[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[help-3dldf] Re: Bison parser for machine-like language
From: |
Laurence Finston |
Subject: |
[help-3dldf] Re: Bison parser for machine-like language |
Date: |
Wed, 5 Jan 2005 19:25:57 +0100 (MET) |
On Wed, 5 Jan 2005, Hans Aberg wrote:
> It looks like something to ask in the newsgroup comp.compilers.
I thought about that, but I don't think my query would be appropriate
there, because I'm specifically interested in using
Bison for the parsing. As far as the language design is concerned, I
think the problem is fairly straightforward, since the language won't have
to do very much. All I want to do is dump information from a `Scanner_Node'
(discussed in previous threads) and then read the dumped file to create
`Id_Maps' and `Id_Map_Entry_Nodes' on a `Scanner_Node' in a subsequent
run of GNU 3DLDF. The `Id_Map_Entry_Nodes' contain the objects created by
parsing an input file, so that `Scanner_Nodes' represent the
state of a parse.
I will probably model the language on Knuth's
MMIX language, which is designed to be like current machine languages, but
it won't have to be so complicated.
I've "compiled" Emacs-Lisp code into byte-code, but I don't really know
what the latter is. I was just planning to use C++'s facilities for
reading and writing binary files. I've tried this with C, but some time
ago, and I've never done it with C++. I've always assumed that byte-code
would be less compact than such binary files, but I've never gone into
this subject before.
Thanks,
Laurence