Path: utzoo!mnetor!uunet!husc6!bbn!uwmcsd1!ig!agate!pasteur!ames!amdahl!pyramid!uccba!hal!ncoast!allbery
From: allbery@ncoast.UUCP (Brandon Allbery)
Newsgroups: comp.emacs
Subject: Re: Need Info: Server Mode
Message-ID: <7488@ncoast.UUCP>
Date: 9 Mar 88 23:03:36 GMT
References: <3590013@hpindda.HP.COM> <262@gmdka.UUCP> <844@unmvax.unm.edu>
Reply-To: allbery@ncoast.UUCP (Brandon Allbery)
Followup-To: comp.emacs
Organization: Cleveland Public Access UN*X, Cleveland, Oh
Lines: 23

As quoted from <844@unmvax.unm.edu> by mike@turing.UNM.EDU (Michael I. Bushnell):
+---------------
| Not so.  In fact, I can prove it.  This article got posted, right???
| I had my emacs stopped (^Z) and was using rn.  Saw this, and decided
| to post.  So, Pnews called up 'ol emacsclient, which printed "Waiting
| for Emacs...".  Then I ^Z-ed out of Pnews, and fg-ed my emacs.
| Whereupon, server read in the header and I wrote this.  Shortly, I
| shall type C-x #, save the buffer, and ^Z emacs and fg Pnews.  And
| then, lo and behold, emacsclient will print "Done."...and if I list
| the article, it's still there.
+---------------

Why so much work?  Why doesn't emacsclient write its message and then SIGCONT
the stopped emacs, which then would discover a message on the socket and
read "keystrokes" from the socket rather than stdin and either display
directly (modulo tostop) or send the display to emacsclient (which could just
"cat" it)?  It'd certainly be easier for the user than having to ^Z/fg all
over the place, and it doesn't sound all that difficult to write.  (Okay,
maybe you can't get away with it -- but I'm fairly certain I could use a
"permanently" backgrounded emacs with message queues under System V to do it.)
-- 
	      Brandon S. Allbery, moderator of comp.sources.misc
       {well!hoptoad,uunet!hnsurg3,cbosgd,sun!mandrill}!ncoast!allbery