That's why I say "regular expressions".
There is not a much better understood class of expressions than REs,
their grammars being among the class of grammars that are equivalent,
in an important mathematical sense, to finite state machines.[1]
Haven't looked at Sun's implementation, but I would not at all be
surprised to see them compile down to FSAs to do acceptance.
-Fred
[1] http://portal.acm.org/citation.cfm?coll=GUIDE&dl=GUIDE&id=48032
On Apr 5, 2007, at 11:46 AM, Polar Humenn wrote:
Then we better have some specific semantics about which "match" is
selected, so it's predictable by thinking about apriori, and not
discovery by subsequent testing.
