jakarta-oro-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Daniel F. Savarese" <...@savarese.org>
Subject Re: Question:
Date Fri, 30 Mar 2001 05:45:46 GMT

>Let R1 be the regular expression 1
>Let R2 be the regular expression 2
>
>R1 < R2 iff all inputs that R1 can match R2 can, but not vice versa.

Although an interesting question, it doesn't have much to do with the
development of the jakarta-oro software and should probably be
directed to a mailing list or newsgroup on lexical analysis.  However,
I will venture to say it is a solvable problem for DFA and even some
NFA representations of regular expressions, but not for Perl5
expressions.  You have to be able to compute closures and show that
the closure of R2 over all of its inputs is a superset of the closure
of R1, which is straightforward for DFAs.

daniel



Mime
View raw message