#!/bin/sh

tirto() {
  id="tirto"
  host="https://tirto.id"

  validateFeed "$id" "$host"

  cat $tmp \
    | shup "body" "a" \
    | grep "<a\shref=\"" \
    | sed "s/<a\shref=\"//" \
    | sed "s/\".*$//" \
    | sed "s/\/?.*//" \
    | sed "s/?.*$//" \
    | sed "s/^\//https:\/\/tirto.id\//" \
    | grep "\S" \
    | uniq \
    | while read line ; do
      # Don't @ me
      url=$line
  
      # Check if post/entry/article already exist
      [ ! -z "$(grep "$url" "$dir/feeds/$id.xml")" ] && echo "Entry already exists."

      # Proceed if doesn't exist
      [ -z "$(grep "$url" "$dir/feeds/$id.xml")" ] && curl -s $url > $tmp

      # Check if content and pubdate are empty, meaning not an article
      content=$(cat $tmp | shup "body" "div" "article" "div[content-text-editor]" | grep "\S" | sed "s/^\s*//" | awk '{printf ("%s", $0)}' | sed "s/<script.*>.*<\/script>//")
      pubdate=$(cat $tmp | grep "datePublished" | grep -o "[0-9]*-[0-9]*-[0-9]*\s[0-9]*:[0-9]*:[0-9]*" | sed "s/$/+0700/" | xargs -i date -d {} -R)

      [ ! -z "$content" ] && [ ! -z "$pubdate" ] \
        && title=$(cat $tmp | shup "title" | sed "s/<titl.*\">//" | sed "s/<\/title>//" | sed "s/^\s*//" | grep "\S") \
        && printf "<!-- content -->\n<item>\n<title>%s</title>\n<pubDate>%s</pubDate>\n<guid>%s</guid>\n<link>%s</link>\n<description><![CDATA[%s]]></description>\n</item>\n" "$title" "$pubdate" "$url" "$url" "$content" > $tmp \
        && sed -i "/<!-- content -->/ {
          r $tmp
          d
        }" "$dir/feeds/$id.xml" \
        && echo "Entry inserted."
    done
}