Front page | perl.bootstrap |
Postings from July 2000
According to Simon Cozens:
> Having a combined lexer-parser may freak out you computer scientists,
> but for us computational linguists, it's standard NLP practise.
I think it's reasonable to try to reduce the coupling, both as a goal
and as one metric of language complexity. But I agree that trying to
eliminate the coupling is probably futile.
In accord with Laziness, it's worth considering a change from the
lexer+BNF parser strategy to something with less of an impedance
mismatch with Perl's nature. Antlr, say, or PCCTS. Or even a hand-
written recursive descent parser. We could use Parse::RecDescent to
prototype it....
--
Chip Salzenberg - a.k.a. - <chip@valinux.com>
"I wanted to play hopscotch with the impenetrable mystery of existence,
but he stepped in a wormhole and had to go in early." // MST3K