youtube_dlc/extractor/googledrive.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_parse_qs
   7 from ..utils import (
   8     determine_ext,
   9     ExtractorError,
  10     int_or_none,
  11     lowercase_escape,
  12     try_get,
  13     update_url_query,
  14 )
  15
  16
  17 class GoogleDriveIE(InfoExtractor):
  18     _VALID_URL = r'''(?x)
  19                         https?://
  20                             (?:
  21                                 (?:docs|drive)\.google\.com/
  22                                 (?:
  23                                     (?:uc|open)\?.*?id=|
  24                                     file/d/
  25                                 )|
  26                                 video\.google\.com/get_player\?.*?docid=
  27                             )
  28                             (?P<id>[a-zA-Z0-9_-]{28,})
  29                     '''
  30     _TESTS = [{
  31         'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1',
  32         'md5': '5c602afbbf2c1db91831f5d82f678554',
  33         'info_dict': {
  34             'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  35             'ext': 'mp4',
  36             'title': 'Big Buck Bunny.mp4',
  37             'duration': 45,
  38         }
  39     }, {
  40         # video can't be watched anonymously due to view count limit reached,
  41         # but can be downloaded (see https://github.com/ytdl-org/youtube-dl/issues/14046)
  42         'url': 'https://drive.google.com/file/d/0B-vUyvmDLdWDcEt4WjBqcmI2XzQ/view',
  43         'only_matching': True,
  44     }, {
  45         # video id is longer than 28 characters
  46         'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit',
  47         'only_matching': True,
  48     }, {
  49         'url': 'https://drive.google.com/open?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  50         'only_matching': True,
  51     }, {
  52         'url': 'https://drive.google.com/uc?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  53         'only_matching': True,
  54     }]
  55     _FORMATS_EXT = {
  56         '5': 'flv',
  57         '6': 'flv',
  58         '13': '3gp',
  59         '17': '3gp',
  60         '18': 'mp4',
  61         '22': 'mp4',
  62         '34': 'flv',
  63         '35': 'flv',
  64         '36': '3gp',
  65         '37': 'mp4',
  66         '38': 'mp4',
  67         '43': 'webm',
  68         '44': 'webm',
  69         '45': 'webm',
  70         '46': 'webm',
  71         '59': 'mp4',
  72     }
  73     _BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext'
  74     _CAPTIONS_ENTRY_TAG = {
  75         'subtitles': 'track',
  76         'automatic_captions': 'target',
  77     }
  78     _caption_formats_ext = []
  79     _captions_xml = None
  80
  81     @staticmethod
  82     def _extract_url(webpage):
  83         mobj = re.search(
  84             r'<iframe[^>]+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P<id>[a-zA-Z0-9_-]{28,})',
  85             webpage)
  86         if mobj:
  87             return 'https://drive.google.com/file/d/%s' % mobj.group('id')
  88
  89     def _download_subtitles_xml(self, video_id, subtitles_id, hl):
  90         if self._captions_xml:
  91             return
  92         self._captions_xml = self._download_xml(
  93             self._BASE_URL_CAPTIONS, video_id, query={
  94                 'id': video_id,
  95                 'vid': subtitles_id,
  96                 'hl': hl,
  97                 'v': video_id,
  98                 'type': 'list',
  99                 'tlangs': '1',
 100                 'fmts': '1',
 101                 'vssids': '1',
 102             }, note='Downloading subtitles XML',
 103             errnote='Unable to download subtitles XML', fatal=False)
 104         if self._captions_xml:
 105             for f in self._captions_xml.findall('format'):
 106                 if f.attrib.get('fmt_code') and not f.attrib.get('default'):
 107                     self._caption_formats_ext.append(f.attrib['fmt_code'])
 108
 109     def _get_captions_by_type(self, video_id, subtitles_id, caption_type,
 110                               origin_lang_code=None):
 111         if not subtitles_id or not caption_type:
 112             return
 113         captions = {}
 114         for caption_entry in self._captions_xml.findall(
 115                 self._CAPTIONS_ENTRY_TAG[caption_type]):
 116             caption_lang_code = caption_entry.attrib.get('lang_code')
 117             if not caption_lang_code:
 118                 continue
 119             caption_format_data = []
 120             for caption_format in self._caption_formats_ext:
 121                 query = {
 122                     'vid': subtitles_id,
 123                     'v': video_id,
 124                     'fmt': caption_format,
 125                     'lang': (caption_lang_code if origin_lang_code is None
 126                              else origin_lang_code),
 127                     'type': 'track',
 128                     'name': '',
 129                     'kind': '',
 130                 }
 131                 if origin_lang_code is not None:
 132                     query.update({'tlang': caption_lang_code})
 133                 caption_format_data.append({
 134                     'url': update_url_query(self._BASE_URL_CAPTIONS, query),
 135                     'ext': caption_format,
 136                 })
 137             captions[caption_lang_code] = caption_format_data
 138         return captions
 139
 140     def _get_subtitles(self, video_id, subtitles_id, hl):
 141         if not subtitles_id or not hl:
 142             return
 143         self._download_subtitles_xml(video_id, subtitles_id, hl)
 144         if not self._captions_xml:
 145             return
 146         return self._get_captions_by_type(video_id, subtitles_id, 'subtitles')
 147
