompt mustmatch)
(list (let ((find-file-default
(and buffer-file-name
- (abbreviate-file-name buffer-file-name)))
- (munge-default-fun
- (lambda ()
- (setq minibuffer-default find-file-default)
- ;; Clear out this hook so it does not interfere
- ;; with any recursive minibuffer usage.
- (pop minibuffer-setup-hook)))
- (minibuffer-setup-hook
- minibuffer-setup-hook))
- (add-hook 'minibuffer-setup-hook munge-default-fun)
- (read-file-name prompt nil default-directory mustmatch))
+ (abbreviate-file-name buffer-file-name))))
+ (minibuffer-with-setup-hook
+ (lambda () (setq minibuffer-default find-file-default))
+ (read-file-name prompt nil default-directory mustmatch)))
t))
(defun find-file (filename &optional wildcards)
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-21 16:20 ` Stefan Monnier
@ 2005-03-22 0:04 ` Luc Teirlinck
2005-03-22 3:34 ` Richard Stallman
1 sibling, 0 replies; 7+ messages in thread
From: Luc Teirlinck @ 2005-03-22 0:04 UTC (permalink / raw)
Cc: rms, emacs-devel
Stefan Monnier wrote:
> I think there is nothing to be done about this. It would be hard to
> fix and it probably isn't worth the trouble.
How 'bout the untested patch below?
Seems to work.
Sincerely,
Luc.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: bug in file-name-shadow-mode
2005-03-21 16:20 ` Stefan Monnier
2005-03-22 0:04 ` Luc Teirlinck
@ 2005-03-22 3:34 ` Richard Stallman
1 sibling, 0 replies; 7+ messages in thread
From: Richard Stallman @ 2005-03-22 3:34 UTC (permalink / raw)
Cc: teirllm, emacs-devel
The use of fset makes this rather ugly.
If the idea is to put a function onto minibuffer-setup-hook just to
execute once, I see no reason to bind anything. Just add the function
and design it to remove itself when run. For safety's sake,
use an unwind-protect to remove the function from the list
in the case where an error prevented it from being run.
So how about this?
(defun find-file-read-args (prompt mustmatch)
(list (let ((find-file-default
(and buffer-file-name
(abbreviate-file-name buffer-file-name)))
(find-file-read-args-hook-fn
(lambda ()
;; Clear out this hook so it does not interfere
;; with any recursive minibuffer usage.
(remove-hook 'minibuffer-setup-hook
find-file-read-args-hook-fn)
(setq minibuffer-default find-file-default))))
(unwind-protect
(progn
(add-hook 'minibuffer-setup-hook
find-file-read-args-hook-fn)
(read-file-name prompt nil default-directory mustmatch))
(remove-hook 'minibuffer-setup-hook
find-file-read-args-hook-fn)))
t))
Meanwhile, do you understand what the purpose of this function is?
Is this what arranges to make M-n bring in the default file name?
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2005-03-22 3:34 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-03-19 20:12 bug in file-name-shadow-mode Luc Teirlinck
2005-03-20 12:59 ` Richard Stallman
2005-03-20 16:26 ` Luc Teirlinck
2005-03-21 1:19 ` Richard Stallman
2005-03-21 16:20 ` Stefan Monnier
2005-03-22 0:04 ` Luc Teirlinck
2005-03-22 3:34 ` Richard Stallman
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).