Re: Behaviour for save-excursion

[martin rudalics <rudalics@gmx.at>]

 > What is the proper behaviour for save-excursion?  The emacs-lisp intro and
 > the docstring seem to have slightly differing explanations - namely, the
 > docstring mentions that it saves current-buffer as well as point and mark,
 > and the emacs-lisp intro does not.  Based on just what it does, I'm not
 > entirely sure which is correct - the following code snippet, executed in
 > *scratch*, returns the "*scratch*" buffer but the buffer displayed is the
 > "bar" buffer, and evaluating current-bufer afterwards returns the "bar"
 > buffer.
 >
 > (progn
 >   (save-excursion
 >     (switch-to-buffer "bar"))
 >   (current-buffer))
 >
 > Either the manual or the docstring should be changed to reflect what it
 > does, but I'm not entirely sure which based on this.

Please read the Elisp manual entry on `save-excursion' which states:

      Likewise, `save-excursion' does not restore window-buffer
      correspondences altered by functions such as `switch-to-buffer'.
      One way to restore these correspondences, and the selected window,
      is to use `save-window-excursion' inside `save-excursion' (*note
      Window Configurations::).

Also the following entry from section 27.2 The Current Buffer of the
Elisp manual might provide further explanation:

      When an editing command returns to the editor command loop, the
   command loop designates the buffer displayed in the selected window as
   current, to prevent confusion: the buffer that the cursor is in when
   Emacs reads a command is the buffer that the command will apply to.
   (*Note Command Loop::.)  Therefore, `set-buffer' is not the way to
   switch visibly to a different buffer so that the user can edit it.  For
   that, you must use the functions described in *Note Displaying
   Buffers::.

martin