Interpreter advice required

jwillans@cs.york.ac.uk (James Willans)
29 Apr 2002 01:52:51 -0400

          From comp.compilers

Related articles
Interpreter advice required jwillans@cs.york.ac.uk (2002-04-29)
Re: Interpreter advice required lars@bearnip.com (2002-05-03)
Re: Interpreter advice required per@bothner.com (Per Bothner) (2002-05-08)
| List of all articles for this month |
From: jwillans@cs.york.ac.uk (James Willans)
Newsgroups: comp.compilers
Date: 29 Apr 2002 01:52:51 -0400
Organization: http://groups.google.com/
Keywords: interpreter, question
Posted-Date: 29 Apr 2002 01:52:50 EDT

Hi,


I often read this newsgroup and was hoping that you may be able to
offer me some advice.


I'm working on implementing an interpreter (in Java) for a small
object-oriented prototype language (based on Abadi's and Cardelli's
sigma calculus, but with a larger syntax). I have little background
in compiler theory, but have been making steady progress towards
achieving my goal. Thus far I have implemented a parser using javacc
which builds an abstract syntax tree. The next step, as far as I can
see, is to traverse this syntax tree and generate a stream of
instructions that will execute on a virtual machine. I plan to do
this by giving each syntax concepts knowledge about how to compile
themselves (probably simply a 'compile' method). The instructions
will then be executed on the virtual machine.


My problem is that I have little understanding about the runtime
organisation of object oriented programs. I have been using Watt's
"Programming language processors in Java" which has helped a lot, but
doesn't deal indepth with how to make decisions about the form the
virtual machine will take given the language that is going to be
interpreted (clearly it is not necessary to simulate a real machine).
In fact, most books I have looked at don't deal with object oriented
programming language interpretation.


Any pointers to useful material is greatly appreciated.


James


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.