[Home] [Groups] - Message: [Prev in Group] [Next in Group]
3652: Re: [MUD-Dev] string parsing
[Full Header] [Plain Text]
From: "Felix A. Croes" <felix@xs1.simplex.nl>
Newsgroups: nu.kanga.list.mud-dev
Date: Fri, 31 Oct 1997 13:02:44 +0100 (MET)
Organization: Kanga.Nu
Nathan Yospe <yospe@hawaii.edu> wrote:
> On Thu, 30 Oct 1997, Felix A. Croes wrote:
>[...]
> :The part that I didn't specify, of course, is which of these
> :production rules is eventually picked: a valid give command
> :matches all of them. My options seem to be to either select
> :the first matching one, or to sort the resulting parse trees
> :by the order of the production rules that were used to generate
> :them (resulting in the "most valid" parsing being first). I
> :am still undecided about this.
>
> I've written quite a bit on this topic in the past. While my current
> parsers are in Java, and the rules of the lexical system are a little
> different, there are similarities... I have finally implemented sentance
> completion. As soon as there is a possible completion, the completed
> sentance is printed in gray text ghosted off of the typed text. Hitting
> tab completes a current gray segment, hitting enter submits black only.
If there is an archive of previous postings on the subject, I would
be very interested.
Your parsers are in Java... does that mean that it is the client which
does the parsing?
Felix Croes