Related articles |
---|
[3 earlier articles] |
Re: Parsing wars bromage@mullauna.cs.mu.OZ.AU (1992-09-02) |
Re: Parsing wars bburshte@pyramid.com (1992-09-03) |
Re: Parsing wars jar@cheops.HQ.Ileaf.COM (1992-09-05) |
Re: Parsing wars dww@inf.fu-berlin.de (1992-09-08) |
Re: Parsing wars bruce@harry.ugcs.caltech.edu (1992-09-09) |
Re: Parsing wars mickunas@m.cs.uiuc.edu (1992-09-10) |
Re: Parsing wars bburshte@pyramid.com (1992-09-13) |
Newsgroups: | comp.compilers |
From: | bburshte@pyramid.com (Boris Burshteyn) |
Organization: | Compilers Central |
Date: | Sun, 13 Sep 1992 21:16:18 GMT |
Keywords: | parse |
References: | 92-09-006 92-09-043 |
jar@cheops.HQ.Ileaf.COM (Jim Roskind x5570) writes [about "strange parsing"]
>The conseqeunces of this extension are quite significant, and the
>resulting parser engine provides a much larger class of parsers than LR
>parsers. ...
The idea of several passes over the reduction stack has also been
expressed in my paper in SIGPLAN NOTICES V.27 N.1, 1992.
A parser built upon a modifiable grammar changes the set of available
context-free rules on the fly while parsing the source. If it cannot
continue, it backtracks and tries to reduce tokens (terminals &
nonterminals) from the stack. It may indeed find something to reduce
because of new rules introduced dynamically. Several backtracking attempts
can be made, and the set of rules may be changed while reparsing the
stack. There is also an example of parsing of expressions before
declaration of variables.
Regards - Boris Burshteyn (bburshte@pyramid.com)
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.