|Backtracking and parsing tools email@example.com (A Johnstone) (2004-10-02)|
|RE: Backtracking and parsing tools firstname.lastname@example.org (Quinn Tyler Jackson) (2004-10-02)|
|Re: Backtracking and parsing tools email@example.com (2004-10-04)|
|RE: Backtracking and parsing tools firstname.lastname@example.org (Quinn Tyler Jackson) (2004-10-09)|
|Re: Backtracking and parsing tools email@example.com (Paul Tarvydas) (2004-10-09)|
|Re: Backtracking and parsing tools firstname.lastname@example.org (Jim Cordy) (2005-01-12)|
|Date:||4 Oct 2004 00:45:47 -0400|
|Organization:||AOL Bertelsmann Online GmbH & Co. KG http://www.germany.aol.com|
|Posted-Date:||04 Oct 2004 00:45:47 EDT|
Quinn Tyler Jackson <email@example.com> schreibt:
>One might say that, in a sense, the parser "looks ahead" for the ")"
>before invoking the deep parse ... or one might say that the parser
>(with the benefit of hindsight) "looks back" when it sees that it is
>worthwhile to do so.
Isn't that something like stepwise refinement? A simple parser for the
LL(1) parts of the grammar, and other techniques for the constructs of
higher degree? Provided that reasonable boundaries for the first
guess exist, such a parser also can be used with incomplete or
erroneous code, as occurs during code input.
BTW, your paper gave me interesting and useful insight into your
concepts. Not that I understood all the theory now, but I got a much
better feeling for MetaS :-)
Return to the
Search the comp.compilers archives again.