yt_dlp/extractor/googledrive.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_parse_qs
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     extract_attributes,
   9     get_element_by_class,
  10     get_element_html_by_id,
  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|drive\.usercontent)\.google\.com/
  23                                 (?:
  24                                     (?:uc|open|download)\?.*?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             'thumbnail': 'https://drive.google.com/thumbnail?id=0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  40         }
  41     }, {
  42         # video can't be watched anonymously due to view count limit reached,
  43         # but can be downloaded (see https://github.com/ytdl-org/youtube-dl/issues/14046)
  44         'url': 'https://drive.google.com/file/d/0B-vUyvmDLdWDcEt4WjBqcmI2XzQ/view',
  45         'only_matching': True,
  46     }, {
  47         # video id is longer than 28 characters
  48         'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit',
  49         'only_matching': True,
  50     }, {
  51         'url': 'https://drive.google.com/open?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  52         'only_matching': True,
  53     }, {
  54         'url': 'https://drive.google.com/uc?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  55         'only_matching': True,
  56     }, {
  57         'url': 'https://drive.usercontent.google.com/download?id=0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  58         'only_matching': True,
  59     }]
  60     _FORMATS_EXT = {
  61         '5': 'flv',
  62         '6': 'flv',
  63         '13': '3gp',
  64         '17': '3gp',
  65         '18': 'mp4',
  66         '22': 'mp4',
  67         '34': 'flv',
  68         '35': 'flv',
  69         '36': '3gp',
  70         '37': 'mp4',
  71         '38': 'mp4',
  72         '43': 'webm',
  73         '44': 'webm',
  74         '45': 'webm',
  75         '46': 'webm',
  76         '59': 'mp4',
  77     }
  78     _BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext'
  79     _CAPTIONS_ENTRY_TAG = {
  80         'subtitles': 'track',
  81         'automatic_captions': 'target',
  82     }
  83     _caption_formats_ext = []
  84     _captions_xml = None
  85
  86     @classmethod
  87     def _extract_embed_urls(cls, url, webpage):
  88         mobj = re.search(
  89             r'<iframe[^>]+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P<id>[a-zA-Z0-9_-]{28,})',
  90             webpage)
  91         if mobj:
  92             yield 'https://drive.google.com/file/d/%s' % mobj.group('id')
  93
  94     def _download_subtitles_xml(self, video_id, subtitles_id, hl):
  95         if self._captions_xml:
  96             return
  97         self._captions_xml = self._download_xml(
  98             self._BASE_URL_CAPTIONS, video_id, query={
  99                 'id': video_id,
 100                 'vid': subtitles_id,
 101                 'hl': hl,
 102                 'v': video_id,
 103                 'type': 'list',
 104                 'tlangs': '1',
 105                 'fmts': '1',
 106                 'vssids': '1',
 107             }, note='Downloading subtitles XML',
 108             errnote='Unable to download subtitles XML', fatal=False)
 109         if self._captions_xml:
 110             for f in self._captions_xml.findall('format'):
 111                 if f.attrib.get('fmt_code') and not f.attrib.get('default'):
 112                     self._caption_formats_ext.append(f.attrib['fmt_code'])
 113
 114     def _get_captions_by_type(self, video_id, subtitles_id, caption_type,
 115                               origin_lang_code=None):
 116         if not subtitles_id or not caption_type:
 117             return
 118         captions = {}
 119         for caption_entry in self._captions_xml.findall(
 120                 self._CAPTIONS_ENTRY_TAG[caption_type]):
 121             caption_lang_code = caption_entry.attrib.get('lang_code')
 122             if not caption_lang_code:
 123                 continue
 124             caption_format_data = []
 125             for caption_format in self._caption_formats_ext:
 126                 query = {
 127                     'vid': subtitles_id,
 128                     'v': video_id,
 129                     'fmt': caption_format,
 130                     'lang': (caption_lang_code if origin_lang_code is None
 131                              else origin_lang_code),
 132                     'type': 'track',
 133                     'name': '',
 134                     'kind': '',
 135                 }
 136                 if origin_lang_code is not None:
 137                     query.update({'tlang': caption_lang_code})
 138                 caption_format_data.append({
 139                     'url': update_url_query(self._BASE_URL_CAPTIONS, query),
 140                     'ext': caption_format,
 141                 })
 142             captions[caption_lang_code] = caption_format_data
 143         return captions
 144
 145     def _get_subtitles(self, video_id, subtitles_id, hl):
 146         if not subtitles_id or not hl:
