Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!samsung!munnari.oz.au!basser!usage!elecvax!spinifex!cameron
From: cameron@spinifex.eecs.unsw.oz (Cameron Simpson)
Newsgroups: comp.editors
Subject: vi's "too dangerous to map that" (was: Re: vi (vs VMS/EDT))
Summary: i tried to set up EDT-like stuff in vi once ...
Message-ID: <511@spinifex.eecs.unsw.oz>
Date: 27 Nov 89 03:23:53 GMT
References: <1989Nov14.223604.17418@cs.eur.nl> <1989Nov21.101249.15226@cs.eur.nl>
Organization: EE & CS, Uni N.S.W., Sydney, Australia
Lines: 22

In article <1989Nov21.101249.15226@cs.eur.nl>, henk@cs.eur.nl (Henk Langeveld) writes:
> Andrej Brodnik (Andy) writes:
> >It is very interesting to read all of this staff about vi :: EDT. But let us
> >do something reasonable. Let us try to enforce vi to behave as EDT on UNIX and
> >EDT to be as vi on VMS. Can we?
> Nice idea, good luck!
> >PS: By the way let us assume to have some DEC compatible (if you would like
> >ANSI) terminal with a keypad as VT1xx has it.

I tried to do this for a friend of mine a couple of years ago. That is, rig
the keypad to act a bit like it does in EDT with `gold keys' (I think), and
effectively be in insert mode all the time, etc. Much mapping of the tail
ends of the escape sequences returned by the keys, etc. It partly worked,
but eventually I ran up against vi's "too dangerous to map that" message.
I can't find any doco about exactly what causes it, though it seems to be
maps which cross insert/command mode boundaries. I'd love to know exactly
what causes it. Then I'd have a better idea of how to work around it.
	- Cameron Simpson
-- 
ACSnet,CSNET:	cameron@eecs.unsw.oz	JANET: eecs.unsw.oz!cameron@ukc
UUCP:		...!uunet!munnari!eecs.unsw.oz!cameron
BITNET/ARPA:	cameron@eecs.unsw.oz.au


Brought to you by Super Global Mega Corp .com