#!/bin/sh
# finds regexps in sequences. Prints positions [and substring matched]
# usage FindRegexp pattern [p] SequenceFile (tbl)
# input in lowers

gawk 'BEGIN{
	regexp=ARGV[1]; # pattern
	if (ARGC==4) {
  		s=1;	# pattern match is printed.
		ARGV[2]="";
	}
	ARGV[1]="";
}
{
	locus=$1;
	sequence=$2;
	cp=0; # position of last match
	nm=0; # number of matches

	while (match(sequence, regexp) > 0) {
		nm++;
		smatch[nm,1]=(cp+=RSTART); # position in string

		smatch[nm,2]=substr(sequence, RSTART, RLENGTH); # matched substring
		sequence=substr(sequence,RSTART+1);
	}
	printf "%s", locus;
	for (i=1;i<=nm;i++)
		if (s) 
			printf " %d %s", smatch[i,1], smatch[i,2];
		else
			printf " %d", smatch[i,1];
	printf "\n";
}' "$@"