Question

I have trouble with parsing a logfile… Here is a logfile sample (in real live there are bash color codes included):

foo completely unintresting line bar
foo [INFO] [#############] bar 
foo [INFO] [#######______] bar 

I need the parts “[#############]” and “[#######__]”. To get these information I drop the unintresting lines with:

CMD_TAIL=`tail -n 3 server.log | sed -r "s/\x1B\[([0-9]{1,3}((;[0-9]{1,3})*)?)?[m|K]//g" | grep -E '\[INFO\] \[[#_]+\]'`

But if I “walk” the CMD_TAIL I get this (whitespace replaced by newline):

1: foo :2
1: [INFO] :2
1: [#############] :2
1: bar :2
...

How do I get the complete line so I can extract the needed information?

Answer 1

In case you’re in a bash-like shell, try setting the internal field separator to newline before your loop:

IFS=$'\n'
for TAIL in $CMD_TAIL; do
    echo "1: $TAIL :2"
done