Extracting a specific pattern from a file...

Hi,

I have a ksh script which looks through various logs for the ip address pattern as well as email address pattern and other pattern still to be defined.  How do I extract the found patterns from the file?  I tried using awk but it seems that I need to know the position of the found pattern to get a proper output:

For instance, I tried this:

cat file.txt |  awk '/[0-9]*\.[0-9]*\.[0-9]*\.[0-9]*/{print $5}'

but this assumes that the ip is found in the fifth position in the file (this doesn't happen all the time).

How do I extract just the item i want?

I tried this also but I'm not sure how to do the print part of the statement:

cat file.txt |  awk '/\([0-9]*\.[0-9]*\.[0-9]*\.[0-9]*\)/{print \1}'

I don't think my awk knows about "\1".... or maybe this is only good for sed?