 147             return
 148         self._download_subtitles_xml(video_id, subtitles_id, hl)
 149         if not self._captions_xml:
 150             return
 151         return self._get_captions_by_type(video_id, subtitles_id, 'subtitles')
 152
 153     def _get_automatic_captions(self, video_id, subtitles_id, hl):
 154         if not subtitles_id or not hl:
 155             return
 156         self._download_subtitles_xml(video_id, subtitles_id, hl)
 157         if not self._captions_xml:
 158             return
 159         track = self._captions_xml.find('track')
 160         if track is None:
 161             return
 162         origin_lang_code = track.attrib.get('lang_code')
 163         if not origin_lang_code:
 164             return
 165         return self._get_captions_by_type(
 166             video_id, subtitles_id, 'automatic_captions', origin_lang_code)
 167
 168     def _real_extract(self, url):
 169         video_id = self._match_id(url)
 170         video_info = compat_parse_qs(self._download_webpage(
 171             'https://drive.google.com/get_video_info',
 172             video_id, 'Downloading video webpage', query={'docid': video_id}))
 173
 174         def get_value(key):
 175             return try_get(video_info, lambda x: x[key][0])
 176
 177         reason = get_value('reason')
 178         title = get_value('title')
 179
 180         formats = []
 181         fmt_stream_map = (get_value('fmt_stream_map') or '').split(',')
 182         fmt_list = (get_value('fmt_list') or '').split(',')
 183         if fmt_stream_map and fmt_list:
 184             resolutions = {}
 185             for fmt in fmt_list:
 186                 mobj = re.search(
 187                     r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
 188                 if mobj:
 189                     resolutions[mobj.group('format_id')] = (
 190                         int(mobj.group('width')), int(mobj.group('height')))
 191
 192             for fmt_stream in fmt_stream_map:
 193                 fmt_stream_split = fmt_stream.split('|')
 194                 if len(fmt_stream_split) < 2:
 195                     continue
 196                 format_id, format_url = fmt_stream_split[:2]
 197                 f = {
 198                     'url': lowercase_escape(format_url),
 199                     'format_id': format_id,
 200                     'ext': self._FORMATS_EXT[format_id],
 201                 }
 202                 resolution = resolutions.get(format_id)
 203                 if resolution:
 204                     f.update({
 205                         'width': resolution[0],
 206                         'height': resolution[1],
 207                     })
 208                 formats.append(f)
 209
 210         source_url = update_url_query(
 211             'https://drive.usercontent.google.com/download', {
 212                 'id': video_id,
 213                 'export': 'download',
 214                 'confirm': 't',
 215             })
 216
 217         def request_source_file(source_url, kind, data=None):
 218             return self._request_webpage(
 219                 source_url, video_id, note='Requesting %s file' % kind,
 220                 errnote='Unable to request %s file' % kind, fatal=False, data=data)
 221         urlh = request_source_file(source_url, 'source')
 222         if urlh:
 223             def add_source_format(urlh):
 224                 nonlocal title
 225                 if not title:
 226                     title = self._search_regex(
 227                         r'\bfilename="([^"]+)"', urlh.headers.get('Content-Disposition'),
 228                         'title', default=None)
 229                 formats.append({
 230                     # Use redirect URLs as download URLs in order to calculate
 231                     # correct cookies in _calc_cookies.
 232                     # Using original URLs may result in redirect loop due to
 233                     # google.com's cookies mistakenly used for googleusercontent.com
 234                     # redirect URLs (see #23919).
 235                     'url': urlh.url,
 236                     'ext': determine_ext(title, 'mp4').lower(),
 237                     'format_id': 'source',
 238                     'quality': 1,
 239                 })
 240             if urlh.headers.get('Content-Disposition'):
 241                 add_source_format(urlh)
 242             else:
 243                 confirmation_webpage = self._webpage_read_content(
 244                     urlh, url, video_id, note='Downloading confirmation page',
 245                     errnote='Unable to confirm download', fatal=False)
 246                 if confirmation_webpage:
 247                     confirmed_source_url = extract_attributes(
 248                         get_element_html_by_id('download-form', confirmation_webpage) or '').get('action')
 249                     if confirmed_source_url:
 250                         urlh = request_source_file(confirmed_source_url, 'confirmed source', data=b'')
 251                         if urlh and urlh.headers.get('Content-Disposition'):
 252                             add_source_format(urlh)
