yt_dlp/extractor/googledrive.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from .youtube import YoutubeIE
   6 from ..utils import (
   7     ExtractorError,
   8     bug_reports_message,
   9     determine_ext,
  10     extract_attributes,
  11     get_element_by_class,
  12     get_element_html_by_id,
  13     int_or_none,
  14     lowercase_escape,
  15     try_get,
  16     update_url_query,
  17 )
  18
  19
  20 class GoogleDriveIE(InfoExtractor):
  21     _VALID_URL = r'''(?x)
  22                         https?://
  23                             (?:
  24                                 (?:docs|drive|drive\.usercontent)\.google\.com/
  25                                 (?:
  26                                     (?:uc|open|download)\?.*?id=|
  27                                     file/d/
  28                                 )|
  29                                 video\.google\.com/get_player\?.*?docid=
  30                             )
  31                             (?P<id>[a-zA-Z0-9_-]{28,})
  32                     '''
  33     _TESTS = [{
  34         'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1',
  35         'md5': '5c602afbbf2c1db91831f5d82f678554',
  36         'info_dict': {
  37             'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  38             'ext': 'mp4',
  39             'title': 'Big Buck Bunny.mp4',
  40             'duration': 45,
  41             'thumbnail': 'https://drive.google.com/thumbnail?id=0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  42         },
  43     }, {
  44         # has itag 50 which is not in YoutubeIE._formats (royalty Free music from 1922)
  45         'url': 'https://drive.google.com/uc?id=1IP0o8dHcQrIHGgVyp0Ofvx2cGfLzyO1x',
  46         'md5': '322db8d63dd19788c04050a4bba67073',
  47         'info_dict': {
  48             'id': '1IP0o8dHcQrIHGgVyp0Ofvx2cGfLzyO1x',
  49             'ext': 'mp3',
  50             'title': 'My Buddy - Henry Burr - Gus Kahn - Walter Donaldson.mp3',
  51             'duration': 184,
  52             'thumbnail': 'https://drive.google.com/thumbnail?id=1IP0o8dHcQrIHGgVyp0Ofvx2cGfLzyO1x',
  53         },
  54     }, {
  55         # video can't be watched anonymously due to view count limit reached,
  56         # but can be downloaded (see https://github.com/ytdl-org/youtube-dl/issues/14046)
  57         'url': 'https://drive.google.com/file/d/0B-vUyvmDLdWDcEt4WjBqcmI2XzQ/view',
  58         'only_matching': True,
  59     }, {
  60         # video id is longer than 28 characters
  61         'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit',
  62         'only_matching': True,
  63     }, {
  64         'url': 'https://drive.google.com/open?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  65         'only_matching': True,
  66     }, {
  67         'url': 'https://drive.google.com/uc?id=0B2fjwgkl1A_CX083Tkowdmt6d28',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://drive.usercontent.google.com/download?id=0ByeS4oOUV-49Zzh4R1J6R09zazQ',
  71         'only_matching': True,
  72     }]
  73     _FORMATS_EXT = {
  74         **{k: v['ext'] for k, v in YoutubeIE._formats.items() if v.get('ext')},
  75         '50': 'm4a',
  76     }
  77     _BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext'
  78     _CAPTIONS_ENTRY_TAG = {
  79         'subtitles': 'track',
  80         'automatic_captions': 'target',
  81     }
  82     _caption_formats_ext = []
  83     _captions_xml = None
  84
  85     @classmethod
  86     def _extract_embed_urls(cls, url, webpage):
  87         mobj = re.search(
  88             r'<iframe[^>]+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P<id>[a-zA-Z0-9_-]{28,})',
  89             webpage)
  90         if mobj:
  91             yield 'https://drive.google.com/file/d/{}'.format(mobj.group('id'))
  92
  93     def _download_subtitles_xml(self, video_id, subtitles_id, hl):
  94         if self._captions_xml:
  95             return
  96         self._captions_xml = self._download_xml(
  97             self._BASE_URL_CAPTIONS, video_id, query={
  98                 'id': video_id,
  99                 'vid': subtitles_id,
 100                 'hl': hl,
 101                 'v': video_id,
 102                 'type': 'list',
 103                 'tlangs': '1',
 104                 'fmts': '1',
 105                 'vssids': '1',
 106             }, note='Downloading subtitles XML',
 107             errnote='Unable to download subtitles XML', fatal=False)
 108         if self._captions_xml:
 109             for f in self._captions_xml.findall('format'):
 110                 if f.attrib.get('fmt_code') and not f.attrib.get('default'):
 111                     self._caption_formats_ext.append(f.attrib['fmt_code'])
 112
 113     def _get_captions_by_type(self, video_id, subtitles_id, caption_type,
 114                               origin_lang_code=None):
 115         if not subtitles_id or not caption_type:
 116             return
 117         captions = {}
 118         for caption_entry in self._captions_xml.findall(
 119                 self._CAPTIONS_ENTRY_TAG[caption_type]):
 120             caption_lang_code = caption_entry.attrib.get('lang_code')
 121             if not caption_lang_code:
 122                 continue
 123             caption_format_data = []
