I have non-standard data, which I'd like to standardise
file:
d101 11001 e101 9665 f101 9663 d102 11002 e102 11003 f102 11004 g102 11005 desired output:
d101 11001 e101 12001 f101 12002 d102 11002 e102 11003 f102 11004 g102 11005 so the logic should be, if length of column2 = 4 it should replace it with incremental numbering of a provided series: in this case 1200 is series, & 1, 2, 3 .. are increments.
$ awk -v n=12000 'length($2)==4 {$2=++n} {print}' file d101 11001 e101 12001 f101 12002 d102 11002 e102 11003 f102 11004 g102 11005 Note that we first increment n and then assign, to use the new value. If we wanted to start printing from 12000 we would use: $2=n++, first assign and then increase.
Using Raku (formerly known as Perl6)
raku -ne 'state $i; print .words[0]~" "; put S/^ \d**4 $/{++$i + 12000}/ given .words[1];' Briefly, raku is called at the command line with the -ne (linewise, non-autoprinting) flags. A state variable is declared (once and only once), for incrementing within the code. In the second statement the .words[0] first whitespace separated 'word' (i.e. column) is printed followed by a ~ tilde concatenated " " blank space (for separating the output columns). In the third statement the S/// (big-S) substitution command searches for \d**4 exactly 4 digits within given .words[1] (the second column). A match is replaced with {++$i + 12000} a computed value (curlies denote code blocks within regexes), and returned (because S/// big-S substitution returns the modified string).
Sample Input:
d101 11001 e101 9665 f101 9663 d102 11002 e102 11003 f102 11004 g102 11005 Sample Output:
d101 11001 e101 12001 f101 12002 d102 11002 e102 11003 f102 11004 g102 11005 