 253                     else:
 254                         self.report_warning(
 255                             get_element_by_class('uc-error-subcaption', confirmation_webpage)
 256                             or get_element_by_class('uc-error-caption', confirmation_webpage)
 257                             or 'unable to extract confirmation code')
 258
 259         if not formats and reason:
 260             if title:
 261                 self.raise_no_formats(reason, expected=True)
 262             else:
 263                 raise ExtractorError(reason, expected=True)
 264
 265         hl = get_value('hl')
 266         subtitles_id = None
 267         ttsurl = get_value('ttsurl')
 268         if ttsurl:
 269             # the video Id for subtitles will be the last value in the ttsurl
 270             # query string
 271             subtitles_id = ttsurl.encode('utf-8').decode(
 272                 'unicode_escape').split('=')[-1]
 273
 274         self.cookiejar.clear(domain='.google.com', path='/', name='NID')
 275
 276         return {
 277             'id': video_id,
 278             'title': title,
 279             'thumbnail': 'https://drive.google.com/thumbnail?id=' + video_id,
 280             'duration': int_or_none(get_value('length_seconds')),
 281             'formats': formats,
 282             'subtitles': self.extract_subtitles(video_id, subtitles_id, hl),
 283             'automatic_captions': self.extract_automatic_captions(
 284                 video_id, subtitles_id, hl),
 285         }
 286
 287
 288 class GoogleDriveFolderIE(InfoExtractor):
 289     IE_NAME = 'GoogleDrive:Folder'
 290     _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/folders/(?P<id>[\w-]{28,})'
 291     _TESTS = [{
 292         'url': 'https://drive.google.com/drive/folders/1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
 293         'info_dict': {
 294             'id': '1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
 295             'title': 'Forrest'
 296         },
 297         'playlist_count': 3,
 298     }]
 299     _BOUNDARY = '=====vc17a3rwnndj====='
 300     _REQUEST = "/drive/v2beta/files?openDrive=true&reason=102&syncType=0&errorRecovery=false&q=trashed%20%3D%20false%20and%20'{folder_id}'%20in%20parents&fields=kind%2CnextPageToken%2Citems(kind%2CmodifiedDate%2CmodifiedByMeDate%2ClastViewedByMeDate%2CfileSize%2Cowners(kind%2CpermissionId%2Cid)%2ClastModifyingUser(kind%2CpermissionId%2Cid)%2ChasThumbnail%2CthumbnailVersion%2Ctitle%2Cid%2CresourceKey%2Cshared%2CsharedWithMeDate%2CuserPermission(role)%2CexplicitlyTrashed%2CmimeType%2CquotaBytesUsed%2Ccopyable%2CfileExtension%2CsharingUser(kind%2CpermissionId%2Cid)%2Cspaces%2Cversion%2CteamDriveId%2ChasAugmentedPermissions%2CcreatedDate%2CtrashingUser(kind%2CpermissionId%2Cid)%2CtrashedDate%2Cparents(id)%2CshortcutDetails(targetId%2CtargetMimeType%2CtargetLookupStatus)%2Ccapabilities(canCopy%2CcanDownload%2CcanEdit%2CcanAddChildren%2CcanDelete%2CcanRemoveChildren%2CcanShare%2CcanTrash%2CcanRename%2CcanReadTeamDrive%2CcanMoveTeamDriveItem)%2Clabels(starred%2Ctrashed%2Crestricted%2Cviewed))%2CincompleteSearch&appDataFilter=NO_APP_DATA&spaces=drive&pageToken={page_token}&maxResults=50&supportsTeamDrives=true&includeItemsFromAllDrives=true&corpora=default&orderBy=folder%2Ctitle_natural%20asc&retryCount=0&key={key} HTTP/1.1"
 301     _DATA = f'''--{_BOUNDARY}
 302 content-type: application/http
 303 content-transfer-encoding: binary
 304
 305 GET %s
 306
 307 --{_BOUNDARY}
 308 '''
 309
 310     def _call_api(self, folder_id, key, data, **kwargs):
 311         response = self._download_webpage(
 312             'https://clients6.google.com/batch/drive/v2beta',
 313             folder_id, data=data.encode('utf-8'),
 314             headers={
 315                 'Content-Type': 'text/plain;charset=UTF-8;',
 316                 'Origin': 'https://drive.google.com',
 317             }, query={
 318                 '$ct': f'multipart/mixed; boundary="{self._BOUNDARY}"',
 319                 'key': key
 320             }, **kwargs)
 321         return self._search_json('', response, 'api response', folder_id, **kwargs) or {}
 322
 323     def _get_folder_items(self, folder_id, key):
 324         page_token = ''
 325         while page_token is not None:
 326             request = self._REQUEST.format(folder_id=folder_id, page_token=page_token, key=key)
 327             page = self._call_api(folder_id, key, self._DATA % request)
 328             yield from page['items']
 329             page_token = page.get('nextPageToken')
 330
 331     def _real_extract(self, url):
 332         folder_id = self._match_id(url)
 333
 334         webpage = self._download_webpage(url, folder_id)
 335         key = self._search_regex(r'"(\w{39})"', webpage, 'key')
 336
 337         folder_info = self._call_api(folder_id, key, self._DATA % f'/drive/v2beta/files/{folder_id} HTTP/1.1', fatal=False)
 338
 339         return self.playlist_from_matches(
 340             self._get_folder_items(folder_id, key), folder_id, folder_info.get('title'),
 341             ie=GoogleDriveIE, getter=lambda item: f'https://drive.google.com/file/d/{item["id"]}')