I have a file with below data
cat test.txt
010001010000010100000000000000000000000000000000000000000000000000000000000000000000000000000000000
two fields in above line represents usage one digit starting from 16
I want the the o/p in such that if 01 then it should print the digit, else put , and check next field.
first two fields 01 means 16 is used so it should be printed in o/p file 2nd two field 00 mean 17 is not used, so it should not be printed in o/p file ans so on.
Like wise there are 4000 fields, means 2000 digits starting from 16-2016
For above line , need output like
16,18,19,22,23,
Here's what I have so far.
cat counter_mfd.txt | cut -d\' -f2 | sed 's/.\{2\}/& /g' | awk -F, '{a[$2]=$1;} END{for(i in a)print i" "a[i];}'
awk: record `00 00 00 00 00 00 00...' too long
This should work as Perl is pretty normal on Solaris:
perl -e '$s=16; foreach (<>=~m/.{2}/g){if($_=="01"){print "$s,"};$s++}' < test.txt
Output
16,18,19,22,23,
Explanation
$s is just the starting value.
foreach (<>=~m/.{2}/g) splits stdin into groups of two characters and then iterates over them setting $_ to each one in turn.
The if checks if 2-digit string is "01" and prints $s if it is.
Then $s is incremented whether the value was printed or not.