youtube_dl/extractor/younow.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import itertools
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     try_get,
  12 )
  13
  14 CDN_API_BASE = 'https://cdn.younow.com/php/api'
  15 MOMENT_URL_FORMAT = '%s/moment/fetch/id=%%s' % CDN_API_BASE
  16
  17
  18 class YouNowLiveIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?younow\.com/(?P<id>[^/?#&]+)'
  20     _TEST = {
  21         'url': 'https://www.younow.com/AmandaPadeezy',
  22         'info_dict': {
  23             'id': 'AmandaPadeezy',
  24             'ext': 'mp4',
  25             'is_live': True,
  26             'title': 'March 26, 2017',
  27             'thumbnail': r're:^https?://.*\.jpg$',
  28             'tags': ['girls'],
  29             'categories': ['girls'],
  30             'uploader': 'AmandaPadeezy',
  31             'uploader_id': '6716501',
  32             'uploader_url': 'https://www.younow.com/AmandaPadeezy',
  33             'creator': 'AmandaPadeezy',
  34         },
  35         'skip': True,
  36     }
  37
  38     @classmethod
  39     def suitable(cls, url):
  40         return (False
  41                 if YouNowChannelIE.suitable(url) or YouNowMomentIE.suitable(url)
  42                 else super(YouNowLiveIE, cls).suitable(url))
  43
  44     def _real_extract(self, url):
  45         username = self._match_id(url)
  46
  47         data = self._download_json(
  48             'https://api.younow.com/php/api/broadcast/info/curId=0/user=%s'
  49             % username, username)
  50
  51         if data.get('errorCode') != 0:
  52             raise ExtractorError(data['errorMsg'], expected=True)
  53
  54         uploader = try_get(
  55             data, lambda x: x['user']['profileUrlString'],
  56             compat_str) or username
  57
  58         return {
  59             'id': uploader,
  60             'is_live': True,
  61             'title': self._live_title(uploader),
  62             'thumbnail': data.get('awsUrl'),
  63             'tags': data.get('tags'),
  64             'categories': data.get('tags'),
  65             'uploader': uploader,
  66             'uploader_id': data.get('userId'),
  67             'uploader_url': 'https://www.younow.com/%s' % username,
  68             'creator': uploader,
  69             'view_count': int_or_none(data.get('viewers')),
  70             'like_count': int_or_none(data.get('likes')),
  71             'formats': [{
  72                 'url': '%s/broadcast/videoPath/hls=1/broadcastId=%s/channelId=%s'
  73                        % (CDN_API_BASE, data['broadcastId'], data['userId']),
  74                 'ext': 'mp4',
  75                 'protocol': 'm3u8',
  76             }],
  77         }
  78
  79
  80 def _extract_moment(item, fatal=True):
  81     moment_id = item.get('momentId')
  82     if not moment_id:
  83         if not fatal:
  84             return
  85         raise ExtractorError('Unable to extract moment id')
  86
  87     moment_id = compat_str(moment_id)
  88
  89     title = item.get('text')
  90     if not title:
  91         title = 'YouNow %s' % (
  92             item.get('momentType') or item.get('titleType') or 'moment')
  93
  94     uploader = try_get(item, lambda x: x['owner']['name'], compat_str)
  95     uploader_id = try_get(item, lambda x: x['owner']['userId'])
  96     uploader_url = 'https://www.younow.com/%s' % uploader if uploader else None
  97
  98     entry = {
  99         'extractor_key': 'YouNowMoment',
 100         'id': moment_id,
 101         'title': title,
 102         'view_count': int_or_none(item.get('views')),
 103         'like_count': int_or_none(item.get('likes')),
 104         'timestamp': int_or_none(item.get('created')),
 105         'creator': uploader,
 106         'uploader': uploader,
 107         'uploader_id': uploader_id,
 108         'uploader_url': uploader_url,
 109         'formats': [{
 110             'url': 'https://hls.younow.com/momentsplaylists/live/%s/%s.m3u8'
 111                    % (moment_id, moment_id),
 112             'ext': 'mp4',
 113             'protocol': 'm3u8_native',
 114         }],
 115     }
 116
 117     return entry
 118
 119
 120 class YouNowChannelIE(InfoExtractor):
 121     _VALID_URL = r'https?://(?:www\.)?younow\.com/(?P<id>[^/]+)/channel'
 122     _TEST = {
 123         'url': 'https://www.younow.com/its_Kateee_/channel',
 124         'info_dict': {
 125             'id': '14629760',
 126             'title': 'its_Kateee_ moments'
 127         },
 128         'playlist_mincount': 8,
 129     }
 130
 131     def _entries(self, username, channel_id):
 132         created_before = 0
 133         for page_num in itertools.count(1):
 134             if created_before is None:
 135                 break
 136             info = self._download_json(
 137                 '%s/moment/profile/channelId=%s/createdBefore=%d/records=20'
 138                 % (CDN_API_BASE, channel_id, created_before), username,
 139                 note='Downloading moments page %d' % page_num)
 140             items = info.get('items')
 141             if not items or not isinstance(items, list):
 142                 break
 143             for item in items:
 144                 if not isinstance(item, dict):
 145                     continue
 146                 item_type = item.get('type')
 147                 if item_type == 'moment':
 148                     entry = _extract_moment(item, fatal=False)
 149                     if entry:
 150                         yield entry
 151                 elif item_type == 'collection':
 152                     moments = item.get('momentsIds')
 153                     if isinstance(moments, list):
 154                         for moment_id in moments:
 155                             m = self._download_json(
 156                                 MOMENT_URL_FORMAT % moment_id, username,
 157                                 note='Downloading %s moment JSON' % moment_id,
 158                                 fatal=False)
 159                             if m and isinstance(m, dict) and m.get('item'):
 160                                 entry = _extract_moment(m['item'])
 161                                 if entry:
 162                                     yield entry
 163                 created_before = int_or_none(item.get('created'))
 164
 165     def _real_extract(self, url):
 166         username = self._match_id(url)
 167         channel_id = compat_str(self._download_json(
 168             'https://api.younow.com/php/api/broadcast/info/curId=0/user=%s'
 169             % username, username, note='Downloading user information')['userId'])
 170         return self.playlist_result(
 171             self._entries(username, channel_id), channel_id,
 172             '%s moments' % username)
 173
 174
 175 class YouNowMomentIE(InfoExtractor):
 176     _VALID_URL = r'https?://(?:www\.)?younow\.com/[^/]+/(?P<id>[^/?#&]+)'
 177     _TEST = {
 178         'url': 'https://www.younow.com/GABO.../20712117/36319236/3b316doc/m',
 179         'md5': 'a30c70eadb9fb39a1aa3c8c0d22a0807',
 180         'info_dict': {
 181             'id': '20712117',
 182             'ext': 'mp4',
 183             'title': 'YouNow capture',
 184             'view_count': int,
 185             'like_count': int,
 186             'timestamp': 1490432040,
 187             'upload_date': '20170325',
 188             'uploader': 'GABO...',
 189             'uploader_id': 35917228,
 190         },
 191     }
 192
 193     @classmethod
 194     def suitable(cls, url):
 195         return (False
 196                 if YouNowChannelIE.suitable(url)
 197                 else super(YouNowMomentIE, cls).suitable(url))
 198
 199     def _real_extract(self, url):
 200         video_id = self._match_id(url)
 201         item = self._download_json(MOMENT_URL_FORMAT % video_id, video_id)
 202         return _extract_moment(item['item'])