Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!sdd.hp.com!decwrl!ucbvax!hplabs!hpl-opus!steinbac
From: steinbac@hpl-opus.HP.COM (Gunter Steinbach)
Newsgroups: comp.editors
Subject: Re: word completion macro for vi
Message-ID: <62420012@hpl-opus.HP.COM>
Date: 29 Jun 90 23:14:05 GMT
References: <717@utsun.s.u-tokyo.ac.jp>
Organization: HP Labs, High Speed Electronics Dept., Palo Alto, CA
Lines: 27

> / hpl-opus:comp.editors / bowles@is.s.u-tokyo.ac.jp (Eric E. Bowles) /
> 10:24 pm  Jun 19, 1990 /

> " Word completion (Komplete, Next, Previous).
> "
> :map! ^K ^[a.^[bmmi?\<^[b"zdt.@z^Mywmx`mPbea ^[dws
> :map! ^N ^[a.^[bdw`xnywmx`mPbea ^[dws
> :map! ^P ^[a.^[bdw`xNywmx`mPbea ^[dws

Great stuff.  I love it, including the additions by L. Quin and S.
Mullender.

However, on my system, an HP9000/350 running HP-UX6.5, the ^N (next)
function doesn't work.  If I type the mapped commands in one by one by
hand, it works fine, but when map!ped to ^N, the "n" command seems to
look not for the last search string but for what follows the "n".  E.g.
if I map! the sequence only up to the "y" following the "n", then the
cursor jumps to the previous "y".  And with the whole sequence map!ped,
it just can't find anything (except in this message, as I just tried
out, because it contains the same sequence, "ywmx`m....").

Has anybody else experienced this?  It could still be some interaction
with another of my many mappings, but I can't figure what.

Any pointers are greatly appreciated.

	 Guenter Steinbach		gunter_steinbach@hplabs.hp.com