Yarn/uqwk scoring rules?

From: Howard Schwartz (theo@ncal.verio.com)
Date: Tue, 14 Jul 1998 13:08:16 -0700 (PDT)

Recently, I included a line like the following:

+1 pattern Date: .*Jul

in my global scoring file for uqwk, which claims its rules are similar
to those of yarn. I discovered that some news articles that have July
after the Date: field, received a score of +2 rather than +1. Why?
Because there was another date line in the header something like this:

NNTP-Date:

Therefore the string, Jul appeared twice after the word ``Date:'' in
the header and the scoring rules counted this as 2 points.

My question: Clearly, when one specifies lines like:

Date:
Subject:

The scoring rules do not interpret this as the word Date or Subject, at
the beginning of a line. Is there some way to specify that these fields
must be at the beginning of a line, for instance:

^Date:
^Subject:

Or, as I suspect, are regular expression special characters such as
``^'' forbidden in connection with words like Date and Subject?

Does yarn or uqwk read the entire header of a news post as one streaming
line, or can it somehow recognize individuals lines within a header?