yt-dlp/youtube_dl/extractor/zattoo.py

# coding: utf-8
from __future__ import unicode_literals

import re
from uuid import uuid4

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_str,
)
from ..utils import (
    ExtractorError,
    int_or_none,
    try_get,
    url_or_none,
    urlencode_postdata,
)


class ZattooBaseIE(InfoExtractor):
    _NETRC_MACHINE = 'zattoo'
    _HOST_URL = 'https://zattoo.com'

    _power_guide_hash = None

    def _login(self):
        username, password = self._get_login_info()
        if not username or not password:
            self.raise_login_required(
                'A valid %s account is needed to access this media.'
                % self._NETRC_MACHINE)

        try:
            data = self._download_json(
                '%s/zapi/v2/account/login' % self._HOST_URL, None, 'Logging in',
                data=urlencode_postdata({
                    'login': username,
                    'password': password,
                    'remember': 'true',
                }), headers={
                    'Referer': '%s/login' % self._HOST_URL,
                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
                })
        except ExtractorError as e:
            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
                raise ExtractorError(
                    'Unable to login: incorrect username and/or password',
                    expected=True)
            raise

        self._power_guide_hash = data['session']['power_guide_hash']

    def _real_initialize(self):
        webpage = self._download_webpage(
            self._HOST_URL, None, 'Downloading app token')
        app_token = self._html_search_regex(
            r'appToken\s*=\s*(["\'])(?P<token>(?:(?!\1).)+?)\1',
            webpage, 'app token', group='token')
        app_version = self._html_search_regex(
            r'<!--\w+-(.+?)-', webpage, 'app version', default='2.8.2')

        # Will setup appropriate cookies
        self._request_webpage(
            '%s/zapi/v2/session/hello' % self._HOST_URL, None,
            'Opening session', data=urlencode_postdata({
                'client_app_token': app_token,
                'uuid': compat_str(uuid4()),
                'lang': 'en',
                'app_version': app_version,
                'format': 'json',
            }))

        self._login()

    def _extract_cid(self, video_id, channel_name):
        channel_groups = self._download_json(
            '%s/zapi/v2/cached/channels/%s' % (self._HOST_URL,
                                               self._power_guide_hash),
            video_id, 'Downloading channel list',
            query={'details': False})['channel_groups']
        channel_list = []
        for chgrp in channel_groups:
            channel_list.extend(chgrp['channels'])
        try:
            return next(
                chan['cid'] for chan in channel_list
                if chan.get('cid') and (
                    chan.get('display_alias') == channel_name or
                    chan.get('cid') == channel_name))
        except StopIteration:
            raise ExtractorError('Could not extract channel id')

    def _extract_cid_and_video_info(self, video_id):
        data = self._download_json(
            '%s/zapi/program/details' % self._HOST_URL,
            video_id,
            'Downloading video information',
            query={
                'program_id': video_id,
                'complete': True
            })

        p = data['program']
        cid = p['cid']

        info_dict = {
            'id': video_id,
            'title': p.get('title') or p['episode_title'],
            'description': p.get('description'),
            'thumbnail': p.get('image_url'),
            'creator': p.get('channel_name'),
            'episode': p.get('episode_title'),
            'episode_number': int_or_none(p.get('episode_number')),
            'season_number': int_or_none(p.get('season_number')),
            'release_year': int_or_none(p.get('year')),
            'categories': try_get(p, lambda x: x['categories'], list),
        }

        return cid, info_dict

    def _extract_formats(self, cid, video_id, record_id=None, is_live=False):
        postdata_common = {
            'https_watch_urls': True,
        }

        if is_live:
            postdata_common.update({'timeshift': 10800})
            url = '%s/zapi/watch/live/%s' % (self._HOST_URL, cid)
        elif record_id:
            url = '%s/zapi/watch/recording/%s' % (self._HOST_URL, record_id)
        else:
            url = '%s/zapi/watch/recall/%s/%s' % (self._HOST_URL, cid, video_id)

        formats = []
        for stream_type in ('dash', 'hls', 'hls5', 'hds'):
            postdata = postdata_common.copy()
            postdata['stream_type'] = stream_type

            data = self._download_json(
                url, video_id, 'Downloading %s formats' % stream_type.upper(),
                data=urlencode_postdata(postdata), fatal=False)
            if not data:
                continue

            watch_urls = try_get(
                data, lambda x: x['stream']['watch_urls'], list)
            if not watch_urls:
                continue

