[BK-2020-03.git] / unitproc / bkfind

#!/bin/bash

# Date: 2020-01-20T17:08Z
#
# Author: Steven Baltakatei Sandoval (baltakatei.com)
#
# License: This bash script, `bkfind`, is licensed under GPLv3 or
# later by Steven Baltakatei Sandoval:
#
#    `bkfind`, a duplicate file finder
#    Copyright (C) 2020  Steven Baltakatei Sandoval (baltakatei.com)
#
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 3 of the License, or
#    any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    A copy of the GNU General Public License may be found at
#    <https://www.gnu.org/licenses/>.
#
# Description: This is a script that searches a specified directory
# for files with a file name containing a specified string. It works
# as follows:
# 
#   - Search specified directory tree for files that have filenames
#     that contain the specified file's filename. List groups of files
#     sharing the same hash first then list files with unique hashes.
#
# Dependencies: find, rhash, uniq, cut, cat, bash. See end of file
#
# Tested on:
#
#   - GNU/Linux Debian 10
#


#==Initialization==
# Use input arguments to define internal script variables.
DIR1="$1"  # Specified directory
FILE1="$2"  # Specified file
DUPLICATES1=""
DUPLICATES2=""
DUPLICATES3=""
UNIQUES1=""
UNIQUES2=""
UNIQUES3=""
RHASH_HASH_TYPE="sha512"
HASH_DISP_LENGTH=16
let HASH_CHAR_LENGTH="512 / 4" # The number of characters returned by the chosen hash function (ex: `rhash --sha512 {}` produces 512/4=128 hexadecimal chars)

# Strip path information from provided file name.
FILEBASE1=$(basename "$FILE1")


#==Main Program==
# Generate list of sha512 hashes and filepaths, save to $HASHLIST1
HASHLIST1="$(find $DIR1 -type f -iname "*$FILEBASE1*" -exec rhash --"$RHASH_HASH_TYPE" {} \;)"

# Specify character position before which characters are dropped from each line with `cut`.
let CUT_POSITION="1 + $HASH_CHAR_LENGTH - $HASH_DISP_LENGTH"

#====Files with duplicate hashes====
# Generate sublist of duplicate entries, save to $DUPLICATES1
DUPLICATES1="$(echo -e "$HASHLIST1" | sort | uniq -D --check-chars=128)"

# Format $DUPLICATES1 for readability by grouping, truncating sha512 hash; save to $DUPLICATES2
DUPLICATES2="$(echo -e "$DUPLICATES1" | uniq --check-chars=128 --group | cut --characters=$CUT_POSITION-)"

#====Files with unique hashes====
# Generate sublist of unique entries, save to $UNIQUES1
UNIQUES1="$(echo -e "$HASHLIST1" | sort | uniq --unique --check-chars=128)"

# Format $UNIQUES1 for readability by truncating sha512 hash; save to $UNIQUES2
UNIQUES2="$(echo -e "$UNIQUES1" | cut --characters=$CUT_POSITION-)"


# List results
echo -e "$DUPLICATES2"
echo -e "$UNIQUES2"

# Dependencies:
#
#     - find (GNU findutils) 4.6.0.225-235f
#       Copyright (C) 2019 Free Software Foundation, Inc.
#       License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>.
#       Written by Eric B. Decker, James Youngman, and Kevin Dalley.
#
#     - RHash v1.3.8
#       License: RHash License <http://rhash.sourceforge.net/license.php>
#
#     - uniq (GNU coreutils) 8.30
#       Copyright (C) 2018 Free Software Foundation, Inc.
#       License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>.
#       This is free software: you are free to change and redistribute it.
#       There is NO WARRANTY, to the extent permitted by law.
#       Written by Richard M. Stallman and David MacKenzie.
#
#     - cut (GNU coreutils) 8.30
#       Copyright (C) 2018 Free Software Foundation, Inc.
#       License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>.
#       This is free software: you are free to change and redistribute it.
#       There is NO WARRANTY, to the extent permitted by law.
#       Written by David M. Ihnat, David MacKenzie, and Jim Meyering.
#
#     - cat (GNU coreutils) 8.30
#       Copyright (C) 2018 Free Software Foundation, Inc.
#       License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>.
#       This is free software: you are free to change and redistribute it.
#       There is NO WARRANTY, to the extent permitted by law.
#       Written by Torbjorn Granlund and Richard M. Stallman.
#
#     - GNU bash, version 5.0.3(1)-release (x86_64-pc-linux-gnu)
#       Copyright (C) 2019 Free Software Foundation, Inc.
#       License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
#       This is free software; you are free to change and redistribute it.
#       There is NO WARRANTY, to the extent permitted by law.