#!/bin/bash


# trap ctrl-c and call ctrl_c()

function ctrl_c() {
        echo "ABORTING DUPE REVIEW"
        cat "$exceptionsfile" | (read -r; printf "%s\n" "$REPLY"; sort -u) > "${sorted_exceptionsfile}"
        mv -f "${sorted_exceptionsfile}" "$exceptionsfile"
        [[ -f "${optimized_exceptionsfile}" ]] && rm -f "${optimized_exceptionsfile}"
        exit -1
}

trap ctrl_c SIGINT SIGTERM


IE="$(pwd)"
exceptionsfile="${IE}/exceptions.txt"
optimized_exceptionsfile=""
sorted_exceptionsfile="$( mktemp /tmp/sorted_exceptions_XXXXXXXX.txt )"

optimization_period="7"
optimized_curr="$(( $(date +"%s") / 86400 ))"
skip_flag=false
DEFAULT_SIMILARITY="Small"

while getopts "hy" opt; do
    case $opt in
    h) echo -e "usage: $0 [-h help] [-y automatically add exceptions]";  exit ;;
    y) skip_flag=true ;;
    ?) echo "error: option -$OPTARG is not implemented"; exit ;;
    esac
done

( [ -e "$exceptionsfile" ] || (touch "$exceptionsfile" && echo "##### LAST TIMESTAMP OF EXCEPTION FILE OPTIMIZATION (IN DAYS SINCE UNIX EPOCH): ${optimized_curr} #####" > "$exceptionsfile") ) && [ ! -w "$exceptionsfile" ] && echo "ERROR: Cannot write to $exceptionsfile" && exit 1

optimized_last="$(cat "$exceptionsfile" | head -n 1 | grep -E "##### LAST TIMESTAMP OF EXCEPTION FILE.*#####" | grep -oE "[0-9]+")"
[ -z "$optimized_last" ]  && echo "ERROR: Cannot parse last optimization date of $exceptionsfile" && exit 1

[[ -n $(find "$(pwd)" -regextype awk -iregex ".*[\[\]\(\)\{\}\*\+\?]+.*") ]] && echo -e "REGEX SYMBOLS FOUND IN FILENAMES, IE IN SHAMBLES. ABORTING.\n" && exit -1

function optimize_exceptions {

    optimized_exceptionsfile="$( mktemp /tmp/optimized_exceptions_XXXXXXXX.txt )"

    optimized_date_rel=$(( optimized_curr - optimized_last ))
    [[ $optimized_date_rel -ge $optimization_period ]] || return 1

    echo "##### LAST TIMESTAMP OF EXCEPTION FILE OPTIMIZATION (IN DAYS SINCE UNIX EPOCH): ${optimized_curr} #####" > "${optimized_exceptionsfile}"

    while read line; do
        lineexpr=$(echo -e "$line" | sed -r "s| \/|\.*\/|g; s|^|\.*|g; s|$|\.*|g")
        [[ -n $( grep -E "##### LAST TIMESTAMP OF EXCEPTION FILE.*#####" <<< "$line" ) || $(grep -a "$lineexpr" exceptions.txt | wc -l) -ge 2 ]] && continue || echo "$line" >> "${optimized_exceptionsfile}"
    done < "$exceptionsfile"

    exceptions_lines=$(wc -l < "$exceptionsfile")
    optimized_exceptions_lines=$(wc -l < "${optimized_exceptionsfile}")
    optimization_diff=$(( exceptions_lines - optimized_exceptions_lines ))
    [[ $optimization_diff -gt 0 ]] || return 1

    while true; do
        read -p "Exceptions file (${exceptionsfile}) has been optimized (${optimization_diff} new optimizations, ${optimized_date_rel} days since last optimization). Would you like to overwrite the existing exceptions file (y/n)? " yn

        case $( tr '[A-Z]' '[a-z]' <<< "$yn" ) in
            y|yes) rm "$exceptionsfile" && cp "${optimized_exceptionsfile}" "$exceptionsfile" && echo "Exceptions file updated." ; break;;
            n|no) echo "Update aborted." ; break;;
            *) echo "Invalid response." ;;
        esac
    done

    rm -f "${optimized_exceptionsfile}"
}

$skip_flag || optimize_exceptions

re='^Minimal|VerySmall|Small|Medium|High|VeryHigh$'

declare SIMILARITY

if $skip_flag ; then
    SIMILARITY="$DEFAULT_SIMILARITY"
else
    while read -p "Enter similarity preset [Default: Small; Allowed: Minimal, VerySmall, Small, Medium, High, VeryHigh]: " SIMILARITY; do

