lc_ext.sh

#!/bin/bash

# The purpose of this script is to find all (PDF) files in the directory
# specified and turn the file extension to lowercase. If another file
# already has the new name, check its MD5 hash to see if it's an
# identical file, and if so delete it. Also delete files with any name
# that have identical MD5 hashes to other files in the directory.

set -eo pipefail

usage () {
    printf '\n%s\n\n' "Usage: $(basename "$0") [dir]"
    exit
}

if [[ -z $1 || ! -d $1 ]]; then
    usage
fi

declare -A regex md5s

regex[fn]='^(.*)\.([^.]*)$'

dir=$(readlink -f "$1")

mapfile -t files < <(find "$dir" -type f -iname "*.pdf")

for (( i = 0; i < ${#files[@]}; i++ )); do
    f="${files[${i}]}"
    f_bn=$(basename "$f")

    if [[ ! $f =~ ${regex[fn]} ]]; then
        continue
    fi

    no_ext="${BASH_REMATCH[1]}"
    ext="${BASH_REMATCH[2]}"
    lc_ext="${ext,,}"

    new_f="${no_ext}.${lc_ext}"
    new_f_bn=$(basename "$new_f")

    f_md5=$(md5sum -b "$f")
    f_md5="${f_md5%% *}"

    if [[ -z ${md5s[${f_md5}]} ]]; then
        md5s["${f_md5}"]=1
    else
        printf '%s\n' "rm: ${f}"
        rm "$f"

        continue
    fi

    if [[ $f != "$new_f" ]]; then
        if [[ -f $new_f ]]; then
            new_f_md5=$(md5sum -b "$new_f")
            new_f_md5="${new_f_md5%% *}"

            if [[ $f_md5 != "$new_f_md5" ]]; then
                continue
            fi
        fi

        printf '%s\n' "mv: ${new_f}"
        mv "$f" "$new_f"
    fi
done