Data Validation

Shell Programming and Scripting
1

I have a comma delimited file that I need to validate the data in one two columns in.

Sample File:

1234,1234,1234,DESCRIPTION,1,1,2
1234,1234,1234,DESCRIPTION,1,1,2
1234,1234,1234,DESCRIPTION,1,1,2
1234,1234,1234,DESCRIPTION,1,1,2

I need to make sure that the second column's entries are numeric. If they have an alpha character, I need to delete that line of data and write it to another file.

I'm not sure where to get started, and any advice would be appreciated.

Thanks!

2 
#!/bin/ksh

validate()
{
    val="$2"
    # remove all numbers = some left = not numbers
    # new=$( echo "$val" | tr -d "[0-9]" )
    # without external command:
    new=${val//[0-9]/}
    [ "$new" != "" ] && echo "$*" >&2 && return 1
    echo "$*"
}

oifs="$IFS"
while read line
do
     IFS=","
     validate $line
     IFS="$oifs"
done

And use this script:

cat file | ./thisscript   > ok.txt  2>err.txt
3

You can try alaso awk solution:

awk -v FS=',' -v OFS=',' '{ 
	if ($2 ~ /^[0-9]*$/)
		print $0
	else
		print $0 >> "err_file"
}' input_file
4

try :

sed -n '/.*,[0-9]*,.*,.*,.*,.*,.*/p' filename