yt_dlp/extractor/curiositystream.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     urlencode_postdata,
  10     compat_str,
  11     ExtractorError,
  12 )
  13
  14
  15 class CuriosityStreamBaseIE(InfoExtractor):
  16     _NETRC_MACHINE = 'curiositystream'
  17     _auth_token = None
  18     _API_BASE_URL = 'https://api.curiositystream.com/v1/'
  19
  20     def _handle_errors(self, result):
  21         error = result.get('error', {}).get('message')
  22         if error:
  23             if isinstance(error, dict):
  24                 error = ', '.join(error.values())
  25             raise ExtractorError(
  26                 '%s said: %s' % (self.IE_NAME, error), expected=True)
  27
  28     def _call_api(self, path, video_id, query=None):
  29         headers = {}
  30         if self._auth_token:
  31             headers['X-Auth-Token'] = self._auth_token
  32         result = self._download_json(
  33             self._API_BASE_URL + path, video_id, headers=headers, query=query)
  34         self._handle_errors(result)
  35         return result['data']
  36
  37     def _real_initialize(self):
  38         email, password = self._get_login_info()
  39         if email is None:
  40             return
  41         result = self._download_json(
  42             self._API_BASE_URL + 'login', None, data=urlencode_postdata({
  43                 'email': email,
  44                 'password': password,
  45             }))
  46         self._handle_errors(result)
  47         self._auth_token = result['message']['auth_token']
  48
  49
  50 class CuriosityStreamIE(CuriosityStreamBaseIE):
  51     IE_NAME = 'curiositystream'
  52     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/video/(?P<id>\d+)'
  53     _TEST = {
  54         'url': 'https://app.curiositystream.com/video/2',
  55         'info_dict': {
  56             'id': '2',
  57             'ext': 'mp4',
  58             'title': 'How Did You Develop The Internet?',
  59             'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
  60         },
  61         'params': {
  62             'format': 'bestvideo',
  63             # m3u8 download
  64             'skip_download': True,
  65         },
  66     }
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         formats = []
  72         for encoding_format in ('m3u8', 'mpd'):
  73             media = self._call_api('media/' + video_id, video_id, query={
  74                 'encodingsNew': 'true',
  75                 'encodingsFormat': encoding_format,
  76             })
  77             for encoding in media.get('encodings', []):
  78                 playlist_url = encoding.get('master_playlist_url')
  79                 if encoding_format == 'm3u8':
  80                     # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol
  81                     formats.extend(self._extract_m3u8_formats(
  82                         playlist_url, video_id, 'mp4',
  83                         m3u8_id='hls', fatal=False))
  84                 elif encoding_format == 'mpd':
  85                     formats.extend(self._extract_mpd_formats(
  86                         playlist_url, video_id, mpd_id='dash', fatal=False))
  87                 encoding_url = encoding.get('url')
  88                 file_url = encoding.get('file_url')
  89                 if not encoding_url and not file_url:
  90                     continue
  91                 f = {
  92                     'width': int_or_none(encoding.get('width')),
  93                     'height': int_or_none(encoding.get('height')),
  94                     'vbr': int_or_none(encoding.get('video_bitrate')),
  95                     'abr': int_or_none(encoding.get('audio_bitrate')),
  96                     'filesize': int_or_none(encoding.get('size_in_bytes')),
  97                     'vcodec': encoding.get('video_codec'),
  98                     'acodec': encoding.get('audio_codec'),
  99                     'container': encoding.get('container_type'),
 100                 }
 101                 for f_url in (encoding_url, file_url):
 102                     if not f_url:
 103                         continue
 104                     fmt = f.copy()
 105                     rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
 106                     if rtmp:
 107                         fmt.update({
 108                             'url': rtmp.group('url'),
 109                             'play_path': rtmp.group('playpath'),
 110                             'app': rtmp.group('app'),
 111                             'ext': 'flv',
 112                             'format_id': 'rtmp',
 113                         })
 114                     else:
 115                         fmt.update({
 116                             'url': f_url,
 117                             'format_id': 'http',
 118                         })
 119                     formats.append(fmt)
 120         self._sort_formats(formats)
 121
 122         title = media['title']
 123
 124         subtitles = {}
 125         for closed_caption in media.get('closed_captions', []):
 126             sub_url = closed_caption.get('file')
 127             if not sub_url:
 128                 continue
 129             lang = closed_caption.get('code') or closed_caption.get('language') or 'en'
 130             subtitles.setdefault(lang, []).append({
 131                 'url': sub_url,
 132             })
 133
 134         return {
 135             'id': video_id,
 136             'formats': formats,
 137             'title': title,
 138             'description': media.get('description'),
 139             'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'),
 140             'duration': int_or_none(media.get('duration')),
 141             'tags': media.get('tags'),
 142             'subtitles': subtitles,
 143         }
 144
 145
 146 class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
 147     IE_NAME = 'curiositystream:collection'
 148     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collections?|series)/(?P<id>\d+)'
 149     _API_BASE_URL = 'https://api.curiositystream.com/v2/collections/'
 150     _TESTS = [{
 151         'url': 'https://curiositystream.com/collections/86',
 152         'info_dict': {
 153             'id': '86',
 154             'title': 'Staff Picks',
 155             'description': 'Wondering where to start? Here are a few of our favorite series and films... from our couch to yours.',
 156         },
 157         'playlist_mincount': 7,
 158     }, {
 159         'url': 'https://app.curiositystream.com/collection/2',
 160         'info_dict': {
 161             'id': '2',
 162             'title': 'Curious Minds: The Internet',
 163             'description': 'How is the internet shaping our lives in the 21st Century?',
 164         },
 165         'playlist_mincount': 16,
 166     }, {
 167         'url': 'https://curiositystream.com/series/2',
 168         'only_matching': True,
 169     }, {
 170         'url': 'https://curiositystream.com/collections/36',
 171         'only_matching': True,
 172     }]
 173
 174     def _real_extract(self, url):
 175         collection_id = self._match_id(url)
 176         collection = self._call_api(collection_id, collection_id)
 177         entries = []
 178         for media in collection.get('media', []):
 179             media_id = compat_str(media.get('id'))
 180             media_type, ie = ('series', CuriosityStreamCollectionIE) if media.get('is_collection') else ('video', CuriosityStreamIE)
 181             entries.append(self.url_result(
 182                 'https://curiositystream.com/%s/%s' % (media_type, media_id),
 183                 ie=ie.ie_key(), video_id=media_id))
 184         return self.playlist_result(
 185             entries, collection_id,
 186             collection.get('title'), collection.get('description'))