Bash script to prune duplicate files

dedupe.sh

#!/bin/bash

# Delete files from src_folder that are present in target_folder
# USAGE: dedupe.sh <src_folder> <target_folder>

# Uncomment for Linux
SHA_CMD=sha1sum
# Uncomment for Mac
# SHA_CMD=shasum

# First build a static list of files in the target directory for fast searching
TMP_LIST="$(echo "$2" | $SHA_CMD | cut -f 1 -d \ ).tmp.list"
find "$2" -type f > $TMP_LIST

# Iterate through files in the source directory
find "$1" -type f | while read i; do

  # Get a SHA for the source file
  SRC_SUM="$($SHA_CMD "$i" | cut -f 1 -d \ )"

  # Check if there is a file in the target directory with the same name
  SRC_NAME="$(basename "$i")"
  TGT_PATH="$(grep "$SRC_NAME"'$' $TMP_LIST | head -1)"

  # Get a SHA for the target file (or empty string if there isn't one)
  TGT_SUM="$($SHA_CMD "$TGT_PATH" | cut -f 1 -d \ )"

  # Compare source & target SHAs
  if [[ $SRC_SUM == $TGT_SUM ]]; then
    echo DELETING DUPLICATE: $i
    rm "$i"
  else
    echo SKIPPING UNIQUE: $i
  fi
done

rm $TMP_LIST