[slime-devel] Re: [RfC] SPC vs. RET in Xref Buffers

[Andrew Gasparovic]

On Aug 8, 2008, at 6:57 PM, Tobias C. Rittweiler wrote:

> Andrew Gasparovic <andrew at andrewgasparovic.com> writes:
>
>> The way the XREF buffer works wrt. emacs windows has always seemed a
>> little strange to me. When you M-. and go straight to a function
>> definition (for example), the source file opens in the same window  
>> you
>> were in when you hit M-. The XREF buffer opens in another one (the
>> opposite one if you work with two windows side by side, as I do).
>>
>> That's not bad by itself, but pressing SPACE on one of the  
>> definitions
>> has the nasty side-effect of ruining my window layout. It replaces
>> both windows with one to display the source file. So if I just want  
>> to
>> go to a particular definition's source file, I either have to accept
>> that and reset my windows by hand, or press ENTER, kill the XREF
>> buffer, and switch to the other window with the source buffer.
>
> I always work with two frames side by side. That said, I believe SPACE
> should now work a bit better even with your setup.
>
>
>> Am I missing something? It seems like other people would be
>> experiencing this more often, but I've never heard anyone
>> complaining....
>
> In the long run, I'd like to have `C-u q' always restore the window
> configuration (not there yet), and `C-x k' never (where it's still  
> made
> sure that potential cleanup functions are run) across all relevant
> buffers.
>
> `q' restores the window configuration only if it hasn't changed
> meanwhile. This works quite reasonable when you use frames, not so  
> much
> when you use a horizontally split windows, I guess. Anyway, if the
> abovely pictured scheme is in place, you'll have full control over it.

That seems to be a pretty good compromise, actually, for people using  
side-by-side windows. Pressing RET in the XREF buffer (or SPACE in the  
future, I guess) to go to a definition, immediately followed by a 'q',  
closes the XREF buffer, leaves the windows alone, and leaves the point  
on the definition selected. I can live with that....

Thanks!
-Andrew



>
>
>  -T.
>
> _______________________________________________
> slime-devel site list
> slime-devel at common-lisp.net
> http://common-lisp.net/mailman/listinfo/slime-devel