As I posted before, I'm working on an Eclipse plugin for JScheme, and
I'm running across some problems, so I figured I'd post to the list to
see what suggestions people came up with.
Basically, for each .scm file the user is editing, I want to be able to
support "intelligent completion" or at the very least, knowledgeable
syntax highlighting. By this I mean, I want to know all (and only) the
symbols that would be loaded up to the cursor point. What's more, I want
this table of symbols to be different for each editor.
If I'm working on on jlib app in one window and a network server app in
another window, I don't want the symbols imported by (or defined in) the
network program to be available in the jlib app, unless of course the
jlib app loads the network server.
One possible solution I've heard is to use a prefix when I load the
file. I would set the symbol prefix to be the full path to the file
(presumably) and this would effectively create namespaces for each file
being edited. Would this be effective? Is there a better way?
mike
--
Michael R Head <bu...@cs...>
Brandeis University CS graduate student
|