            for watch in watch_urls:
                if not isinstance(watch, dict):
                    continue
                watch_url = url_or_none(watch.get('url'))
                if not watch_url:
                    continue
                format_id_list = [stream_type]
                maxrate = watch.get('maxrate')
                if maxrate:
                    format_id_list.append(compat_str(maxrate))
                audio_channel = watch.get('audio_channel')
                if audio_channel:
                    format_id_list.append(compat_str(audio_channel))
                preference = 1 if audio_channel == 'A' else None
                format_id = '-'.join(format_id_list)
                if stream_type in ('dash', 'dash_widevine', 'dash_playready'):
                    this_formats = self._extract_mpd_formats(
                        watch_url, video_id, mpd_id=format_id, fatal=False)
                elif stream_type in ('hls', 'hls5', 'hls5_fairplay'):
                    this_formats = self._extract_m3u8_formats(
                        watch_url, video_id, 'mp4',
                        entry_protocol='m3u8_native', m3u8_id=format_id,
                        fatal=False)
                elif stream_type == 'hds':
                    this_formats = self._extract_f4m_formats(
                        watch_url, video_id, f4m_id=format_id, fatal=False)
                elif stream_type == 'smooth_playready':
                    this_formats = self._extract_ism_formats(
                        watch_url, video_id, ism_id=format_id, fatal=False)
                else:
                    assert False
                for this_format in this_formats:
                    this_format['preference'] = preference
                formats.extend(this_formats)
        self._sort_formats(formats)
        return formats

    def _extract_video(self, channel_name, video_id, record_id=None, is_live=False):
        if is_live:
            cid = self._extract_cid(video_id, channel_name)
            info_dict = {
                'id': channel_name,
                'title': self._live_title(channel_name),
                'is_live': True,
            }
        else:
            cid, info_dict = self._extract_cid_and_video_info(video_id)
        formats = self._extract_formats(
            cid, video_id, record_id=record_id, is_live=is_live)
        info_dict['formats'] = formats
        return info_dict


class QuicklineBaseIE(ZattooBaseIE):
    _NETRC_MACHINE = 'quickline'
    _HOST_URL = 'https://mobiltv.quickline.com'


class QuicklineIE(QuicklineBaseIE):
    _VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<channel>[^/]+)/(?P<id>[0-9]+)'

    _TEST = {
        'url': 'https://mobiltv.quickline.com/watch/prosieben/130671867-maze-runner-die-auserwaehlten-in-der-brandwueste',
        'only_matching': True,
    }

    def _real_extract(self, url):
        channel_name, video_id = re.match(self._VALID_URL, url).groups()
        return self._extract_video(channel_name, video_id)


class QuicklineLiveIE(QuicklineBaseIE):
    _VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<id>[^/]+)'

    _TEST = {
        'url': 'https://mobiltv.quickline.com/watch/srf1',
        'only_matching': True,
    }

    @classmethod
    def suitable(cls, url):
        return False if QuicklineIE.suitable(url) else super(QuicklineLiveIE, cls).suitable(url)

    def _real_extract(self, url):
        channel_name = video_id = self._match_id(url)
        return self._extract_video(channel_name, video_id, is_live=True)


class ZattooIE(ZattooBaseIE):
    _VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<channel>[^/]+?)/(?P<id>[0-9]+)[^/]+(?:/(?P<recid>[0-9]+))?'

    # Since regular videos are only available for 7 days and recorded videos
    # are only available for a specific user, we cannot have detailed tests.
    _TESTS = [{
        'url': 'https://zattoo.com/watch/prosieben/130671867-maze-runner-die-auserwaehlten-in-der-brandwueste',
        'only_matching': True,
    }, {
        'url': 'https://zattoo.com/watch/srf_zwei/132905652-eishockey-spengler-cup/102791477/1512211800000/1514433500000/92000',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        channel_name, video_id, record_id = re.match(self._VALID_URL, url).groups()
        return self._extract_video(channel_name, video_id, record_id)


class ZattooLiveIE(ZattooBaseIE):
    _VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<id>[^/]+)'

    _TEST = {
        'url': 'https://zattoo.com/watch/srf1',
        'only_matching': True,
    }

    @classmethod
    def suitable(cls, url):
        return False if ZattooIE.suitable(url) else super(ZattooLiveIE, cls).suitable(url)

    def _real_extract(self, url):
        channel_name = video_id = self._match_id(url)
        return self._extract_video(channel_name, video_id, is_live=True)