--- links/csv2js.pl	2007/08/23 13:27:56	7
+++ links/csv2js.pl	2007/08/23 14:08:06	8
@@ -111,10 +111,38 @@
 	warn "prefixes found = ",dump($p);
 	my @sorted = sort { $p->{$b} <=> $p->{$a} } keys %$p;
 	my $strip = shift @sorted || return @_;
-	warn "longest preffix: '$strip'\n";
+	warn "longest prefix: '$strip' (stripped)\n";
 	return map { my $v = $_; $v =~ s/^\Q$strip\E//i; $v; } @_;
 }
 
+sub group_by {
+	my ( $group, $detail ) = ( shift, shift );
+	my $what = shift;
+	confess "expected CODE as first argument!" unless ref($what) eq 'CODE';
+	my @data = @_;
+
+	my $stat;
+	my @details;
+
+	foreach my $i ( 0 .. $#data ) {
+		my $v = $data[$i];
+		my ( $by, $rest ) = $what->($v);
+#		warn "## group_by: $i $v -> $by\n";
+		push @{ $stat->{$by} }, $i;
+		$details[$i] = $rest;
+	}
+
+	warn "group_by '$group' stats = ",dump( $stat );
+
+	foreach my $g ( keys %$stat ) {
+		foreach my $r ( @{ $stat->{$g} } ) {
+			warn "## $group $g $r\n";
+			$dump[$r]->{$group} = $g;
+			$dump[$r]->{$detail} = $details[$r];
+		}
+	}
+}
+
 while (<$csv_fh>) {
     $csv_parser->parse($_);
     my @fields = $csv_parser->fields;
@@ -145,10 +173,6 @@
 
 				my $v = clean( $sv[$j] );
 
-				if ( $j == 0 ) {
-					$h->{ $l . '_short' } = $v;
-				}
-
 				if ( my $human = human( $v ) ) {
 					$h->{ $l . '_' . $j . '_human' } = $human;
 				} else {
@@ -201,11 +225,17 @@
 
 warn "split_stats = ", dump( $split_stats ), "\n";
 
+#
+# split prefix from label_0
+#
 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 );
 
+# group products by manufacturers
+group_by( qw/manufacturer player_name/, sub { $_[0] =~ m/^(\S+)\s+(.+)/; ($1,$2) }, @stripped );
+
 warn "dump = ", dump( @dump ), "\n";
 
 print "features: .", join(', .', keys %$split_stats), "\n";