Extract *.m3u8 playlist urls from link to anime page on online.anidub.com

README.md

Скрипт позволяет создать плейлист в формате \*.m3u для просмотра аниме с https://online.anidub.com/

Использование

Для создания плейлиста нужно выполнить скрипт с двумя параметрами: ссылку на страницу аниме (например это) и путь куда сохранять плейлист (например /tmp/pl.m3u). Команда будет иметь вид

$ python3 extract.py --output pl.m3u \
    --url "https://online.anidub.com/anime/10415-galakticheskie-zheleznye-dorogi-ginga-tetsudou-monogatari-01-iz-26.html"

Требования

• python3
• установленные зависимости из requirements.txt
• любой плейер способный проигрывать плейлисты в форматах:
  • *.m3u - используется для создания плейлиста с сериями
  • *.m3u8 - используется для https://video.sibnet.ru
  • *.mp4 - используется для https://www.stormo.tv

extract.py

#!/usr/bin/env python3
import re

import requests
import requests_cache
import click

from lxml import etree
from cssselect import GenericTranslator


requests_cache.install_cache(backend='memory')


class Video(object):
    def __init__(self, number, provider_url):
        self.number = number
        self.provider_url = provider_url
        self.title = None
        self.duration = None
        self.url = None


class Player(object):
    def __init__(self, anidub_url):
        self.anidub_url = anidub_url
        self.playlist = []

    def _extract_videos(self):
        raise NotImplemented

    def _extract_video_url(self, url):
        raise NotImplemented

    def _extract_video_title(self, url):
        raise NotImplemented

    def _extract_video_duration(self, url):
        raise NotImplemented

    @property
    def name(self):
        raise NotImplemented

    @property
    def REG(self):
        raise NotImplemented

    def count_series(self):
        if not self.playlist:
            self._extract_videos()
        return len(self.playlist)

    def process_playlist(self):
        with click.progressbar(self.playlist, label='Loading: ') as playlist:
            for video in playlist:
                video.url = self._extract_video_url(video.provider_url)
                video.title = self._extract_video_title(video.provider_url)
                video.duration = self._extract_video_duration(video.provider_url)


class SibnetPlayer(Player):
    name = 'Sibnet'
    REG = re.compile('"/v/[a-z0-9]*/[0-9]*[.]m3u8"')

    def _extract_videos(self):
        html = requests.get(self.anidub_url).text
        selector = GenericTranslator().css_to_xpath('#sel > option')
        document = etree.HTML(html)
        for el in document.xpath(selector):
            url, number = el.get('value').split('|')
            self.playlist.append(Video(number, url))

    def _extract_video_url(self, url):
        html = requests.get(url).text
        try:
            link = self.REG.search(html).group(0)[1:-1]
            return 'https://video.sibnet.ru{}'.format(link)
        except e:
            return ''

    def _extract_video_title(self, url):
        html = requests.get(url).text
        document = etree.HTML(html)
        return document.find('.//meta[@property="og:title"]').get('content')

    def _extract_video_duration(self, url):
        html = requests.get(url).text
        document = etree.HTML(html)
        return document.find('.//meta[@property="og:duration"]').get('content')


class StormoPlayer(Player):
    name = 'Stormo'
    REG = re.compile('https://www\.stormo\.tv/get_file/[0-9]*/[a-z0-9]*/'
                     '[0-9]*/[0-9]*/[0-9]*[.]mp4/')

    def _extract_videos(self):
        html = requests.get(self.anidub_url).text
        selector = GenericTranslator().css_to_xpath('#sel3 > option')
        document = etree.HTML(html)
        for el in document.xpath(selector):
            url, number = el.get('value').split('|')
            self.playlist.append(Video(number, url))

    def _extract_video_url(self, url):
        html = requests.get(url).text
        try:
            return self.REG.search(html).group(0)
        except e:
            return ''

    def _extract_video_title(self, url):
        html = requests.get(url).text
        document = etree.HTML(html)
        return document.find('.//title').text

    def _extract_video_duration(self, url):
        return -1


@click.command()
@click.option('-u', '--url', required=True, help='anidub url')
@click.option('-o', '--output', type=click.File('w'), required=True,
              help='where to put VLC playlist (*.m3u)')
def extract(url=None, output=None):
    players = [SibnetPlayer(url), StormoPlayer(url)]
    series_count = max(player.count_series() for player in players)
    players = [p for p in players if p.count_series() == series_count]
    output.write('#EXTM3U\n\n')
    for player in players[:1]:
        click.secho('--= {} =--'.format(player.name), fg='green', bold=True)
        output.write('# --= {} =--\n\n'.format(player.name))
        player.process_playlist()
        click.secho('#\tURL', fg='blue')
        for video in player.playlist:
            click.echo('{}\t{}'.format(video.number, video.url))
            output.write('#EXTINF:{}, {}\n'.format(video.duration, video.title))
            output.write('{}\n'.format(video.url))


if __name__ == "__main__":
    extract()

requirements.txt

Click==7.0
cssselect==1.0.3
lxml==4.2.5
requests==2.20.0
requests-cache==0.4.13