Question

I’ve tried doing this myself with the following:

for i in $(grep something toprefix.log); do sed -i -e 's/^/prefix/' $i; done

First of all, apologies for the bone-headedness – I know the above doesn’t work!

Basically, I want to apply a prefix to matching lines in a log file using grep, and sed or awk.

prefix-matching
prefix-matching
not-matching
prefix-matching
not-matching
not-matching
prefix-matching

Thanks in advance.

Answer 1

$ sed -i '/something/s/^/prefix/' toprefix.log

Edit: For details see http://www.gnu.org/software/sed/manual/html_node/Addresses.html

Edit: Removed ‘<‘ since ‘-i’ is used.

Edit: Tabs are \t, so

$ sed -i '/something/s/^/prefix\t/' toprefix.log

adds prefix and a tab.