sorry about the original text formating, it totally screwed up the way it looked like.
But yeah, thanks for the clarification, I think at least this:

match something that is not a number: "[\\d$]"

should write as:

match something that is not a number: "[^\\d$]"

OPERATING REVENUES:
publishing        $   42,419    $   44,754     $  46,203
collegiate marketing and production services      97
ASSOCIATION MANAGEMENT SERVICES      16
wireless                   8,883              8,129               7,507
                               51,302             52,883             53,823


On 8/2/06, John Maddock < john@johnmaddock.co.uk> wrote:
Winson Yung wrote:
>> Thank you John, right the difference is in the header. Your regular
>> expression however doesn't make sense to me though, where is the
>> part that matchs the sub-total header?

consume whitespace: "\\s+"
match something that is not a number: "[\\d$]"
consume the rest of that line: "[^\r\n]*[\r\n]+"
consume the next line as well: "[^\r\n]+[\r\n]+"

All of that was in a (?: ... )+ block since it can be repeated several times
before we get to a line with no header.

Clear now?

John.


_______________________________________________
Boost-users mailing list
Boost-users@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-users