|Interpreter. What is the best way ? firstname.lastname@example.org (1991-11-14)|
|Re: Interpreter. What is the best way ? email@example.com (1991-11-14)|
|Re: Interpreters (and now, byte code standards) firstname.lastname@example.org (1991-11-15)|
|Re: Interpreter. What is the best way ? email@example.com (1991-11-16)|
|From:||firstname.lastname@example.org (Gavin Thomas Nicol)|
|Date:||Thu, 14 Nov 91 18:16:59 JST|
I'm not sure if this is a question that really fits here, but I am have
started to write an application extension language, that combines aspects
of C and PASCAL (ie. it should be easy for either kind of programmer to
pick up on, even though it will be a subset.)
This language will be interpreted, and herein lies the question. For
this kind of system, is it better to compile the program into bytecode,
into a parse tree (like gawk), or into tokenised input ? Obviously,
interpreting the sources directly is NOT the quickest way of doing things.
I should also note that I want to be able to include this language into
various applications, and want it to be easily extensible. For example, in
order to register an new function (in the source) I just want to add a
declaraction to a table.
If anyone has any comments, please post to the net, or email me. I will
Thanks in advance.
Gavin Nicol (alias "nick" or "nick-san")
NEC Scientific Information System Development Ltd.
R&D KSP Bldg.
100-1 Sakato Takatsu-ku, Kawasaki, Kanagawa-ken 213,
Phone : <Japan> (044) 812-8411
Return to the
Search the comp.compilers archives again.