yt_dlp/extractor/zoom.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     js_to_json,
   6     parse_filesize,
   7     parse_resolution,
   8     str_or_none,
   9     traverse_obj,
  10     url_basename,
  11     urlencode_postdata,
  12     urljoin,
  13 )
  14
  15
  16 class ZoomIE(InfoExtractor):
  17     IE_NAME = 'zoom'
  18     _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom\.us/)rec(?:ording)?/(?P<type>play|share)/(?P<id>[\w.-]+)'
  19     _TESTS = [{
  20         'url': 'https://economist.zoom.us/rec/play/dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  21         'md5': 'ab445e8c911fddc4f9adc842c2c5d434',
  22         'info_dict': {
  23             'id': 'dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  24             'ext': 'mp4',
  25             'title': 'China\'s "two sessions" and the new five-year plan',
  26         },
  27         'skip': 'Recording requires email authentication to access',
  28     }, {
  29         # play URL
  30         'url': 'https://ffgolf.zoom.us/rec/play/qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
  31         'md5': '2c4b1c4e5213ebf9db293e88d9385bee',
  32         'info_dict': {
  33             'id': 'qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
  34             'ext': 'mp4',
  35             'title': 'Prépa AF2023 - Séance 5 du 11 avril - R20/VM/GO',
  36         },
  37     }, {
  38         # share URL
  39         'url': 'https://us02web.zoom.us/rec/share/hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
  40         'md5': '90fdc7cfcaee5d52d1c817fc03c43c9b',
  41         'info_dict': {
  42             'id': 'hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
  43             'ext': 'mp4',
  44             'title': 'Timea Andrea Lelik\'s Personal Meeting Room',
  45         },
  46         'skip': 'This recording has expired',
  47     }, {
  48         # view_with_share URL
  49         'url': 'https://cityofdetroit.zoom.us/rec/share/VjE-5kW3xmgbEYqR5KzRgZ1OFZvtMtiXk5HyRJo5kK4m5PYE6RF4rF_oiiO_9qaM.UTAg1MI7JSnF3ZjX',
  50         'md5': 'bdc7867a5934c151957fb81321b3c024',
  51         'info_dict': {
  52             'id': 'VjE-5kW3xmgbEYqR5KzRgZ1OFZvtMtiXk5HyRJo5kK4m5PYE6RF4rF_oiiO_9qaM.UTAg1MI7JSnF3ZjX',
  53             'ext': 'mp4',
  54             'title': 'February 2022 Detroit Revenue Estimating Conference',
  55             'duration': 7299,
  56             'formats': 'mincount:3',
  57         },
  58     }]
  59
  60     def _get_page_data(self, webpage, video_id):
  61         return self._search_json(
  62             r'window\.__data__\s*=', webpage, 'data', video_id, transform_source=js_to_json)
  63
  64     def _get_real_webpage(self, url, base_url, video_id, url_type):
  65         webpage = self._download_webpage(url, video_id, note=f'Downloading {url_type} webpage')
  66         try:
  67             form = self._form_hidden_inputs('password_form', webpage)
  68         except ExtractorError:
  69             return webpage
  70
  71         password = self.get_param('videopassword')
  72         if not password:
  73             raise ExtractorError(
  74                 'This video is protected by a passcode, use the --video-password option', expected=True)
  75         is_meeting = form.get('useWhichPasswd') == 'meeting'
  76         validation = self._download_json(
  77             base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''),
  78             video_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({
  79                 'id': form[('meet' if is_meeting else 'file') + 'Id'],
  80                 'passwd': password,
  81                 'action': form.get('action'),
  82             }))
  83         if not validation.get('status'):
  84             raise ExtractorError(validation['errorMessage'], expected=True)
  85         return self._download_webpage(url, video_id, note=f'Re-downloading {url_type} webpage')
  86
  87     def _real_extract(self, url):
  88         base_url, url_type, video_id = self._match_valid_url(url).group('base_url', 'type', 'id')
  89         query = {}
  90
  91         if url_type == 'share':
  92             webpage = self._get_real_webpage(url, base_url, video_id, 'share')
  93             meeting_id = self._get_page_data(webpage, video_id)['meetingId']
  94             redirect_path = self._download_json(
  95                 f'{base_url}nws/recording/1.0/play/share-info/{meeting_id}',
  96                 video_id, note='Downloading share info JSON')['result']['redirectUrl']
  97             url = urljoin(base_url, redirect_path)
  98             query['continueMode'] = 'true'
  99
 100         webpage = self._get_real_webpage(url, base_url, video_id, 'play')
 101         file_id = self._get_page_data(webpage, video_id)['fileId']
 102         if not file_id:
 103             # When things go wrong, file_id can be empty string
 104             raise ExtractorError('Unable to extract file ID')
 105
 106         data = self._download_json(
 107             f'{base_url}nws/recording/1.0/play/info/{file_id}', video_id, query=query,
 108             note='Downloading play info JSON')['result']
 109
 110         subtitles = {}
 111         for _type in ('transcript', 'cc', 'chapter'):
 112             if data.get('%sUrl' % _type):
 113                 subtitles[_type] = [{
 114                     'url': urljoin(base_url, data['%sUrl' % _type]),
 115                     'ext': 'vtt',
 116                 }]
 117
 118         formats = []
 119
 120         if data.get('viewMp4Url'):
 121             formats.append({
 122                 'format_note': 'Camera stream',
 123                 'url': data['viewMp4Url'],
 124                 'width': int_or_none(traverse_obj(data, ('viewResolvtions', 0))),
 125                 'height': int_or_none(traverse_obj(data, ('viewResolvtions', 1))),
 126                 'format_id': 'view',
 127                 'ext': 'mp4',
 128                 'filesize_approx': parse_filesize(str_or_none(traverse_obj(data, ('recording', 'fileSizeInMB')))),
 129                 'preference': 0
 130             })
 131
 132         if data.get('shareMp4Url'):
 133             formats.append({
 134                 'format_note': 'Screen share stream',
 135                 'url': data['shareMp4Url'],
 136                 'width': int_or_none(traverse_obj(data, ('shareResolvtions', 0))),
 137                 'height': int_or_none(traverse_obj(data, ('shareResolvtions', 1))),
 138                 'format_id': 'share',
 139                 'ext': 'mp4',
 140                 'preference': -1
 141             })
 142
 143         view_with_share_url = data.get('viewMp4WithshareUrl')
 144         if view_with_share_url:
 145             formats.append({
 146                 **parse_resolution(self._search_regex(
 147                     r'_(\d+x\d+)\.mp4', url_basename(view_with_share_url), 'resolution', default=None)),
 148                 'format_note': 'Screen share with camera',
 149                 'url': view_with_share_url,
 150                 'format_id': 'view_with_share',
 151                 'ext': 'mp4',
 152                 'preference': 1
 153             })
 154
 155         return {
 156             'id': video_id,
 157             'title': str_or_none(traverse_obj(data, ('meet', 'topic'))),
 158             'duration': int_or_none(data.get('duration')),
 159             'subtitles': subtitles,
 160             'formats': formats,
 161             'http_headers': {
 162                 'Referer': base_url,
 163             },
 164         }