X-Git-Url: http://git.hungrycats.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=faster-dupemerge;h=9920a26bf2bb531bd2a22f2339b83c601aee2cb3;hb=af4afacf460ad2f6c2eafe7632bb55f2060823c6;hp=be6e29bc678bd78aec0d92fef00f030e8d829ef3;hpb=629ed745a2842e156405cbb197ab854037ad7798;p=dupemerge diff --git a/faster-dupemerge b/faster-dupemerge index be6e29b..9920a26 100755 --- a/faster-dupemerge +++ b/faster-dupemerge @@ -87,6 +87,7 @@ my $collapse_timestamp = 0; my $collapse_zero = 0; my $skip_compares = 0; my $skip_hashes = 0; +my $progress = 0; my $verbose = 0; my $debug = 0; my $dry_run = 0; @@ -102,6 +103,7 @@ sub digest { if ($skip_hashes) { return "SKIPPING HASHES"; } else { + print STDERR 'H' if $progress; my $digest = &really_digest($filename); $hash_bytes += -s $filename; $hash_files++; @@ -132,24 +134,33 @@ hard links). --debug show all steps in duplication discovery process (implies --verbose) - --dry-run do not lock files or make changes to filesystem + --dry-run do not lock files or make changes to filesystem --find pass next options (up to --) to find command - --humane human-readable statistics (e.g. 1 048 576) + --humane human-readable statistics (e.g. 1 048 576) --lock FILE exit immediately (status 10) if unable to obtain a flock(LOCK_EX|LOCK_NB) on FILE - --lock-rm remove lock file at exit + --lock-rm remove lock file at exit + + --progress output single-character progress indicators: + C - compare + H - hash + L - link(2) + R - rename(2) + S, s - lstat(2) (see source for details) + U - unlink(2) + . - all inodes with similar attributes done --sort pass next options (up to --) to sort command --timestamps mtime may be different for identical files - --skip-compare skip byte-by-byte file comparisons + --skip-compare skip byte-by-byte file comparisons - --skip-hash skip calculation of hash function on files + --skip-hash skip calculation of hash function on files --trust old name for --skip-compare (trust the hash function) @@ -172,6 +183,8 @@ while ($#ARGV >= 0) { $skip_compares = 1; } elsif ($arg eq '--skip-hash') { $skip_hashes = 1; + } elsif ($arg eq '--progress') { + $progress = 1; } elsif ($arg eq '--verbose') { $verbose = 1; } elsif ($arg eq '--lock-rm') { @@ -269,6 +282,7 @@ sub link_files { my $quoted_from = tick_quote($from); my $quoted_to = tick_quote($to); + print STDERR "\n" if $progress; print STDERR "ln -f $quoted_from $quoted_to\n"; return if $dry_run; @@ -279,10 +293,13 @@ sub link_files { $inode_base =~ s:^.*/::os; my $tmp_to = File::Temp::tempnam($inode_dir, ".$inode_base."); print STDERR "\tlink: $from -> $tmp_to\n" if $debug; + print STDERR 'L' if $progress; link($from, $tmp_to) or die "link: $from -> $tmp_to: $!"; print STDERR "\trename: $tmp_to -> $to\n" if $debug; + print STDERR 'R' if $progress; unless (rename($tmp_to, $to)) { my $saved_bang = $!; + print STDERR 'U' if $progress; unlink($tmp_to) or warn "unlink: $tmp_to: $!"; # Try, possibly in vain, to clean up die "rename: $tmp_to -> $from: $saved_bang"; } @@ -358,6 +375,7 @@ link_start: incumbent_file: foreach my $incumbent_file (@incumbent_names) { + print STDERR 'S' if $progress; my ($incumbent_dev,$incumbent_ino,$incumbent_mode,$incumbent_nlink,$incumbent_uid,$incumbent_gid,$incumbent_rdev,$incumbent_size,$incumbent_atime,$incumbent_mtime,$incumbent_ctime,$incumbent_blksize,$incumbent_blocks) = lstat($incumbent_file); print STDERR "\t\tINCUMBENT dev=$incumbent_dev ino=$incumbent_ino mode=$incumbent_mode nlink=$incumbent_nlink uid=$incumbent_uid gid=$incumbent_gid rdev=$incumbent_rdev size=$incumbent_size atime=$incumbent_atime mtime=$incumbent_mtime ctime=$incumbent_ctime blksize=$incumbent_blksize blocks=$incumbent_blocks _=$incumbent_file\n" if $debug; @@ -378,6 +396,7 @@ incumbent_file: candidate_file: foreach my $candidate_file (@candidate_names) { + print STDERR 's' if $progress; my ($candidate_dev,$candidate_ino,$candidate_mode,$candidate_nlink,$candidate_uid,$candidate_gid,$candidate_rdev,$candidate_size,$candidate_atime,$candidate_mtime,$candidate_ctime,$candidate_blksize,$candidate_blocks) = lstat($candidate_file); print STDERR "\t\t\tCANDIDATE dev=$candidate_dev ino=$candidate_ino mode=$candidate_mode nlink=$candidate_nlink uid=$candidate_uid gid=$candidate_gid rdev=$candidate_rdev size=$candidate_size atime=$candidate_atime mtime=$candidate_mtime ctime=$candidate_ctime blksize=$candidate_blksize blocks=$candidate_blocks _=$candidate_file\n" if $debug; @@ -409,6 +428,7 @@ candidate_file: my $quoted_incumbent_file = tick_quote($incumbent_file); my $quoted_candidate_file = tick_quote($candidate_file); print STDERR "cmp $quoted_incumbent_file $quoted_candidate_file\n" if $debug; + print STDERR 'C' if $progress; if (compare($incumbent_file, $candidate_file)) { $compare_differences++; $identical = 0; @@ -545,6 +565,7 @@ candidate_file: end_merge: + print STDERR '.' if $progress; print STDERR "Merge done.\n" if $debug; undef %inode_to_file_name; }