On Sat, 22 Jun 2002, Nadav Har'El wrote:
> On Sat, Jun 22, 2002, Shlomi Fish wrote about "[Haifux] The Scannerless Generalized
>LR Parser":
> > It describes a tool that combines the functionality of Lex and Yacc with
> > some improvements. I did not try it yet, but it might be worth to take a
> > look.
>
> Note that in some respects, Lex is highly overrated. It is quite easy to
> write a lexical analyzer for most languages you can think of, in straight
> C code that is still readable and quite easily extendable. All you need to
> do is is to write the yylex() function yourself.
>
One should note that neither Lex and Yacc force the other upon the
programmer. Lex can sometimes be convenient if you don't feel like
re-inventing your own wheel. When working in Perl, Lex is just a small
amount of user-land syntactic sugar, and doesn't do too much.
> So in that sense, Yacc (with C, for example) already combines the
> functionality of Lex and Yacc :)
>
I'm not sure that you write here. The way I understood it this parser can
change the behaviour of the tokenizer based on what the parser told him
(or vice versa - ?). It may be possible with Yacc, but I'm not sure it
would be straightforward.
With Rindolf, I'll probably need to modify the
behaviour of the parser and lexer at run-time (I'd like people to be able
to declare new operators). Same thing with Perl 6 if I understood Larry
Wall correctly.
Regards,
Shlomi Fish
>
>
----------------------------------------------------------------------
Shlomi Fish [EMAIL PROTECTED]
Home Page: http://t2.technion.ac.il/~shlomif/
Home E-mail: [EMAIL PROTECTED]
He who re-invents the wheel, understands much better how a wheel works.
--------------------------------------------------------------------------
Haifa Linux Club Mailing List (http://linuxclub.il.eu.org)
To unsub send an empty message to [EMAIL PROTECTED]
