--- links/csv2js.pl	2007/08/23 13:27:14	6
+++ links/csv2js.pl	2007/08/23 13:27:56	7
@@ -75,7 +75,7 @@
 sub human {
 	my $s = shift;
 
-	if ( $s =~ m/(\d+)\s*($multiplier_regex)/i) {
+	if ( $s =~ m/^\s*(\d+)\s*($multiplier_regex)\s*$/i) {
 		my ( $v, $m ) = ( $1, lc($2) );
 		my $factor = $multiplier->{$m};
 		confess "can't find multiplier $m" unless defined $factor;
@@ -86,6 +86,35 @@
 	return;
 }
 
+sub strip_prefix {
+	my @data = @_;
+	my $prefix = shift @data;
+
+	my $p;
+
+	foreach my $d ( @data ) {
+		my $chomp = length($prefix);
+		# find end of common string
+		$chomp-- while(
+			lc(substr( $prefix, 0, $chomp )) ne lc(substr( $d, 0, $chomp ))
+			&&
+			$chomp > 0
+		);
+		if ( $chomp == 0 ) {
+			warn "no common prefix in ",dump( @_ );
+			return @_;
+		}
+
+		my $prefix = substr( $prefix, 0, $chomp );
+		$p->{$prefix}++;
+	}
+	warn "prefixes found = ",dump($p);
+	my @sorted = sort { $p->{$b} <=> $p->{$a} } keys %$p;
+	my $strip = shift @sorted || return @_;
+	warn "longest preffix: '$strip'\n";
+	return map { my $v = $_; $v =~ s/^\Q$strip\E//i; $v; } @_;
+}
+
 while (<$csv_fh>) {
     $csv_parser->parse($_);
     my @fields = $csv_parser->fields;
@@ -126,7 +155,7 @@
 					$h->{ $l . '_' . $j } = $v;
 				}
 
-				$split_stats->{$v}->{$j}++;
+				$split_stats->{$v}->{pos}->{$j}++;
 				$split_stats->{$v}->{sum}++;
 				push @{ $split_stats->{$v}->{rec}->{$#dump + 1} }, $j;
 			}
@@ -170,7 +199,12 @@
 	}
 }
 
-#warn "split_stats = ", dump( $split_stats ), "\n";
+warn "split_stats = ", dump( $split_stats ), "\n";
+
+my @all = map { $_->{label_0} || die "no label_0 for ",dump($_) } @dump;
+warn "all = ",dump(@all);
+my @stripped = strip_prefix( @all );
+$dump[$_]->{label_0} = $stripped[$_] foreach ( 0 .. $#stripped );
 
 warn "dump = ", dump( @dump ), "\n";