🚀 go-pugleaf

In article <BzD9vv.440@world.std.com>, dwh@world.std.com (Denis W
Haskin) writes:
>Actually, while we're at it, I find that my 'forward session'
>character
>for the terminal server also comes in and out of commission when I use
>Emacs; that is, Emacs seems to disable it.  Any ideas here?

If your terminal server works the same as a DECserver 300, then I would guess
that Emacs is turning pasthru or passall mode on and off. I haven't used Emacs
on VMS in years, but the version I used put the terminal in passall mode for 
the duration of the Emacs session.

---
Randy_McGee@EMail.Mot.Com
Motorola CIO Distributed Systems Support Group

In article <BzD9vv.440@world.std.com> dwh@world.std.com (Denis W Haskin) writes:
>I've built and installed Gnu Emacs 18.58 on a VMS 5.5 system and it is
>mostly working alright, with one minor but fairly annoying problem.
>
>It appears that Emacs isn't keeping up with keystrokes, and if I use the
>arrow keys (i.e. multi-byte keystrokes) at any decent speed, it
>misinterprets them and I get garbage, including deleted text, inserted
>text, random capitalizations, etc.


We've been having this problem since 18.54 and VMS 5.  Emacs simply
doesn't seem to be able to keep with fast character input.  I've made
inquiries about this in the past, and received a few suggestions for
patches (though don't remember what they were at the moment), but
nothing seems to help.

The problem manifests itself mostly when Emacs is used via vt tubes
(vt220, 240, 320, etc.) which are connected thru DEC terminal servers.
But, I've seen Emacs dropping keys even when running as an X11 client.
Whenever it drops a key when run under X11 it will issue a beep (so
I notice right away and thus it's not so bad), which makes me wonder
how it "knows" it lost a key, and if it does, why can't it just keep
up.

I always hoped that later versions will correct that.  I see that as
of 18.58 it's still a problem. :-(

Sorry I couldn't be more of a help.

>Actually, while we're at it, I find that my 'forward session' character
>for the terminal server also comes in and out of commission when I use
>Emacs; that is, Emacs seems to disable it.  Any ideas here?

It's probably because Emacs switches the terminal session to passthru
mode, so all characters are passed to it instead of being grabbed by
the terminal server.  To forward session while in Emacs, try hitting
break (which puts you back into local on DEC's servers) and then the
forward-session char.

		- Steve -
-- 
Internet: stevef@bony1.bony.com   | / If you could turn on the lights \
bang : {uupsi,uunet}!bony1!stevef ||| quickly enough, you could see   ||
                                  ||| what the dark looks like...     ||
  Opinions? What opinions?        | \_________________________________/

>>>>> On Fri, 18 Dec 92 14:17:05 GMT, stevef@bony1.bony.com (Steve Faiwiszewski) said:

SF> In article <BzD9vv.440@world.std.com> dwh@world.std.com (Denis W Haskin) writes:
>I've built and installed Gnu Emacs 18.58 on a VMS 5.5 system and it is
>mostly working alright, with one minor but fairly annoying problem.
>
>It appears that Emacs isn't keeping up with keystrokes, and if I use the
>arrow keys (i.e. multi-byte keystrokes) at any decent speed, it
>misinterprets them and I get garbage, including deleted text, inserted
>text, random capitalizations, etc.


SF> The problem manifests itself mostly when Emacs is used via vt tubes
SF> (vt220, 240, 320, etc.) which are connected thru DEC terminal servers.
SF> But, I've seen Emacs dropping keys even when running as an X11 client.
SF> Whenever it drops a key when run under X11 it will issue a beep (so
SF> I notice right away and thus it's not so bad), which makes me wonder
SF> how it "knows" it lost a key, and if it does, why can't it just keep
SF> up.

Emacs does NOT know... And does NOT beep. This is done by the whatever server.

I got another suggestion. Keys are fetched by Emacs using ASTs. If you type
fast, you might get more active ASTs than you process' ASTLIM. The answer
might be to increase ASTLIM, either on per-user basis, or by increasing
PQL_MASTLM in SYSGEN.

SF> I always hoped that later versions will correct that.  I see that as
SF> of 18.58 it's still a problem. :-(

I have the same problem. I haven't tried the ASTLIM thingy... it's just
a theory so far.

>Actually, while we're at it, I find that my 'forward session' character
>for the terminal server also comes in and out of commission when I use
>Emacs; that is, Emacs seems to disable it.  Any ideas here?

True, Emacs switches the terminal to PASTHRU.

--
!+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++!
! Richard Levitte, System manager         ! tel: int+46-8-790 64 23           !
! Royal Institute of Technology           ! fax: int+46-8-791 76 54           !
! Department of Telecommunication Theory  ! Internet: richard@ttt.kth.se      !
! S-100 44 Stockholm, Sweden              !                                   !
!-----------------------------------------------------------------------------!