At 7:56 PM +0100 8/24/02, Simon Cozens wrote:
>The reason I decided to use a tree rather than bytecode was nothing to do
>with efficiency, backtracking or anything else. It was about flexibility.
>More specifically, the flexibility to modify or rewrite portions of the
>grammar, and particularly to be able to do so on the fly, during a match
>if need be.
Right. That's one of the reasons we're keeping the ASTs around for
compiled code. (And even if we weren't, it's more than enough of a
reason to keep the tree form of the grammars around)
--
Dan
--------------------------------------"it's like this"-------------------
Dan Sugalski even samurai
[EMAIL PROTECTED] have teddy bears and even
teddy bears get drunk
