Re: The case of directly-executable parsers

Ranjit Mathew <rmathew@gmail.com>
31 Jan 2006 09:56:02 -0500

          From comp.compilers
Related articles
The case of directly-executable parsers momchil.velikov@gmail.com (2006-01-28)
Re: The case of directly-executable parsers derek@_NOSPAM_knosof.co.uk (Derek M. Jones) (2006-01-28)
Re: The case of directly-executable parsers momchil.velikov@gmail.com (momchil.velikov@gmail.com) (2006-01-30)
Re: The case of directly-executable parsers clint@0lsen.net (Clint Olsen) (2006-01-31)
Re: The case of directly-executable parsers rmathew@gmail.com (Ranjit Mathew) (2006-01-31)
Re: The case of directly-executable parsers 148f3wg02@sneakemail.com (Karsten Nyblad) (2006-01-31)
Re: The case of directly-executable parsers momchil.velikov@gmail.com (momchil.velikov@gmail.com) (2006-02-02)
| List of all articles for this month |
From: Ranjit Mathew <rmathew@gmail.com>
Newsgroups: comp.compilers
Date: 31 Jan 2006 09:56:02 -0500
Organization: Compilers Central
References: 06-01-104
Keywords: parse, yacc, comment
Posted-Date: 31 Jan 2006 09:56:02 EST

momchil.velikov@gmail.com wrote:
>
> What could be the reason for this parser performing so poorly? Or for
> the Bison parser performing so well? Has anyone got some recent


GNU Bison uses a different algorithm from YACC. See:


http://cvs.savannah.nongnu.org/viewcvs/*checkout*/bison/bison/REFERENCES?rev=1.2


I quote:


    "Also, Bison uses a faster but less space-efficient encoding for the
    parse tables (see Corbett's PhD thesis from Berkeley, "Static
    Semantics in Compiler Error Recovery", June 1985, Report No. UCB/CSD
    85/251), and more modern technique for generating the look-ahead sets.
    (See Frank DeRemer and Thomas Pennello, "Efficient Computation of
    LALR(1) Look-Ahead Sets", ACM Transactions on Programming Languages
    and Systems (TOPLAS) 4, 4 (October 1982), 615-649. Their
    technique is the standard one now.)"


HTH,
Ranjit.


--
Ranjit Mathew Email: rmathew AT gmail DOT com
Bangalore, INDIA. Web: http://rmathew.com/
[Every Unix-ish system I know has used Corbett's version of yacc for at least
the past ten years. -John]

Post a followup to this message

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