DELETE all those duplicate files but one based on md5 hash comparision in the current directory tree

find . -type f -print0|xargs -0 md5sum|sort|perl -ne 'chomp;$ph=$h;($h,$f)=split(/\s+/,$_,2);print "$f"."\x00" if ($h eq $ph)'|xargs -0 rm -v --
This one-liner will the *delete* without any further confirmation all 100% duplicates but one based on their md5 hash in the current directory tree (i.e including files in its subdirectories). Good for cleaning up collections of mp3 files or pictures of your dog|cat|kids|wife being present in gazillion incarnations on hd. md5sum can be substituted with sha1sum without problems. The actual filename is not taken into account-just the hash is used. Whatever sort thinks is the first filename is kept. It is assumed that the filename does not contain 0x00. As per the good suggestion in the first comment, this one does a hard link instead: find . -xdev -type f -print0 | xargs -0 md5sum | sort | perl -ne 'chomp; $ph=$h; ($h,$f)=split(/\s+/,$_,2); if ($h ne $ph) { $k = $f; } else { unlink($f); link($k, $f); }'
Sample Output
removed `./duplicate0.mp3'
removed `./1/duplicate1.mp3'
removed `./2/duplicate2.mp3'

19
2009-06-07 03:14:06

These Might Interest You

  • If you have the fdupes command, you'll save a lot of typing. It can do recursive searches (-r,-R) and it allows you to interactively select which of the duplicate files found you wish to keep or delete.


    22
    fdupes -r .
    Vilemirth · 2011-02-19 17:02:30 0
  • Calculates md5 sum of files. sort (required for uniq to work). uniq based on only the hash. use cut ro remove the hash from the result.


    18
    find -type f -exec md5sum '{}' ';' | sort | uniq --all-repeated=separate -w 33 | cut -c 35-
    infinull · 2009-08-04 07:05:12 1
  • Improvement of the command "Find Duplicate Files (based on size first, then MD5 hash)" when searching for duplicate files in a directory containing a subversion working copy. This way the (multiple dupicates) in the meta-information directories are ignored. Can easily be adopted for other VCS as well. For CVS i.e. change ".svn" into ".csv": find -type d -name ".csv" -prune -o -not -empty -type f -printf "%s\n" | sort -rn | uniq -d | xargs -I{} -n1 find -type d -name ".csv" -prune -o -type f -size {}c -print0 | xargs -0 md5sum | sort | uniq -w32 --all-repeated=separate Show Sample Output


    2
    find -type d -name ".svn" -prune -o -not -empty -type f -printf "%s\n" | sort -rn | uniq -d | xargs -I{} -n1 find -type d -name ".svn" -prune -o -type f -size {}c -print0 | xargs -0 md5sum | sort | uniq -w32 --all-repeated=separate
    2chg · 2010-01-28 09:45:29 0
  • This works on Mac OS X using the `md5` command instead of `md5sum`, which works similarly, but has a different output format. Note that this only prints the name of the duplicates, not the original file. This is handy because you can add `| xargs rm` to the end of the command to delete all the duplicates while leaving the original.


    2
    find . -type f -exec md5 '{}' ';' | sort | uniq -f 3 -d | sed -e "s/.*(\(.*\)).*/\1/"
    noahspurrier · 2012-01-14 08:54:12 4

What Others Think

Very cute. Alternatively, if you want to keep the appearance but still save space, you can use the "hardlink" command. This searches recursively for duplicates and replaces them all with hard links.
flatcap · 471 weeks and 5 days ago
fdupes its just simple! If you just want one of the files type: fdupes -fN
debuti · 471 weeks and 5 days ago

What do you think?

Any thoughts on this command? Does it work on your machine? Can you do the same thing with only 14 characters?

You must be signed in to comment.

What's this?

commandlinefu.com is the place to record those command-line gems that you return to again and again. That way others can gain from your CLI wisdom and you from theirs too. All commands can be commented on, discussed and voted up or down.

Share Your Commands



Stay in the loop…

Follow the Tweets.

Every new command is wrapped in a tweet and posted to Twitter. Following the stream is a great way of staying abreast of the latest commands. For the more discerning, there are Twitter accounts for commands that get a minimum of 3 and 10 votes - that way only the great commands get tweeted.

» http://twitter.com/commandlinefu
» http://twitter.com/commandlinefu3
» http://twitter.com/commandlinefu10

Subscribe to the feeds.

Use your favourite RSS aggregator to stay in touch with the latest commands. There are feeds mirroring the 3 Twitter streams as well as for virtually every other subset (users, tags, functions,…):

Subscribe to the feed for: