X-Git-Url: http://git.hungrycats.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=faster-dupemerge;h=9753e5aacb2a74ffb9b89fa0a4eec5636d600c90;hb=a4de3164f2ee36a4680c9153fda3b28435a77328;hp=14360dc8c802d83c89132fe5780b7070fd0896f0;hpb=a433cb10cd4e633289a448649f9509e1b24d9c7e;p=dupemerge

diff --git a/faster-dupemerge b/faster-dupemerge
index 14360dc..9753e5a 100755
--- a/faster-dupemerge
+++ b/faster-dupemerge
@@ -86,6 +86,7 @@ my $collapse_access = 0;
 my $collapse_timestamp = 0;
 my $collapse_zero = 0;
 my $skip_compares = 0;
+my $skip_compare_preference = 0;
 my $skip_hashes = 0;
 my $skip_hashes_threshold = 0;
 my $progress = 0;
@@ -184,7 +185,7 @@ while ($#ARGV >= 0) {
 	} elsif ($arg eq '--zeros') {
 		$collapse_zero = 1;
 	} elsif ($arg eq '--trust' || $arg eq '--skip-compare') {
-		$skip_compares = 1;
+		$skip_compares = $skip_compare_preference = 1;
 	} elsif ($arg =~ /^--skip-hash(?:=(\d+)([KkMmGgTt]?))?$/os) {
 		my ($quantity, $unit) = ($1, $2);
 		$unit ||= '_';
@@ -200,7 +201,8 @@ while ($#ARGV >= 0) {
 			t => 1000*1000*1000*1000,
 			T => 1024*1024*1024*1024,
 		);
-		$skip_hashes = $skip_hashes_threshold = $quantity * $scale{$unit};
+		$skip_hashes = 0;
+		$skip_hashes_threshold = $quantity * $scale{$unit};
 	} elsif ($arg eq '--progress') {
 		$progress = 1;
 	} elsif ($arg eq '--verbose') {
@@ -239,10 +241,6 @@ while ($#ARGV >= 0) {
 	}
 }
 
-if ($skip_hashes && $skip_compares) {
-	die "Cannot skip both hashes and compares.\n";
-}
-
 @directories or usage;
 
 if (defined($lock_file) && !$dry_run) {
@@ -357,6 +355,7 @@ hash_file:
 		foreach my $filename (sort keys(%{$inode_to_file_name{$candidate}})) {
 			print STDERR "\t\tDigesting file $filename\n" if $debug;
 			if ((-l $filename) || ! -f _) {
+				print STDERR "\n" if $progress;
 				warn "Bogon file " . tick_quote($filename);
 				$input_bogons++;
 				delete $inode_to_file_name{$candidate}->{$filename};
@@ -542,7 +541,7 @@ candidate_file:
 
 									if ($link_done) {
 										delete $inode_to_file_name{$to_inode}->{$to_file};
-										$inode_to_file_name{$from_inode}->{$to_file} = undef;
+										$inode_to_file_name{$from_inode}->{$to_file} = undef unless ($dry_run);
 										$hash_to_inode{$digest} = [ $from_inode ];
 
 										$hard_links++;
@@ -589,6 +588,9 @@ end_merge:
 	undef %inode_to_file_name;
 }
 
+my $last_time = 0;
+my $last_size = 0;
+
 while (<FIND>) {
 	my ($weak_key, $size, $dev, $ino, $name) = m/^((\d+) \d+ \d+ \d+ -?[\d.]+) (\d+) (\d+) (.+)\0$/so;
 	die "read error: $!\nLast input line was '$_'" unless defined($name);
@@ -597,7 +599,22 @@ while (<FIND>) {
 
 	print STDERR "weak_key=$weak_key inode=$inode name=$name\n" if $debug;
 
-	$skip_hashes = $size >= $skip_hashes_threshold;
+	if ($skip_hashes_threshold && $size >= $skip_hashes_threshold) {
+		$skip_hashes = 1;
+		$skip_compares = 0;
+	} else {
+		$skip_hashes = 0;
+		$skip_compares = $skip_compare_preference;
+	}
+
+	if ($progress) {
+		my $this_time = time();
+		if ($this_time != $last_time && $size != $last_size) {
+			$last_time = $this_time;
+			$last_size = $size;
+			print STDERR "($size)";
+		}
+	}
 
 	$input_links++;
 	merge_files if $weak_key ne $current_key;