tags:

views:

37

answers:

2
$ cat read.sh 
#!bin/bash

// how can I read the columnwise data to awk-script?
awk '{sum+=$1} END {print sum}' read
$ cat data 
1
2
3
4
5
$ . ./read.sh <data
awk: cmd. line:1: fatal: cannot open file `read' for reading (No such file or directory)
+1  A: 

Remove the filename from the end of the awk command:

Change

awk '{sum+=$1} END {print sum}' read

to

awk '{sum+=$1} END {print sum}' 

The 1st one tell awk to get the input from a file named read where as the 2nd one tells awk to get the input from standard input.

The way you are running the script: ./read.sh <data
You are supplying the input through standard input.

Alternatively if you always want the script to read input from the file named data, you can do:

awk '{sum+=$1} END {print sum}' data

and run the script as: ./read.sh

codaddict
You can also use - as the filename for stdin instead of leaving it blank. I find that easier to read.
drawnonward
A: 

Also, your she-bang line is garbled; it should be #!/bin/bash. But this should work instead:

#!/usr/bin/awk -f

{ sum += $1 }
END { print sum }
Mark Edgar