yt_dlp/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     get_element_by_class,
  11     int_or_none,
  12     lowercase_escape,
  13     try_get,
  14     update_url_query,
  15 )
  16
  17
  18 class GoogleDriveIE(InfoExtractor):
  19     _VALID_URL = r'''(?x)
  20                         https?://
  21                             (?:
  22                                 (?:docs|drive)\.google\.com/
  23                                 (?:
  24                                     (?:uc|open)\?.*?id=|
  25                                     file/d/
  26                                 )|
  27                                 video\.google\.com/get_player\?.*?docid=
  28                             )
  29                             (?P<id>[a-zA-Z0-9_-]{28,})
  30                     '''
  31     _TESTS = [{
  32         'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1',
  33         'md5': '5c602afbbf2c1db91831f5d82f678554',
  34         'info_dict': {
  35             'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  36             'ext': 'mp4',
  37             'title': 'Big Buck Bunny.mp4',
  38             'duration': 45,
  39         }
  40     }, {
  41         # video can't be watched anonymously due to view count limit reached,
  42         # but can be downloaded (see https://github.com/ytdl-org/youtube-dl/issues/14046)
  43         'url': 'https://drive.google.com/file/d/0B-vUyvmDLdWDcEt4WjBqcmI2XzQ/view',
  44         'only_matching': True,
  45     }, {
  46         # video id is longer than 28 characters
  47         'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit',
  48         'only_matching': True,
  49     }, {
  50         'url': 'https://drive.google.com/open?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  51         'only_matching': True,
  52     }, {
  53         'url': 'https://drive.google.com/uc?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  54         'only_matching': True,
  55     }]
  56     _FORMATS_EXT = {
  57         '5': 'flv',
  58         '6': 'flv',
  59         '13': '3gp',
  60         '17': '3gp',
  61         '18': 'mp4',
  62         '22': 'mp4',
  63         '34': 'flv',
  64         '35': 'flv',
  65         '36': '3gp',
  66         '37': 'mp4',
  67         '38': 'mp4',
  68         '43': 'webm',
  69         '44': 'webm',
  70         '45': 'webm',
  71         '46': 'webm',
  72         '59': 'mp4',
  73     }
  74     _BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext'
  75     _CAPTIONS_ENTRY_TAG = {
  76         'subtitles': 'track',
  77         'automatic_captions': 'target',
  78     }
  79     _caption_formats_ext = []
  80     _captions_xml = None
  81
  82     @staticmethod
  83     def _extract_url(webpage):
  84         mobj = re.search(
  85             r'<iframe[^>]+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P<id>[a-zA-Z0-9_-]{28,})',
  86             webpage)
  87         if mobj:
  88             return 'https://drive.google.com/file/d/%s' % mobj.group('id')
  89
  90     def _download_subtitles_xml(self, video_id, subtitles_id, hl):
  91         if self._captions_xml:
  92             return
  93         self._captions_xml = self._download_xml(
  94             self._BASE_URL_CAPTIONS, video_id, query={
  95                 'id': video_id,
  96                 'vid': subtitles_id,
  97                 'hl': hl,
  98                 'v': video_id,
  99                 'type': 'list',
 100                 'tlangs': '1',
 101                 'fmts': '1',
 102                 'vssids': '1',
 103             }, note='Downloading subtitles XML',
 104             errnote='Unable to download subtitles XML', fatal=False)
 105         if self._captions_xml:
 106             for f in self._captions_xml.findall('format'):
 107                 if f.attrib.get('fmt_code') and not f.attrib.get('default'):
 108                     self._caption_formats_ext.append(f.attrib['fmt_code'])
 109
 110     def _get_captions_by_type(self, video_id, subtitles_id, caption_type,
 111                               origin_lang_code=None):
 112         if not subtitles_id or not caption_type:
 113             return
 114         captions = {}
 115         for caption_entry in self._captions_xml.findall(
 116                 self._CAPTIONS_ENTRY_TAG[caption_type]):
 117             caption_lang_code = caption_entry.attrib.get('lang_code')
 118             if not caption_lang_code:
 119                 continue
 120             caption_format_data = []
 121             for caption_format in self._caption_formats_ext:
 122                 query = {
 123                     'vid': subtitles_id,
 124                     'v': video_id,
 125                     'fmt': caption_format,
 126                     'lang': (caption_lang_code if origin_lang_code is None
 127                              else origin_lang_code),
 128                     'type': 'track',
 129                     'name': '',
 130                     'kind': '',
 131                 }
 132                 if origin_lang_code is not None:
 133                     query.update({'tlang': caption_lang_code})
 134                 caption_format_data.append({
 135                     'url': update_url_query(self._BASE_URL_CAPTIONS, query),
 136                     'ext': caption_format,
 137                 })
 138             captions[caption_lang_code] = caption_format_data
 139         return captions
 140
 141     def _get_subtitles(self, video_id, subtitles_id, hl):
 142         if not subtitles_id or not hl:
 143             return
 144         self._download_subtitles_xml(video_id, subtitles_id, hl)
 145         if not self._captions_xml:
 146             return
 147         return self._get_captions_by_type(video_id, subtitles_id, 'subtitles')
 148
 149     def _get_automatic_captions(self, video_id, subtitles_id, hl):
 150         if not subtitles_id or not hl:
 151             return
 152         self._download_subtitles_xml(video_id, subtitles_id, hl)
 153         if not self._captions_xml:
