bug#13017: 24.2.90; minibuf-eldef-setup-minibuffer should append minibuf-eldef-update-minibuffer to post-command-hook

bug#13017: 24.2.90; minibuf-eldef-setup-minibuffer should append minibuf-eldef-update-minibuffer to post-command-hook

[Christopher Schmidt]

[-- Attachment #1: Type: text/plain, Size: 384 bytes --]

In minibuf-eldef-setup-minibuffer there is this form

    (add-hook 'post-command-hook #'minibuf-eldef-update-minibuffer nil t)

I think this is not correct.  The APPEND parameter of add-hook should be
t so all changes made to the minibuffer content by previous functions in
post-command-hook are taken into account by
minibuf-eldef-update-minibuffer.

Here is a patch for the trunk.

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: minibuf-eldef.diff --]
[-- Type: text/x-diff, Size: 1035 bytes --]

=== modified file 'lisp/ChangeLog'
--- lisp/ChangeLog	2012-11-27 16:40:04 +0000
+++ lisp/ChangeLog	2012-11-28 00:07:24 +0000
@@ -1,3 +1,8 @@
+2012-11-28  Christopher Schmidt  <christopher@ch.ristopher.com>
+
+	* minibuf-eldef.el (minibuf-eldef-setup-minibuffer): Append
+	minibuf-eldef-update-minibuffer to post-command-hook.  (Bug#)
+
 2012-11-27  Glenn Morris  <rgm@gnu.org>
 
 	* emacs-lisp/derived.el (derived-mode-make-docstring):

=== modified file 'lisp/minibuf-eldef.el'
--- lisp/minibuf-eldef.el	2012-11-07 20:43:38 +0000
+++ lisp/minibuf-eldef.el	2012-11-28 00:05:45 +0000
@@ -140,7 +140,7 @@
 	    (minibuffer-contents-no-properties))
       (setq minibuf-eldef-initial-buffer-length (point-max))
       (add-to-list 'minibuf-eldef-frobbed-minibufs (current-buffer))
-      (add-hook 'post-command-hook #'minibuf-eldef-update-minibuffer nil t))))
+      (add-hook 'post-command-hook #'minibuf-eldef-update-minibuffer t t))))
 
 ;; post-command-hook to swap prompts when necessary
 (defun minibuf-eldef-update-minibuffer ()


[-- Attachment #3: Type: text/plain, Size: 21 bytes --]


        Christopher

bug#13017: 24.2.90; minibuf-eldef-setup-minibuffer should append minibuf-eldef-update-minibuffer to post-command-hook

[Stefan Monnier]

> In minibuf-eldef-setup-minibuffer there is this form

>     (add-hook 'post-command-hook #'minibuf-eldef-update-minibuffer nil t)

> I think this is not correct.  The APPEND parameter of add-hook should be
> t so all changes made to the minibuffer content by previous functions in
> post-command-hook are taken into account by
> minibuf-eldef-update-minibuffer.

That sounds reasonable, but I'd be happier to know of a concrete example
where this was needed.


        Stefan

bug#13017: 24.2.90; minibuf-eldef-setup-minibuffer should append minibuf-eldef-update-minibuffer to post-command-hook

[Lars Ingebrigtsen]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> In minibuf-eldef-setup-minibuffer there is this form
>
>>     (add-hook 'post-command-hook #'minibuf-eldef-update-minibuffer nil t)
>
>> I think this is not correct.  The APPEND parameter of add-hook should be
>> t so all changes made to the minibuffer content by previous functions in
>> post-command-hook are taken into account by
>> minibuf-eldef-update-minibuffer.
>
> That sounds reasonable, but I'd be happier to know of a concrete example
> where this was needed.

Christopher, do you have an example that demonstrates why this is needed?

-- 
(domestic pets only, the antidote for overdose, milk.)
  bloggy blog http://lars.ingebrigtsen.no/

bug#13017: 24.2.90; minibuf-eldef-setup-minibuffer should append minibuf-eldef-update-minibuffer to post-command-hook

[Christopher Schmidt]

Tags: notabug

Lars Ingebrigtsen <larsi@gnus.org> writes:
> Christopher, do you have an example that demonstrates why this is
> needed?

IIRC, I played around with another prompt-simplifying form in
post-command-hook.  I do not have the source code any more.

It is easy to come up with an artificial sample which demonstrates this
issue.  Yet, even if minibuf-eldef-update-minibuffer is appended to
post-command-hook, does this change actually improve the situation?  I
do not think so.  It is not hard to come up with another sample which
requires the initial behaviour.

This issue is probably not a bug and does not matter to me any more.
Closing...