file deduping

file_dedupe.sh

#!/bin/zsh
typeset -A files_by_sum
find $1 -maxdepth 1 | while read file; do
  if [ -f $file ]; then #don't look at directories
    sum=$(cat $file | md5sum | awk '{print $1}')
    current=${files_by_sum[$sum]}
    if [ $current ]; then
      rm $file
      ln -s $(basename $current) $file
    fi
    files_by_sum[$sum]=$file
  fi
done