        SIMILARITY=${SIMILARITY:-"$DEFAULT_SIMILARITY"}

        if ! [[ $SIMILARITY =~ $re ]]; then
            echo "error: Invalid input" >&2; continue
        fi

        break
    done
fi


mapfile -t matches < <( czkawka-cli image -d "$IE" -s "$SIMILARITY" | sed -r "s| \- [0-9]*x[0-9]* \- [0-9]*(\.[0-9]*)? [KMGTP]iB \- [a-zA-Z ]*$||g ; s|[0-9]+ images which have similar friends||g ; s|\"||g ; s|^Found ||g ; s|.*-------MESSAGES-------.*||g ; s|.*Properly loaded.*||g ; s|.*Properly saved to file.*||g " \
| head -n -5 | awk -v RS=  '{$1=$1; t=!/\//; if(NR>1 && t) print ""; print; if(t) print ""}' )


count=0

for match in "${matches[@]}"
do
    matches[$count]=$( echo "$match" | sed -r "s|\ \/|\n\/|g" | sort | tr '\n' ' ' | sed -r "s| $|\n|" )
    ((count++))
done


printf '%s\n' "${matches[@]}" | while IFS= read -r line
do
    lineexpr=$(echo -e "$line" | sed -r "s| \/|\.*\/|g; s|^|\.*|g; s|$|\.*|g")
    [[ -n $(grep -a -E "$lineexpr" "$exceptionsfile") ]] && continue || echo "$lineexpr"
    if $skip_flag ; then
        echo "$line" >> "$exceptionsfile"
    else
        echo -e "$line" | sed -r "s|\ \/home\/|\n\/home\/|g" | /bin/feh --info "printf '%S     %wx%h'" --zoom max --scale-down -g 1280x720 -B black -d --action1 "gio trash %F" --action2 "cat %L | sed -r 's|^\.|$(pwd)|g' | sort | tr '\n' ' ' | paste -s -d ' ' | sed 's| $||g' >> \"$exceptionsfile\"" --action3 "$HOME/stuf/scripts/scriptlets/feh_action3.sh %F %L" -f -
    fi
done

cat "$exceptionsfile" | (read -r; printf "%s\n" "$REPLY"; sort -u) > "${sorted_exceptionsfile}"
mv -f "${sorted_exceptionsfile}" "$exceptionsfile"

# EVERYTHING IS HOW IT'S SUPPOSED TO BE, THERE CAN BE NO IMPROVEMENTS, NOTHING ELSE WORKED
# | sed -e "s|\(\/.*\)|\"\1\"|g"
# ; s|\ |\\\ |g

#THIS IS BEST PRACTICE FOR ENVIRONMENT VARIABLES, CONSIDER CORRECTING ~/stuf STUFF
# -t 95% WORKS FOR ELIMINATING THE CHAFF, BUT IT MIGHT MISS SOME

#feh . --action1 "cat %L | sed -r 's|^\.|$(pwd)|g' | sort | tr '\n' ' ' "
#czkawka_cli image -d "/home/jay/Tempsktop" -s "High" | sed -r "s| \- [0-9]*x[0-9]* \- [0-9]*\.[0-9]* [KMGTP]iB \- [a-zA-Z ]*$||g ; s|Found [0-9]+ images which have similar friends||g " | head -n -5 | awk -v RS=  '{$1=$1; t=!/\//; if(NR>1 && t) print ""; print; if(t) print ""}' | xargs -n 1| sort | xargs


# IF IT WORKS WITH QUOTES, ADD QUOTES