|Intermediate compiling email@example.com (Niklas Elmqvist) (1996-11-10)|
|Re: Intermediate compiling kuznetso@MIT.EDU (1996-11-12)|
|Re: Intermediate compiling dgay@barnowl.CS.Berkeley.EDU (1996-11-12)|
|Re: Intermediate compiling J.C.Highfield@maybeso.demon.co.uk (Julian Highfield) (1996-11-18)|
|From:||kuznetso@MIT.EDU (Eugene Kuznetsov)|
|Date:||12 Nov 1996 21:58:35 -0500|
Niklas Elmqvist <firstname.lastname@example.org> wrote:
> The problem is that parsing this (the text string) during run-time is
> not very efficient (i.e. slow). We got pointed in the direction of
> compiling the catches into some pseudo-binary code, i.e. code that is
> very fast for the MUD to read (a bunch of byte opcodes), but which is
> not true binary code.
Well, if you were willing to change the syntax a little bit, you could
use java instead. Chances are there is already an implementation on
your platform, but you can also make a quick little interpreter for
yourself (or license one). The main benefit of this is that it will
use a reasonably standard language & compiled format which is much
quicker interpreted than re-parsing the source. And you should be
able to use other people's code without having to design and implement
your own compilers, virtual machines, etc.
Tcl is also a good idea, although I'd lean towards java for something
like this for a number of reasons.
Return to the
Search the comp.compilers archives again.