GNU bug report logs - #16661
24.3.50; standalone minibuffer frame gets renamed with name of aother frame

Previous Next

Full log

View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: Drew Adams <drew.adams <at> oracle.com>
Cc: 16661 <at> debbugs.gnu.org
Subject: bug#16661: 24.3.50; standalone minibuffer frame gets renamed with name of aother frame
Date: Sat, 22 Feb 2014 10:55:56 +0200

> Date: Fri, 21 Feb 2014 20:23:47 -0800 (PST)
> From: Drew Adams <drew.adams <at> oracle.com>
> Cc: 16661 <at> debbugs.gnu.org
> 
> > How about adding a Lisp backtrace to the info you already put there?
> 
> OK, I have some more info about this.  This is the backtrace:
> 
> 000 this-cmd: nil, last-cmd: nil
> 	(name . "drews-lisp-20")
>   backtrace()
>   ad-Advice-modify-frame-parameters(#<subr modify-frame-parameters> #<frame Emacs minibuffer                         show/hide: hold CTRL + click in window 03bec000> ((name . "drews-lisp-20")))
>   apply(ad-Advice-modify-frame-parameters #<subr modify-frame-parameters> (#<frame Emacs minibuffer                         show/hide: hold CTRL + click in window 03bec000> ((name . "drews-lisp-20"))))
>   modify-frame-parameters(#<frame Emacs minibuffer                         show/hide: hold CTRL + click in window 03bec000> ((name . "drews-lisp-20")))
>   rename-frame()
>   run-hooks(window-setup-hook)
> 
> I do this in my init file:
> 
> (when (and (if (fboundp 'display-graphic-p) (display-graphic-p) window-system)
>            (eq system-type 'windows-nt) (fboundp 'rename-frame))
>   (add-hook 'window-setup-hook 'rename-frame))

I'm confused: window-setup-hook is called only once, as part of Emacs
startup.  Are you saying that all these cases of renaming happened
during Emacs startup, and only at that time?  I somehow decided,
probably erroneously, that the renaming randomly happened during a
running session.

> Why should the `buffer-name' of the minibuffer frame be
> different from the frame name, for the minibuffer frame?

AFAIK, buffer-name in the minibuffer always returns the name of the
buffer which caused minibuffer to be entered.  Anything else would be
highly confusing, e.g. "M-: (buffer-name) RET" surely must return the
name of the invoking buffer.

IOW, I think that buffer-name in the minibuffer is unreliable, if you
want to get the name of the minibuffer itself.

Previous Next

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