[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Dired, Buffer selection via C-M-L
- To: Philip E. Agre <AGRE at MIT-AI>
- Subject: Dired, Buffer selection via C-M-L
- From: Robert W. Kerns <RWK at SCRC-TENEX>
- Date: Sat ,8 May 82 11:11:00 EDT
- Cc: BUG-ZWEI at MIT-AI
- In-reply-to: The message of 7 May 82 09:57-EDT from Philip E. Agre <AGRE at MIT-AI>
Date: 7 May 1982 09:57-EDT
From: Philip E. Agre <AGRE at MIT-AI>
In M-X Dired, if you hit "H" on column 1 on line 1 then you get an error
Function NIL not defined (though C-U H doesn't blow up),
Fixed in 210.17.
and if you Abort out
of the error then the Dired gets buried under N other buffers and you can't
even get at it with C-n C-M-L for arbitrarily large n. Both behaviors seem
wrong.
Well, if you could get at it with C-n C-M-L it would turn up at annoying times.
Most of the time when you're done with a DIRED, you're done with it.
Which reminds me, why not generalize C-n C-M-L to mean "move to the n-th
buffer from the top on the list C-X C-B gives you" rather than just "move to
the n-th previously selected buffer" (which isn't really an accurate
description of its operation anyway, since it seems like it really refers to
just the same list C-X C-B gives you anway)?
Sounds pretty accurate to me, since C-X C-B lists the buffers in
most-recently-selected order. It is ambiguous, however. Got a better
way to phrase it? Anyway, I'm not sure what you're trying to say here.
It sounds to me like "Why not generalize it to do X rather than Y, which
isn't really accurate because it really does X anyway?". Am I mis-parsing
something?
One answer might be that you
want C-1 C-M-L to rotate only that part of the "buffer stack" whose buffers
have actually been selected at that point (as opposed to having been loaded
using explicit calls to zwei:load-file-into-zmacs, as with the zwei-load
hack in defsystem, for which see batali;sys >). The response to that is that
all you have to do is to be careful to define the notion of buffer stack
rotation so that only selected buffers rotate. Since I use buffer stack
rotation only rarely and C-n C-M-L for large n frequently, this doesn't seem
like a bad trade-off to me.
All you have to do is to not put buffers you don't want to participate in
buffer stack rotation on the buffer history list (or remove them). The name
of the variable is *ZMACS-BUFFER-HISTORY*. Of course, this means you can't
use C-M-L on them the first time.
I must confess I don't understand why you'd want to use C-large-n C-M-L.
C-X B or C-X C-B seem faster to me than counting my buffers.