 154             return
 155         track = self._captions_xml.find('track')
 156         if track is None:
 157             return
 158         origin_lang_code = track.attrib.get('lang_code')
 159         if not origin_lang_code:
 160             return
 161         return self._get_captions_by_type(
 162             video_id, subtitles_id, 'automatic_captions', origin_lang_code)
 163
 164     def _real_extract(self, url):
 165         video_id = self._match_id(url)
 166         video_info = compat_parse_qs(self._download_webpage(
 167             'https://drive.google.com/get_video_info',
 168             video_id, query={'docid': video_id}))
 169
 170         def get_value(key):
 171             return try_get(video_info, lambda x: x[key][0])
 172
 173         reason = get_value('reason')
 174         title = get_value('title')
 175         if not title and reason:
 176             raise ExtractorError(reason, expected=True)
 177
 178         formats = []
 179         fmt_stream_map = (get_value('fmt_stream_map') or '').split(',')
 180         fmt_list = (get_value('fmt_list') or '').split(',')
 181         if fmt_stream_map and fmt_list:
 182             resolutions = {}
 183             for fmt in fmt_list:
 184                 mobj = re.search(
 185                     r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
 186                 if mobj:
 187                     resolutions[mobj.group('format_id')] = (
 188                         int(mobj.group('width')), int(mobj.group('height')))
 189
 190             for fmt_stream in fmt_stream_map:
 191                 fmt_stream_split = fmt_stream.split('|')
 192                 if len(fmt_stream_split) < 2:
 193                     continue
 194                 format_id, format_url = fmt_stream_split[:2]
 195                 f = {
 196                     'url': lowercase_escape(format_url),
 197                     'format_id': format_id,
 198                     'ext': self._FORMATS_EXT[format_id],
 199                 }
 200                 resolution = resolutions.get(format_id)
 201                 if resolution:
 202                     f.update({
 203                         'width': resolution[0],
 204                         'height': resolution[1],
 205                     })
 206                 formats.append(f)
 207
 208         source_url = update_url_query(
 209             'https://drive.google.com/uc', {
 210                 'id': video_id,
 211                 'export': 'download',
 212             })
 213
 214         def request_source_file(source_url, kind):
 215             return self._request_webpage(
 216                 source_url, video_id, note='Requesting %s file' % kind,
 217                 errnote='Unable to request %s file' % kind, fatal=False)
 218         urlh = request_source_file(source_url, 'source')
 219         if urlh:
 220             def add_source_format(urlh):
 221                 formats.append({
 222                     # Use redirect URLs as download URLs in order to calculate
 223                     # correct cookies in _calc_cookies.
 224                     # Using original URLs may result in redirect loop due to
 225                     # google.com's cookies mistakenly used for googleusercontent.com
 226                     # redirect URLs (see #23919).
 227                     'url': urlh.geturl(),
 228                     'ext': determine_ext(title, 'mp4').lower(),
 229                     'format_id': 'source',
 230                     'quality': 1,
 231                 })
 232             if urlh.headers.get('Content-Disposition'):
 233                 add_source_format(urlh)
 234             else:
 235                 confirmation_webpage = self._webpage_read_content(
 236                     urlh, url, video_id, note='Downloading confirmation page',
 237                     errnote='Unable to confirm download', fatal=False)
 238                 if confirmation_webpage:
 239                     confirm = self._search_regex(
 240                         r'confirm=([^&"\']+)', confirmation_webpage,
 241                         'confirmation code', default=None)
 242                     if confirm:
 243                         confirmed_source_url = update_url_query(source_url, {
 244                             'confirm': confirm,
 245                         })
 246                         urlh = request_source_file(confirmed_source_url, 'confirmed source')
 247                         if urlh and urlh.headers.get('Content-Disposition'):
 248                             add_source_format(urlh)
 249                     else:
 250                         self.report_warning(
 251                             get_element_by_class('uc-error-subcaption', confirmation_webpage)
 252                             or get_element_by_class('uc-error-caption', confirmation_webpage)
 253                             or 'unable to extract confirmation code')
 254
 255         if not formats and reason:
 256             self.raise_no_formats(reason, expected=True)
 257
 258         self._sort_formats(formats)
 259
 260         hl = get_value('hl')
 261         subtitles_id = None
 262         ttsurl = get_value('ttsurl')
 263         if ttsurl:
 264             # the video Id for subtitles will be the last value in the ttsurl
 265             # query string
 266             subtitles_id = ttsurl.encode('utf-8').decode(
 267                 'unicode_escape').split('=')[-1]
 268
 269         self._downloader.cookiejar.clear(domain='.google.com', path='/', name='NID')
 270
 271         return {
 272             'id': video_id,
 273             'title': title,
 274             'thumbnail': 'https://drive.google.com/thumbnail?id=' + video_id,
 275             'duration': int_or_none(get_value('length_seconds')),
 276             'formats': formats,
 277             'subtitles': self.extract_subtitles(video_id, subtitles_id, hl),
 278             'automatic_captions': self.extract_automatic_captions(
 279                 video_id, subtitles_id, hl),
 280         }