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