I have a list file which contains:
1.1.1.1:50
1.1.1.2:60
1.1.1.3:70
I am trying to make two variables and execute an internal application to hash the first column with the second one. What I did is these loops, but they give me 9 possibilities and I need only three:
This is my script
#!/bin/bash
for a in $(cat list | cut -d ":" -f 1 )
do
for b in $(cat list | cut -d ":" -f 2 )
do
echo $a,$b # this is example
done
done
sed 's/:/ /' list |
while read ipv4 port
do
echo "IPv4 address: $ipv4; Port number: $port"
done
The only thing to be aware of is that if you set other variables in the loop, they will be run in a subshell and won't affect the rest of the script. There are ways around that if need be (look up process substitution).
sed -r 's/(.*):(.*)/IPv4 address: \1; Port number: \2/' listecho in the original is marked as "this is an example", the same applies to my answer. You're right; if the objective is simply to echo data to standard output, there's no need for the loop. It's likely that the objective is to run wget or curl or telnet or something using the IP address and port number, in which case the simple sed substitution isn't sufficient. The port numbers don't really give the game away — they're fake too, for all practical purposes. (The question says "running application to hash the two values".)$IFS instead of sed: while IFS=":" read ip port ; do echo "$ip -> $port" ; done < listsed or similar techniques in preference to tinkering with IFS. I find it more robust and the cost negligible. There are those who insist that you should do everything in the shell that can possibly be done in the shell (rather than use the shell to execute an external command) to avoid overhead. That's fine — but I often find the concern for 'efficiency' over-rated and the reduced clarity of the script less helpful. YMMV.while read over for .. in, and you showed that!