Shell Programming and Scripting

I have a file about cpu load like this (the first column is time, the second column is cpu load):

11:34:01 0.10
11:34:37 0.05
11:35:13 0.03
11:35:59 0.06
11:38:00 0.02
11:38:09 0.17
11:39:48 0.06

I need to find the max value of the second column so I write the script as follows

Code:

max=0
cat a.txt | while read line
do
value=$(echo $line | awk '{print $2}')
if [ $(echo "$max < $value" | bc) -eq 1 ]; then
max=$value
fi
done
echo $max

this one bring the wrong max value. But when I edit with

Code:

while read line
do
..
done < a.txt

it brings the correct value. So why these bring the different answers. Please explain to me.

In the first case the while loop is executed in a subshell ( except in Kornshell ) because of the pipe. This means that any variables set in the loop lose their value once the loop finishes and are no longer available outside the loop..

I get it. Thank you so much.

In bash 4.2 or later, you can do

Code:

shopt -s lastpipe

so that the last pipe runs in the current shell.

Code:

max=$(awk '{if($2>m)m=$2}END{print m}' a.txt)

is simpler and more efficient.

I've forgotten about the powerful awk. thank you for your solution.

just sort by second column, and retrieve the first value by head -1

Code:

 
$ sort -nr -k2 input.txt
11:38:09 0.17
11:34:01 0.10
11:39:48 0.06
11:35:59 0.06
11:34:37 0.05
11:35:13 0.03
11:38:00 0.02