dir_cksum.sh

#!/bin/sh

#
# This script may be useful when you want to ensure that
# some file tree that your copied from one machine to another one
# is not corrupted.
# In this program we calculate total size of all files in some directory
# (recursively) using du (-b means apparent size in bytes; st_size field
# struct stat used).
# And as another control method cksum program used, and with help of awk,
# we calculate checksum of all crc checksums that we got from cksum.
# And... also we count total size of files in bytes using data from cksum
# (just for sure). Of course, this number should be equal to the first one.
# Example:
# $ ./calc.sh /mnt/alex/Files
# 59994260119
# ed623
# 59994260119
#

# Implementation of hash() function in gawk was borrowed from
# https://riptutorial.com/awk/example/12547/computing-a-hash-of-a-string

# We use busybox here, because I've used this script on my NAS
# but, of course, you can simply set busybox_path="" and all should works just fine
# in standart linux/*bsd environment
busybox_path="busybox"
tmpfile=$(mktemp)

cat > $tmpfile <<'EOF'

BEGIN {
  for(n=0;n<256;n++) {
    ord[sprintf("%c",n)] = n
  }

  res = ""
  total = 0;
}

function hash(text, _prime, _modulo, _ax, _chars, _i) {
  _prime = 104729;
  _modulo = 1048576;
  _ax = 0;
  split(text, _chars, "");
  for (_i=1; _i <= length(text); _i++) {
    _ax = (_ax * _prime + ord[_chars[_i]]) % _modulo;
  };
  return sprintf("%05x", _ax)
}

{
#  print $1;
  tmp = "$1";
  res = hash(sprintf("%s%s", tmp, res));
  total += $2;
}

END {
  print res;
  print total;
}

EOF

$busybox_path find "$@" -type f -exec $busybox_path du -sb {} \; |\
$busybox_path awk 'BEGIN {total = 0;} { total += $1; } END { print total; }'

$busybox_path find "$@" -type f -exec $busybox_path cksum {} \; |\
$busybox_path awk -f $tmpfile

rm $tmpfile