I have a file, file.txt, like this:
start cal
end cal
start eff
end eff
start cal
error
end cal
start dod
end dod
What I want to get is the line between the last set of 'start cal' and 'end cal'. I am using tac and grep to do it. But in vain. Any help please? Below is my code:
tac file.txt | grep -m1 'start cal*end cal'
What am I doing wrong? For the example above, I need the command to return 'error'.
Grep is not good tool for this task. It does line based matching. You can consider using awk.
if you have single line in between:
tac file|awk '/end cal/{p=1;next}/start cal/{exit}p'
If it could be multiple lines:
tac file|awk '/end cal/{p=1;next}/start cal/{exit}p'|tac
It outputs:
error
/pattern/ { commands } variable syntax? I don't recognize it.
I don't know if grep can do it, but here it is in gawk. I'll be the first to admit, it's not exactly pretty.
tac file.txt | gawk '/start cal/ { exit } { if (echo) print } /end cal/ { echo = 1 }'
n.b. Kent's answer is better polished.
You're doing two things wrong:
start cal*end cal matches start ca, followed by 0 or more ls, followed by end cal