i'll try a commandline-freehander without net...
for lumpi in `sort -u -k2,2 file|awk '{print $2}'`
do
echo "$lumpi occurs `grep $lumpi file|wc -l` times"
done
Let me know if it works...
Regards, Ralph.
-----Ursprüngliche Nachricht-----
Von: Steven Thornhill [SMTP:Steven.Thornhill@SOUTHERNELECTRIC.CO.UK]
Gesendet am: Donnerstag, 18. Februar 1999 16:52
An: NV-L@UCSBVM.ucsb.edu; Schiffinger Ralph 2100
Betreff: Re: Counting lines in a file containing exact fields
You can create an array by awking the second field and sorting
(unique), then use this array with a loop to count the number of lines
that match each element of the array. If you need an outline of a
script let me know.
Steve
______________________________ Reply Separator
_________________________________
Subject: Counting lines in a file containing exact fields
Author: Lucy Premus <lpremus@METLIFE.COM> at SEInternet
Date: 18/02/99 10:06
Is there a korn shell function to count lines in a file that have one field
thats the same. For example, if my file looks as follows:
server1 NYHO
server2 NYHO
server3 NYHO
server4 RISC
server5 RISC
server6 NYMLB
server7 NYMLB
server8 NYMLB
server9 NYMLB
I want the script to read the file and count up that there are 3 NYHO, 2
RISC,
and 4 NYMLB. However theres a catch, in reality my files will be much larger
than this containing much more than 3 different instances. I don't want to
have
to compare each line to the literal (ie. NYHO, RISC or NYMLB) because theres
just too many of them. Is there any easier way?
|