Shell Programming and Scripting

Im writing a KSH script to read a simple text file and add a delimiter. Ive written the following script but it runs very slow. I initially used the cut command to substring the input record then switched to this version using awk to substring... both run too slow. Any ideas how to make this more efficient? Its taking around 10min for a 40k record file(2 meg)... very slow for such small data.

data file(input.txt):

99919xxxxxxxZ3700162111
99921xxxxxxxZ3300563111
99922xxxxxxxZ5800143111
99923xxxxxxxZ5500056111
99925xxxxxxxZ5100366111

script:

#!/bin/ksh
while read x
do
f1=$(echo "$x"|awk '{print substr($0,1,5)}')
f2=$(echo "$x"|awk '{print substr($0,6,4)}')
f3=$(echo "$x"|awk '{print substr($0,10,2)}')
f4=$(echo "$x"|awk '{print substr($0,12,3)}')
f5=$(echo "$x"|awk '{print substr($0,15,1)}')
f6=$(echo "$x"|awk '{print substr($0,16,2)}')
f7=$(echo "$x"|awk '{print substr($0,18,5)}')
f8=$(echo "$x"|awk '{print substr($0,23,1)}')
f9=$(echo "$x"|awk '{print substr($0,24,1)}')
f10=$(echo "$x"|awk '{print substr($0,25,1)}')

delimrecord="$f1|$f2|$f3|$f4|$f5|$f6|$f7|$f8|$f9|$f10"
echo $delimrecord >> output.txt
done < input.txt

You invoke an external program 10 times for every record - that's why it is slow

try something like this:

Code:

awk ' {
printf("%s|%s|%s|%s|%s|%s|%s|%s|%s|%s\n",
      substr($0,1,5) ,
      substr($0,6,4) ,
      substr($0,10,2),
      substr($0,12,3),
      substr($0,15,1),
      substr($0,16,2),
      substr($0,18,5),
      substr($0,23,1),
      substr($0,24,1),
      substr($0,25,1) ) } ' infile.txt > outfile.txt

Thanks Jim. Thats alot more efficient than the mess I created!