 148     def _get_automatic_captions(self, video_id, subtitles_id, hl):
 149         if not subtitles_id or not hl:
 150             return
 151         self._download_subtitles_xml(video_id, subtitles_id, hl)
 152         if not self._captions_xml:
 153             return
 154         track = self._captions_xml.find('track')
 155         if track is None:
 156             return
 157         origin_lang_code = track.attrib.get('lang_code')
 158         if not origin_lang_code:
 159             return
 160         return self._get_captions_by_type(
 161             video_id, subtitles_id, 'automatic_captions', origin_lang_code)
 162
 163     def _real_extract(self, url):
 164         video_id = self._match_id(url)
 165         video_info = compat_parse_qs(self._download_webpage(
 166             'https://drive.google.com/get_video_info',
 167             video_id, query={'docid': video_id}))
 168
 169         def get_value(key):
 170             return try_get(video_info, lambda x: x[key][0])
 171
 172         reason = get_value('reason')
 173         title = get_value('title')
 174         if not title and reason:
 175             raise ExtractorError(reason, expected=True)
 176
 177         formats = []
 178         fmt_stream_map = (get_value('fmt_stream_map') or '').split(',')
 179         fmt_list = (get_value('fmt_list') or '').split(',')
 180         if fmt_stream_map and fmt_list:
 181             resolutions = {}
 182             for fmt in fmt_list:
 183                 mobj = re.search(
 184                     r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
 185                 if mobj:
 186                     resolutions[mobj.group('format_id')] = (
 187                         int(mobj.group('width')), int(mobj.group('height')))
 188
 189             for fmt_stream in fmt_stream_map:
 190                 fmt_stream_split = fmt_stream.split('|')
 191                 if len(fmt_stream_split) < 2:
 192                     continue
 193                 format_id, format_url = fmt_stream_split[:2]
 194                 f = {
 195                     'url': lowercase_escape(format_url),
 196                     'format_id': format_id,
 197                     'ext': self._FORMATS_EXT[format_id],
 198                 }
 199                 resolution = resolutions.get(format_id)
 200                 if resolution:
 201                     f.update({
 202                         'width': resolution[0],
 203                         'height': resolution[1],
 204                     })
 205                 formats.append(f)
 206
 207         source_url = update_url_query(
 208             'https://drive.google.com/uc', {
 209                 'id': video_id,
 210                 'export': 'download',
 211             })
 212
 213         def request_source_file(source_url, kind):
 214             return self._request_webpage(
 215                 source_url, video_id, note='Requesting %s file' % kind,
 216                 errnote='Unable to request %s file' % kind, fatal=False)
 217         urlh = request_source_file(source_url, 'source')
 218         if urlh:
 219             def add_source_format(urlh):
 220                 formats.append({
 221                     # Use redirect URLs as download URLs in order to calculate
 222                     # correct cookies in _calc_cookies.
 223                     # Using original URLs may result in redirect loop due to
 224                     # google.com's cookies mistakenly used for googleusercontent.com
 225                     # redirect URLs (see #23919).
 226                     'url': urlh.geturl(),
 227                     'ext': determine_ext(title, 'mp4').lower(),
 228                     'format_id': 'source',
 229                     'quality': 1,
 230                 })
 231             if urlh.headers.get('Content-Disposition'):
 232                 add_source_format(urlh)
 233             else:
 234                 confirmation_webpage = self._webpage_read_content(
 235                     urlh, url, video_id, note='Downloading confirmation page',
 236                     errnote='Unable to confirm download', fatal=False)
 237                 if confirmation_webpage:
 238                     confirm = self._search_regex(
 239                         r'confirm=([^&"\']+)', confirmation_webpage,
 240                         'confirmation code', fatal=False)
 241                     if confirm:
 242                         confirmed_source_url = update_url_query(source_url, {
 243                             'confirm': confirm,
 244                         })
 245                         urlh = request_source_file(confirmed_source_url, 'confirmed source')
 246                         if urlh and urlh.headers.get('Content-Disposition'):
 247                             add_source_format(urlh)
 248
 249         if not formats and reason:
 250             raise ExtractorError(reason, expected=True)
 251
 252         self._sort_formats(formats)
 253
 254         hl = get_value('hl')
 255         subtitles_id = None
 256         ttsurl = get_value('ttsurl')
 257         if ttsurl:
 258             # the video Id for subtitles will be the last value in the ttsurl
 259             # query string
 260             subtitles_id = ttsurl.encode('utf-8').decode(
 261                 'unicode_escape').split('=')[-1]
 262
 263         self._downloader.cookiejar.clear(domain='.google.com', path='/', name='NID')
 264
 265         return {
 266             'id': video_id,
 267             'title': title,
 268             'thumbnail': 'https://drive.google.com/thumbnail?id=' + video_id,
 269             'duration': int_or_none(get_value('length_seconds')),
 270             'formats': formats,
 271             'subtitles': self.extract_subtitles(video_id, subtitles_id, hl),
 272             'automatic_captions': self.extract_automatic_captions(
 273                 video_id, subtitles_id, hl),
 274         }