#!/bin/bash

# usage: epub2txtMany *.epub

# needs:    https://github.com/kevinboone/epub2txt2 , 
#           unzip, zipinfo (for cover extraction)
# wishes:   catimg (for displaying covers in term)

# does: Converts epub to txt and tries to extract cover image

# tmp dir
cleanup () {
  [ -n "$tmp" ] && rm -fr "$tmp"
}
mktmp () {
  tmp="/tmp/$RANDOM-$$-epub2txtMany"
  trap cleanup EXIT SIGTERM SIGINT 
  mkdir -m 700 "$tmp" || { echo '!! unable to create a tmp dir' >&2; tmp=; exit 1; }
}
mktmp
# end tmp dir

# main
while [ $# -gt 0 ]; do
    file=$(readlink -f "$1")
    baseext=$(basename "${1}")     # file.ext
    base="${baseext%.*}"           # file
    dir="$(dirname "${file}")"
    echo "--- $base ---"
    #set -x
    # convert to text
    epub2txt -a -n "$file" > "${dir}/${base}.txt" || exit 1
    # extract cover image (chatgpt stuff)
    unzip -q "$file" -d "$tmp"
    pushd "$tmp" >/dev/null || exit
    # Find the content.opf file
    OPF_FILE=$(find . -name "*.opf" | head -n 1)
    # Check if the OPF file was found
    if [ -z "$OPF_FILE" ]; then
      echo "No content.opf file found."

    fi
    # Extract cover image reference
    COVER_ID=$(grep -i "cover" "$OPF_FILE" | grep -o 'content="[^"]*' | cut -d '"' -f2)
    COVER_HREF=$(grep -i "$COVER_ID" "$OPF_FILE" | grep -o 'href="[^"]*' | cut -d '"' -f2)
    # Check if the cover href was found
    if [ -z "$COVER_HREF" ]; then
      echo "No cover image reference found in the content.opf file."

    fi
    # Find the cover image file
    COVER_IMAGE=$(dirname "$OPF_FILE")/$COVER_HREF
    # Check if the cover image file exists
    if [ ! -f "$COVER_IMAGE" ]; then
      echo "Cover image file not found."
    else
      # Copy the cover image to the current directory
      cp "$COVER_IMAGE" "${dir}/${base}.jpg" >/dev/null && catimg -H "${dir}/${base}.jpg"
    fi
    # clean the tmp dir of everything
    cleanup && mktmp
    popd >/dev/null || exit
    #set +x

    shift
done