collect A_ counts (original data stats) only on first loop

diff --git a/upari.pl b/upari.pl
index 3430566..cbb178f 100755 (executable)
--- a/upari.pl
+++ b/upari.pl
@@ -10,6 +10,8 @@ use Data::Dump qw(dump);
 use String::Similarity;
 use Storable;
 
+my $debug = $ENV{DEBUG};
+
 my @files = qw( 1.csv 2.csv 3.csv 4.csv );
 
 our $stat;
@@ -20,7 +22,7 @@ sub duplicate {
        my $use;
        $use->{$_}++ foreach ( map { @{ $stat->{_}->{$_} } } @for );
        my $duplicate = grep { $use->{$_} > 1 } keys %$use;
-       print "XXX use @for ",dump($use),$/ if $duplicate;
+       print "XXX use @for ",dump($use),$/ if $debug && $duplicate;
        return $duplicate;
 }
 
@@ -44,13 +46,13 @@ sub candidates {
                                #print "XXX use $num $key ",dump($use);
                                my $duplicate = grep { $use->{$_} > 1 } keys %$use;
                                if ( $duplicate ) {
-                                       print "XXX $limit_sim suggest duplicate  $num $key SKIP duplicate ",dump($use), $/;
+                                       print "XXX $limit_sim suggest duplicate  $num $key SKIP duplicate ",dump($use), $/ if $debug;
                                        $stat->{suggest}->{duplicate}++;
                                } else {
                                        push @candidates, { key => $key, s => $s };
                                }
                        } else {
-                               print "XXX $limit_sim candidates $key missing\n";
+                               print "XXX $limit_sim candidates $key missing\n" if $debug;
                        }
                }
        }
@@ -121,6 +123,7 @@ foreach my $nr ( 1 .. 4 ) {
 
 
 
+my $first = 1;
 # 0.9 - 0.7 -- 0.6 is too lax
 foreach my $limit ( 0.7, 0.6 ) {
 warn "XXX limit $limit\n";
@@ -135,15 +138,17 @@ foreach my $id ( sort keys %{ $stat->{_} } ) {
                #print "SKIP[$id]";
                next;
        }
-       $stat->{A_count}->{ scalar @val }++;
-       $stat->{A_count_total}++;
-
-       #$stat->{A_count_val_dup}->{ join(' ', @val) }++; # with duplicates
-
        my $u;
        $u->{$_}++ foreach @val;
        my @u_v = sort keys %$u;
-       $stat->{A_count_val}->{ join(' ', @u_v ) }++; # without duplicates
+       if ( $first == 1 ) {
+               $stat->{A_count}->{ scalar @val }++;
+               $stat->{A_count_total}++;
+
+               #$stat->{A_count_val_dup}->{ join(' ', @val) }++; # with duplicates
+
+               $stat->{A_count_val}->{ join(' ', @u_v ) }++; # without duplicates
+       }
 
        if ( $#u_v < 3 ) { # single, double
                my ( $num, $key_id ) = split(/-/,$id,2);
@@ -204,7 +209,7 @@ foreach my $id ( sort keys %{ $stat->{_} } ) {
        $stat->{"B${limit}_count_val"}->{ join(' ', @u_v ) }++; # without duplicates
 }
 
-
+$first++;
 } # for $limit
 
 print "# stat = ",dump( $stat );