As others have already stated, you've posted insufficient information for us to provide a solution. It sounds like you probably want code like this:

for my $line (@lines) {
    next if $line =~ /\d/;
    print $line if $line =~ $regex_that_is_working;
}
[download]

The regex you posted (with the vague "it's not working") has a lot of potential problems:

• You have a boundary assertion (\b) in one place and two plain 'b' letters elsewhere: are they also meant to be assertions?
• You have one alternation (with |) and later an escaped pipe ('\|'): was that supposed to be another alternation?
• You're using a capture but don't access what is captured. What's the intent here?
• You have '0-9' which I suspect should be '[0-9]'; I'm even wondering if '!0-9' should be '[^0-9]'.
• You have what appears to be anchor assertions ('^' and '$') in strange positions; there is no 'm' modifier to indicate a multiline string.
• There are other bits I just gave up trying to guess at.

Take a look "perlintro: Regular expressions". This has introductory information on that page; it also has many links to more detailed documentation on other pages. In addition, search for /regular expression/ on the main "perldoc - perl" page.

The "How do I post a question effectively?" page, on this site, will give you an idea of what you need to post to get better answers from us.