Option to generate list of unique files

[SeeSpotRun]

In response to http://askubuntu.com/questions/686763/how-do-i-view-non-matching-files-only-in-fdupes; working first attempt at https://github.com/SeeSpotRun/rmlint/tree/feature/uniques. Usage: rmlint -o uniques ...

I think it could be quite handy, eg for recovering just the unique files from an old backup:
$ rmlint -o uniques -k ~ // /mnt/backup

Discussion welcome, including best way to copy the found files (short shell script I guess).

[SeeSpotRun]

Here's a simulated usecase using rsync to recover unique files from an old backup:

$ # make some files
$ mkdir originals && cd originals
$ for i in {1..200}; do dd if=/dev/urandom bs=1 count=${#i} of="file $i"; done
...
$ # make a backup
$ cd .. && cp -a originals backup
...
$ # make some changes (add some, change some, delete some)
$ cd originals
$ for i in {200..300}; do dd if=/dev/urandom bs=1 count=${#i} of="file $i"; done
$ for i in {99..110}; do dd if=/dev/urandom bs=1 count=${#i} of="file $i"; done
$ for i in {9..20}; do rm "file $i"; done
$ cd ..
...
$ # use rmlint to find unique files in the backup
$ rmlint -o uniques -k originals/ // backup/ | tee uniques.txt
...
$ # use rsync to archive the unique files
$ mkdir recovered
$ rsync -av --files-from=uniques.txt / recovered/ 
...
$ # did it work?
$ tree recovered/
...
$ # great, now we can wipe the backup drive
$ rm -rf backup

[sahib]

I merged this feature to develop, closing therefore.