main

#!/bin/bash
# A script to help download from this site without signing up

_scriptRoot="$(dirname $(realpath $0))"
_projectName="$(basename ${_scriptRoot})"

_projectCommit="$(git -C "${_scriptRoot}" rev-parse --short HEAD 2>/dev/null)"
_projectTag="$(git -C "${_scriptRoot}" describe --tags 2>/dev/null)"
_userAgent="${_projectName}/${_projectTag} ${_projectCommit:-HEAD}"

#  _____                 _   _
# |  ___|   _ _ __   ___| |_(_) ___  _ __  ___
# | |_ | | | | '_ \ / __| __| |/ _ \| '_ \/ __|
# |  _|| |_| | | | | (__| |_| | (_) | | | \__ \
# |_|   \__,_|_| |_|\___|\__|_|\___/|_| |_|___/
#

function checkDep {
  if ! which which >/dev/null 2>&1
  then
    echo "Need which for dep resolution"
    exit 1
  fi
}

function urldecode { # Capable of decoding double-encoded strings.
  iters=0
  result="${*}"
  until [ -n "$result" ] && ! [[ $result =~ %[0-9]{2} ]] || [ $iters -ge 10 ]
  do
    : "${result:-${*}//+/ }"             # Noop with a first argument holding the result of replacing plusses with spaces.
    result="$(echo -e "${_//%/\\x}")"    # Take the last command's last argument and replace %'s with a \x hex escape. Process with echo.
    ((iters++))
  done

  echo $result
}

function printHelp {
  echo    "Usage:"
  echo    "$0"

  echo -e "\t\$url\t(game-soundtracks URL for OST to be downloaded)"
  echo -e "\t-threads 5\t(Optional threaded downloading)"
  echo -e "\t-threads 5\t(Optional threaded downloading)"
}

#  ____
# |  _ \ _ __ ___ _ __
# | |_) | '__/ _ \ '_ \
# |  __/| | |  __/ |_) |
# |_|   |_|  \___| .__/
#                |_|

checkDep curl wget

while [ $# -gt 0 ]
do
  case "$(tr '[:upper:]' '[:lower:]'<<<$1)" in
    -threads|-parallel)
      if ! which parallel >/dev/null 2>&1
      then
        echo "Can't find parallel in path"
        exit 1
      fi

      if ! [[ $2 =~ [0-9]+ ]]
      then
        echo "thread count must be a number."
        exit 1
      fi

      threads="$2"
      shift
    ;;
  -url)
      fullUrl="$2"
      shift
    ;;
    *)
      if [ -z "$fullUrl" ] && [[ $1 =~ http.*game-soundtracks ]]
      then
        echo "URL: $1"
        fullUrl="$1"
      else
        printHelp
      fi
        ;;
  esac
  shift
done

if [ -z "${fullUrl}" ]
then
  echo "Can't run without a /game-soundtracks/ album OST URL."
  exit 1
fi

ostPath="${fullUrl##*/}"
baseUrl=$(awk -F'/' '{print $1 "/" $2 "/" $3}' <<< "${fullUrl}")

# Cache as much as we can to reduce repeated visits.
cachePath="/tmp/${ostPath}"
cachePathIndex="/tmp/${fullUrl##*/}/index.html"
mkdir -p "${cachePath}"

if [ ! -f "${cachePathIndex}" ]
then
  pageHtml="$(curl -A "${_userAgent}" -Ss "${fullUrl}")"
  if [ -n "${pageHtml}" ]
  then
    echo "${pageHtml}" > "${cachePathIndex}"
  else
    echo "Failed to load the OST page for "
    exit 1
  fi
else
  pageHtml="$(cat "${cachePathIndex}")"
fi


ostTitle="$(grep -Po '(?<=<h2>).*(?=\<a)' <<< "${pageHtml}")"
ostDir="${_scriptRoot}/${ostTitle}"
trackList="$(awk -F'"' '/href="\/game-soundtracks.*mp3"/ {for(i=1;i<=NF;i++) if ($i ~ "game-soundtracks" && !seen[$i]) {print $i; ++seen[$i]}}' <<< "${pageHtml}")"


for track in ${trackList}
do
  trackPath="${track##*/}"
  trackName="$(urldecode "${track##*/}")"
  trackUrlFull="${baseUrl}${track}"

  # Cache here too
  cachePathTrack="${cachePath}/${trackPath}.html"
  if [ ! -f "${cachePathTrack}" ]
  then
    echo "Fetching track page: ${trackName}..."
    pageHtmlTrack="$(curl -A "${_userAgent}" -Ss "${baseUrl}${track}")"
    if [ -n "${pageHtmlTrack}" ]
    then
      echo "${pageHtmlTrack}" > "${cachePathTrack}"
    else
      echo "Failed to get the page for $(urldecode "${trackName}")"
    fi
  else
    pageHtmlTrack="$(cat "${cachePathTrack}")"
  fi


  downloadUrls+=("$(awk -F'"' '/.flac/ {for(i=1;i<=NF;i++) if ($i ~ ".flac") {print $i}}' <<< "${pageHtmlTrack}")")
done


if [ ${#downloadUrls[@]} -gt 0 ]
then
  mkdir -pv "${ostDir}"
  if [ -z "$threads" ]
  then
    echo "No threading specified, downloading sequentially with wget to ${ostTitle} in the script root"
    # --timestamping over --no-clobber allows resumption of partially finished files -
    # though this doesn't seem to be working for khinsider.com. Content-Length and Timestamps may be disabled on the webserver.
    wget --continue --directory-prefix="${ostDir}" --quiet --show-progress --timestamping ${downloadUrls[@]} 
  else
    for URL in ${downloadUrls[@]}
    do
      echo "echo \"Downloading $(urldecode "${URL##*/}")...\" >&2 && wget --continue --directory-prefix=\"${ostDir}\" --quiet --timestamping \"${URL}\""
    done | parallel -j${threads}
  fi
else
  echo "Failed to extract download URLs. Exiting."
  exit 1
fi