script/showdups - show all duplicate files in current repo

script/showdups

@@ -0,0 +1,38 @@
+#!/usr/bin/env perl
+
+use v5.10;
+use warnings;
+
+open my $fh, "-|", "git ls-tree -r HEAD"
+  or die $!;
+
+my %fs_sha1_path;
+
+while ( my $line = <$fh> ) {
+    chomp $line;
+
+    if ( $line =~ m{ \A \d+ \s+ \w+ \s+ (\w+) \s+ (.*) \z }x ) {
+        my $sha1= $1;
+        my $path= $2;
+        if ( -f $path and !-l $path ) {
+            my $fs= -s $path;
+            $fs and push @{ $fs_sha1_path{ $fs }{ $sha1 } }, $path;
+        };
+    }
+}
+
+for my $fs ( sort { $a <=> $b } keys %fs_sha1_path ) {
+    for my $sha1 (keys %{ $fs_sha1_path{ $fs } } ) {
+       if ( @{ $fs_sha1_path{ $fs }{ $sha1 } } > 1 ) {
+           for my $path ( sort @{ $fs_sha1_path{ $fs }{ $sha1 } } ) {
+               say sprintf "%10d %s %s", $fs, substr($sha1,0,7)."..".substr($sha1,-3), $path;
+            }
+            say '-', ' -' x 39;
+        }
+    }
+}
+
+exit;
+
+__END__
+