l1ving_youtube-dl/youtube_dl/extractor/seeso.py

# coding: utf-8
from __future__ import unicode_literals

from .common import InfoExtractor
import urlparse
import os.path
import json
from ..utils import (
    sanitized_Request,
    urlencode_postdata
)


class SeesoIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?seeso\.com/view/episode/(?P<id>[0-9]+)'
    _TEST = {
        # 'md5': 'TODO: md5 sum of the first 10241 bytes of the video file (use --test)',
        'params': {
            'username': 'emailhere',
            'password': 'passwordhere'
        },
        'url': 'https://www.seeso.com/view/episode/799241283849',
        'info_dict': {
            'id': '799241283849',
            'ext': 'mp4',
            'series': 'Bajillion Dollar Propertie$',
            'title': 'Farsi Lessons',
            'thumbnail': 'https://chaosic.akamaized.net/NBCOTT_-_Production/360/907/160831_3092487_Farsi_Lessons.jpg',
            'description': 'Amir leads Victoria into an embarrassing trap, the Bros meet an even more obnoxious bro '
                           'who’s looking for the ultimate sex pad, while rival brokers Serge and Gio decide to get '
                           'rough with  Glenn.'
        }
    }

    def _real_extract(self, url):
        username, password = self._get_login_info()
        video_id = self._match_id(url)

        if username is None or password is None:
            return

        _API_AUTH_URL = "https://www.seeso.com/api/v3/auth/login"
        auth_json = {
            "username": username,
            "password": password,
            "userOnly": True,
            "isRememberMe": True,
            "isWeb": True
        }
        # TODO Change JSON gets to .get() as referenced in readme

        # Send auth POST request and get token from response
        auth_request = sanitized_Request(_API_AUTH_URL, urlencode_postdata(auth_json))
        auth_response = json.loads(self._download_webpage(auth_request, '', note='Getting auth token...'))
        auth_token = auth_response['user']['token']

        # Use the public unauthenticated API to get the video's info
        _VIDEO_INFO_URL = 'https://feed.theplatform.com/f/NZILfC/nbcott-prod-all-media?byAvailabilityState=' \
                          'available&byId=%s&form=cjson'
        request = sanitized_Request(_VIDEO_INFO_URL % video_id)
        json_data = self._download_json(request, video_id)
        entry = json_data.get('entries')[0]

        # Custom fields
        public_url = entry["publicUrl"]

        # Template fields
        series = entry["nbc-chaos$show"]                    # Show name
        title = entry["nbc-chaos$shortTitle"]               # Episode Name
        season_number = entry["nbc-chaos$seasonNumber"]
        episode_number = entry["nbc-chaos$episodeNumber"]
        thumbnail = entry["defaultThumbnailUrl"]
        description = entry["nbc-chaos$shortDescription"]

        # Got the show's public URL. Now we need to parse out the videoID
        public_url_id = os.path.split(urlparse.urlparse(public_url).path)[-1]

        # Get the master m3u8 which lists formats
        m3u8_url = 'https://link.theplatform.com/s/NZILfC/' \
                   'media/{0}?feed=All%20Media%20Feed&auth={1}&vpaid=script,flash' \
                   '&formats=m3u,mpeg4'.format(public_url_id, auth_token)
        formats = []
        for entry in self._extract_m3u8_formats(m3u8_url, video_id, m3u8_id='m3u8'):
            formats.append(entry)

        self._sort_formats(formats)

        return {
                'id': video_id,
                'title': title,
                'thumbnail': thumbnail,
                'series': series,
                'season_number': season_number,
                'episode_number': episode_number,
                'description': description,
                'url': '',
                'formats': formats
        }