_ Git - youtube-dl/blob - youtube_dl/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):
  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)
  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     def _extract_media_info(self, media):
  50         video_id = compat_str(media['id'])
  51         limelight_media_id = media['limelight_media_id']
  52         title = media['title']
  53
  54         formats = []
  55         for encoding in media.get('encodings', []):
  56             m3u8_url = encoding.get('master_playlist_url')
  57             if m3u8_url:
  58                 formats.extend(self._extract_m3u8_formats(
  59                     m3u8_url, video_id, 'mp4', 'm3u8_native',
  60                     m3u8_id='hls', fatal=False))
  61             encoding_url = encoding.get('url')
  62             file_url = encoding.get('file_url')
  63             if not encoding_url and not file_url:
  64                 continue
  65             f = {
  66                 'width': int_or_none(encoding.get('width')),
  67                 'height': int_or_none(encoding.get('height')),
  68                 'vbr': int_or_none(encoding.get('video_bitrate')),
  69                 'abr': int_or_none(encoding.get('audio_bitrate')),
  70                 'filesize': int_or_none(encoding.get('size_in_bytes')),
  71                 'vcodec': encoding.get('video_codec'),
  72                 'acodec': encoding.get('audio_codec'),
  73                 'container': encoding.get('container_type'),
  74             }
  75             for f_url in (encoding_url, file_url):
  76                 if not f_url:
  77                     continue
  78                 fmt = f.copy()
  79                 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url)
  80                 if rtmp:
  81                     fmt.update({
  82                         'url': rtmp.group('url'),
  83                         'play_path': rtmp.group('playpath'),
  84                         'app': rtmp.group('app'),
  85                         'ext': 'flv',
  86                         'format_id': 'rtmp',
  87                     })
  88                 else:
  89                     fmt.update({
  90                         'url': f_url,
  91                         'format_id': 'http',
  92                     })
  93                 formats.append(fmt)
  94         self._sort_formats(formats)
  95
  96         subtitles = {}
  97         for closed_caption in media.get('closed_captions', []):
  98             sub_url = closed_caption.get('file')
  99             if not sub_url:
 100                 continue
 101             lang = closed_caption.get('code') or closed_caption.get('language') or 'en'
 102             subtitles.setdefault(lang, []).append({
 103                 'url': sub_url,
 104             })
 105
 106         return {
 107             'id': video_id,
 108             'formats': formats,
 109             'title': title,
 110             'description': media.get('description'),
 111             'thumbnail': media.get('image_large') or media.get('image_medium') or media.get('image_small'),
 112             'duration': int_or_none(media.get('duration')),
 113             'tags': media.get('tags'),
 114             'subtitles': subtitles,
 115         }
 116
 117
 118 class CuriosityStreamIE(CuriosityStreamBaseIE):
 119     IE_NAME = 'curiositystream'
 120     _VALID_URL = r'https?://app\.curiositystream\.com/video/(?P<id>\d+)'
 121     _TEST = {
 122         'url': 'https://app.curiositystream.com/video/2',
 123         'md5': '262bb2f257ff301115f1973540de8983',
 124         'info_dict': {
 125             'id': '2',
 126             'ext': 'mp4',
 127             'title': 'How Did You Develop The Internet?',
 128             'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
 129         }
 130     }
 131
 132     def _real_extract(self, url):
 133         video_id = self._match_id(url)
 134         media = self._call_api('media/' + video_id, video_id)
 135         return self._extract_media_info(media)
 136
 137
 138 class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
 139     IE_NAME = 'curiositystream:collection'
 140     _VALID_URL = r'https?://app\.curiositystream\.com/collection/(?P<id>\d+)'
 141     _TEST = {
 142         'url': 'https://app.curiositystream.com/collection/2',
 143         'info_dict': {
 144             'id': '2',
 145             'title': 'Curious Minds: The Internet',
 146             'description': 'How is the internet shaping our lives in the 21st Century?',
 147         },
 148         'playlist_mincount': 12,
 149     }
 150
 151     def _real_extract(self, url):
 152         collection_id = self._match_id(url)
 153         collection = self._call_api(
 154             'collections/' + collection_id, collection_id)
 155         entries = []
 156         for media in collection.get('media', []):
 157             entries.append(self._extract_media_info(media))
 158         return self.playlist_result(
 159             entries, collection_id,
 160             collection.get('title'), collection.get('description'))