 124             for caption_format in self._caption_formats_ext:
 125                 query = {
 126                     'vid': subtitles_id,
 127                     'v': video_id,
 128                     'fmt': caption_format,
 129                     'lang': (caption_lang_code if origin_lang_code is None
 130                              else origin_lang_code),
 131                     'type': 'track',
 132                     'name': '',
 133                     'kind': '',
 134                 }
 135                 if origin_lang_code is not None:
 136                     query.update({'tlang': caption_lang_code})
 137                 caption_format_data.append({
 138                     'url': update_url_query(self._BASE_URL_CAPTIONS, query),
 139                     'ext': caption_format,
 140                 })
 141             captions[caption_lang_code] = caption_format_data
 142         return captions
 143
 144     def _get_subtitles(self, video_id, subtitles_id, hl):
 145         if not subtitles_id or not hl:
 146             return
 147         self._download_subtitles_xml(video_id, subtitles_id, hl)
 148         if not self._captions_xml:
 149             return
 150         return self._get_captions_by_type(video_id, subtitles_id, 'subtitles')
 151
 152     def _get_automatic_captions(self, video_id, subtitles_id, hl):
 153         if not subtitles_id or not hl:
 154             return
 155         self._download_subtitles_xml(video_id, subtitles_id, hl)
 156         if not self._captions_xml:
 157             return
 158         track = self._captions_xml.find('track')
 159         if track is None:
 160             return
 161         origin_lang_code = track.attrib.get('lang_code')
 162         if not origin_lang_code:
 163             return
 164         return self._get_captions_by_type(
 165             video_id, subtitles_id, 'automatic_captions', origin_lang_code)
 166
 167     def _real_extract(self, url):
 168         video_id = self._match_id(url)
 169         video_info = urllib.parse.parse_qs(self._download_webpage(
 170             'https://drive.google.com/get_video_info',
 171             video_id, 'Downloading video webpage', query={'docid': video_id}))
 172
 173         def get_value(key):
 174             return try_get(video_info, lambda x: x[key][0])
 175
 176         reason = get_value('reason')
 177         title = get_value('title')
 178
 179         formats = []
 180         fmt_stream_map = (get_value('fmt_stream_map') or '').split(',')
 181         fmt_list = (get_value('fmt_list') or '').split(',')
 182         if fmt_stream_map and fmt_list:
 183             resolutions = {}
 184             for fmt in fmt_list:
 185                 mobj = re.search(
 186                     r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
 187                 if mobj:
 188                     resolutions[mobj.group('format_id')] = (
 189                         int(mobj.group('width')), int(mobj.group('height')))
 190
 191             for fmt_stream in fmt_stream_map:
 192                 fmt_stream_split = fmt_stream.split('|')
 193                 if len(fmt_stream_split) < 2:
 194                     continue
 195                 format_id, format_url = fmt_stream_split[:2]
 196                 ext = self._FORMATS_EXT.get(format_id)
 197                 if not ext:
 198                     self.report_warning(f'Unknown format {format_id}{bug_reports_message()}')
 199                 f = {
 200                     'url': lowercase_escape(format_url),
 201                     'format_id': format_id,
 202                     'ext': ext,
 203                 }
 204                 resolution = resolutions.get(format_id)
 205                 if resolution:
 206                     f.update({
 207                         'width': resolution[0],
 208                         'height': resolution[1],
 209                     })
 210                 formats.append(f)
 211
 212         source_url = update_url_query(
 213             'https://drive.usercontent.google.com/download', {
 214                 'id': video_id,
 215                 'export': 'download',
 216                 'confirm': 't',
 217             })
 218
 219         def request_source_file(source_url, kind, data=None):
 220             return self._request_webpage(
 221                 source_url, video_id, note=f'Requesting {kind} file',
 222                 errnote=f'Unable to request {kind} file', fatal=False, data=data)
 223         urlh = request_source_file(source_url, 'source')
 224         if urlh:
 225             def add_source_format(urlh):
 226                 nonlocal title
 227                 if not title:
 228                     title = self._search_regex(
 229                         r'\bfilename="([^"]+)"', urlh.headers.get('Content-Disposition'),
 230                         'title', default=None)
 231                 formats.append({
 232                     # Use redirect URLs as download URLs in order to calculate
 233                     # correct cookies in _calc_cookies.
 234                     # Using original URLs may result in redirect loop due to
 235                     # google.com's cookies mistakenly used for googleusercontent.com
 236                     # redirect URLs (see #23919).
