Message-ID: <296@aluxe.UUCP>
Date: Sun, 12-Aug-84 19:42:25 EDT
Article-I.D.: aluxe.296
Posted: Sun Aug 12 19:42:25 1984
Date-Received: Mon, 13-Aug-84 01:05:20 EDT
Organization: AT&T Bell Laboratories, Allentown, PA
Lines: 38

 > From: info-mac@uw-beaver (info-mac)
 > Newsgroups: fa.info-mac
 > 
 > From: winkler@harvard.ARPA (Dan Winkler)
 > The beta version of MacTerminal does allow you to move the cursor
 > using the mouse.  You point the mouse, hold down the option key,
 > and click and MacTerminal sends the appropriate cursor addressing
 > commands to position the cursor.
 > 

How does MacTerminal tell the host where the cursor is at?  

If when you position the cursor with the mouse a bunch of arrow key sequences 
are transmitted relative to where you were, it would take quite a while to
tramsmit, (2.6 sec worst case at 1200bd).  This would have the
advantage that TERMCAP understands what the arrow keys are, and would
be usable on most programs.

Another method would be to let the user move the cursor locally, and REQUIRE
the host to ask where the cursor is at when it matters, ( \E[6n for a vt100).
This method is much faster and cleaner, but rewrites to programs would be
necessary.

The final method would to have MacTerminal send an escape sequence,
(\E[Pl;PcM or something ) when he/she releases the mouse button at
a new cursor location.  This is probably the best, but again, TERMCAP
does not support cursor entry of this sort (as far as I know); and
rewrites would be necesary to editors.... .

I do not have access to the new Beta MacTerminal yet, but I would like
to know what thoughts there were on a standard way to "mouse" the cursor
on screen driven programs, and how MacTerminal did it.

Lonnie R. Abelbeck
AT&T Bell Laboratories
mhuxi!aluxe!lra