CODEWITHSUNDEEP
unixawk
San
San

Reputation: 31

How to print the value of a column using awk command?

I am trying to write a script that would print the CPU utilisation and memory utilisation result to a csv file. But I would like to only print the values that lie in the specific columns named "CPU", "USERNAME" and "MEMORY".

Could you kindly tell me how can I achieve the same using awk script?Using awk, I have found that awk'{print $2,$5}' prints the values of the 2nd and 5th column. But I would like to print the values by providing the column name instead of the column number.

Is it possible to do that?

The sample script I have written is:

#!/bin/sh
#This is a comment
#This is my first shell script
echo "System Status Report"
date
echo "CPU Utilisation Report"
prstat -a > CPU.xls

The sample output is:

PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP 
26652 root 22M 9648K sleep 1 0 94:34.26 0.2% gpp/1 
26854 root 708M 697M sleep 3 0 36:23.19 0.1% logscan/1 
25167 oracle 1254M 1225M sleep 1 0 14:07.01 0.0% oracle/10 
NPROC USERNAME SIZE RSS MEMORY TIME CPU 
41 root 1000M 834M 1.1% 143:47.47 0.4% 
60 oracle 72G 70G 99% 65:07.13 0.1%

I would like to get the following output in the file:

USERNAME CPU 
root 0.2%  
root 0.1%
oracle 0.0%  
USERNAME MEMORY 
root 1.1% 
oracle 99%

Upvotes: 0

Views: 1855

Answers (1)

Amarpreet Singh
Amarpreet Singh

Reputation: 2270

Yes, definitely. This is the solution you can implement. Say temp.tmp is the file that contains the data. 

# cat temp.tmp
Sno CPU USERNAME
1 50 foo
2 60 bar
3 65 foobar

t.awk has the awk code

# cat t.awk
NR==1 {
    for (i=1; i<=NF; i++) {
        ix[$i] = i
     }
}
NR>1 {
    print $ix[c1], $ix[c2]
}

Now lets see if it works. 

# awk -f t.awk c1=CPU c2=USERNAME temp.tmp
50 foo
60 bar
65 foobar

I think this is what you want isn't it ?

Upvotes: 1

Related Questions