 237                     'url': urlh.url,
 238                     'ext': determine_ext(title, 'mp4').lower(),
 239                     'format_id': 'source',
 240                     'quality': 1,
 241                 })
 242             if urlh.headers.get('Content-Disposition'):
 243                 add_source_format(urlh)
 244             else:
 245                 confirmation_webpage = self._webpage_read_content(
 246                     urlh, url, video_id, note='Downloading confirmation page',
 247                     errnote='Unable to confirm download', fatal=False)
 248                 if confirmation_webpage:
 249                     confirmed_source_url = extract_attributes(
 250                         get_element_html_by_id('download-form', confirmation_webpage) or '').get('action')
 251                     if confirmed_source_url:
 252                         urlh = request_source_file(confirmed_source_url, 'confirmed source', data=b'')
 253                         if urlh and urlh.headers.get('Content-Disposition'):
 254                             add_source_format(urlh)
 255                     else:
 256                         self.report_warning(
 257                             get_element_by_class('uc-error-subcaption', confirmation_webpage)
 258                             or get_element_by_class('uc-error-caption', confirmation_webpage)
 259                             or 'unable to extract confirmation code')
 260
 261         if not formats and reason:
 262             if title:
 263                 self.raise_no_formats(reason, expected=True)
 264             else:
 265                 raise ExtractorError(reason, expected=True)
 266
 267         hl = get_value('hl')
 268         subtitles_id = None
 269         ttsurl = get_value('ttsurl')
 270         if ttsurl:
 271             # the video Id for subtitles will be the last value in the ttsurl
 272             # query string
 273             subtitles_id = ttsurl.encode().decode(
 274                 'unicode_escape').split('=')[-1]
 275
 276         self.cookiejar.clear(domain='.google.com', path='/', name='NID')
 277
 278         return {
 279             'id': video_id,
 280             'title': title,
 281             'thumbnail': 'https://drive.google.com/thumbnail?id=' + video_id,
 282             'duration': int_or_none(get_value('length_seconds')),
 283             'formats': formats,
 284             'subtitles': self.extract_subtitles(video_id, subtitles_id, hl),
 285             'automatic_captions': self.extract_automatic_captions(
 286                 video_id, subtitles_id, hl),
 287         }
 288
 289
 290 class GoogleDriveFolderIE(InfoExtractor):
 291     IE_NAME = 'GoogleDrive:Folder'
 292     _VALID_URL = r'https?://(?:docs|drive)\.google\.com/drive/folders/(?P<id>[\w-]{28,})'
 293     _TESTS = [{
 294         'url': 'https://drive.google.com/drive/folders/1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
 295         'info_dict': {
 296             'id': '1dQ4sx0-__Nvg65rxTSgQrl7VyW_FZ9QI',
 297             'title': 'Forrest',
 298         },
 299         'playlist_count': 3,
 300     }]
 301     _BOUNDARY = '=====vc17a3rwnndj====='
 302     _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"
 303     _DATA = f'''--{_BOUNDARY}
 304 content-type: application/http
 305 content-transfer-encoding: binary
 306
 307 GET %s
 308
 309 --{_BOUNDARY}
 310 '''
 311
 312     def _call_api(self, folder_id, key, data, **kwargs):
 313         response = self._download_webpage(
 314             'https://clients6.google.com/batch/drive/v2beta',
 315             folder_id, data=data.encode(),
 316             headers={
 317                 'Content-Type': 'text/plain;charset=UTF-8;',
 318                 'Origin': 'https://drive.google.com',
 319             }, query={
 320                 '$ct': f'multipart/mixed; boundary="{self._BOUNDARY}"',
 321                 'key': key,
 322             }, **kwargs)
 323         return self._search_json('', response, 'api response', folder_id, **kwargs) or {}
 324
 325     def _get_folder_items(self, folder_id, key):
 326         page_token = ''
 327         while page_token is not None:
 328             request = self._REQUEST.format(folder_id=folder_id, page_token=page_token, key=key)
 329             page = self._call_api(folder_id, key, self._DATA % request)
 330             yield from page['items']
 331             page_token = page.get('nextPageToken')
 332
 333     def _real_extract(self, url):
 334         folder_id = self._match_id(url)
 335
 336         webpage = self._download_webpage(url, folder_id)
 337         key = self._search_regex(r'"(\w{39})"', webpage, 'key')
 338
 339         folder_info = self._call_api(folder_id, key, self._DATA % f'/drive/v2beta/files/{folder_id} HTTP/1.1', fatal=False)
 340
 341         return self.playlist_from_matches(
 342             self._get_folder_items(folder_id, key), folder_id, folder_info.get('title'),
 343             ie=GoogleDriveIE, getter=lambda item: f'https://drive.google.com/file/d/{item["id"]}')