[OS X Emacs] Aquamacs Cursor movement and word wrapping: C-e, C-a, C-n, C-p

Tom Van Vleck thvv at multicians.org
Thu Apr 22 22:29:05 EDT 2010

On Apr 22, 2010, at 9:56 PM, David Reitter wrote:

> On Apr 22, 2010, at 9:39 PM, Tom Van Vleck wrote:
>> Changing the meaning of C-e, C-a, C-n, C-p would make Aquamacs  
>> unusable for me.  It would break saved macros.  It would mean that  
>> these keystrokes meant different things depending on what machine I  
>> was logged into.  I would have to relearn editing skills I learned  
>> over 30 years ago.  I couldn't use Aquamacs if you changed this  
>> fundamental meaning.
> Could you clarify please: do not change them from their Emacs 23  
> settings?
> In Emacs 23, all of these apply to the visual line (visual-line-mode  
> = soft word wrapping on).  That's what the commands do in Emacs 23.
> This is different from Emacs 22, where C-a moved to the beginning of  
> the (long) buffer line, no matter what the wrapping was.  Native  
> word wrapping didn't exist (longlines-mode excluded).  When I  
> introduced word wrapping in Emacs 22 in one of the Aquamacs 1.x  
> versions, I took care to give C-a, C-e, C-n, C-p their meaning based  
> on visual lines, based on user feedback.

I think C-a should move to the beginning of a buffer line, and  
similarly for the others.
Making it move to the beginning of a visual line means that where the  
cursor ends up
would depend on the width of the window and how the line was wrapped.

When I record a keystroke macro, I often begin it with C-a to make  
sure I am at the beginning of the buffer line before e.g. searching  
for something.  If C-a sometimes went to somewhere else, then the  
macro would not work correctly.  Sometimes keystroke macros get named  
and saved.  I have a few such in my customization, though I would have  
to read them intensely to see if they will break if the meaning of  
commands is altered.

If Emacs 23 has changed the meaning of C-a C-e C-n C-p, or made the  
meaning depend on some mode, then I think that is a mistake.

> So do I understand you correctly that you want the bindings to be  
> same in 2.x and in 1.x, i.e. as in Emacs 23?
> I'm confused as you mention macros, which seem to require buffer- 
> based semantics.

I thought buffer-based semantics was how Emacs worked.  It is what my  
fingers expect.  I would like this behavior to continue.

More information about the MacOSX-Emacs mailing list