yt_dlp/extractor/zoom.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     str_or_none,
   6     js_to_json,
   7     parse_filesize,
   8     traverse_obj,
   9     urlencode_postdata,
  10     urljoin,
  11 )
  12
  13
  14 class ZoomIE(InfoExtractor):
  15     IE_NAME = 'zoom'
  16     _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom\.us/)rec(?:ording)?/(?P<type>play|share)/(?P<id>[\w.-]+)'
  17     _TESTS = [{
  18         'url': 'https://economist.zoom.us/rec/play/dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  19         'md5': 'ab445e8c911fddc4f9adc842c2c5d434',
  20         'info_dict': {
  21             'id': 'dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  22             'ext': 'mp4',
  23             'title': 'China\'s "two sessions" and the new five-year plan',
  24         },
  25         'skip': 'Recording requires email authentication to access',
  26     }, {
  27         # play URL
  28         'url': 'https://ffgolf.zoom.us/rec/play/qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
  29         'md5': '2c4b1c4e5213ebf9db293e88d9385bee',
  30         'info_dict': {
  31             'id': 'qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
  32             'ext': 'mp4',
  33             'title': 'Prépa AF2023 - Séance 5 du 11 avril - R20/VM/GO',
  34         },
  35     }, {
  36         # share URL
  37         'url': 'https://us02web.zoom.us/rec/share/hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
  38         'md5': '90fdc7cfcaee5d52d1c817fc03c43c9b',
  39         'info_dict': {
  40             'id': 'hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
  41             'ext': 'mp4',
  42             'title': 'Timea Andrea Lelik\'s Personal Meeting Room',
  43         },
  44     }]
  45
  46     def _get_page_data(self, webpage, video_id):
  47         return self._search_json(
  48             r'window\.__data__\s*=', webpage, 'data', video_id, transform_source=js_to_json)
  49
  50     def _get_real_webpage(self, url, base_url, video_id, url_type):
  51         webpage = self._download_webpage(url, video_id, note=f'Downloading {url_type} webpage')
  52         try:
  53             form = self._form_hidden_inputs('password_form', webpage)
  54         except ExtractorError:
  55             return webpage
  56
  57         password = self.get_param('videopassword')
  58         if not password:
  59             raise ExtractorError(
  60                 'This video is protected by a passcode, use the --video-password option', expected=True)
  61         is_meeting = form.get('useWhichPasswd') == 'meeting'
  62         validation = self._download_json(
  63             base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''),
  64             video_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({
  65                 'id': form[('meet' if is_meeting else 'file') + 'Id'],
  66                 'passwd': password,
  67                 'action': form.get('action'),
  68             }))
  69         if not validation.get('status'):
  70             raise ExtractorError(validation['errorMessage'], expected=True)
  71         return self._download_webpage(url, video_id, note=f'Re-downloading {url_type} webpage')
  72
  73     def _real_extract(self, url):
  74         base_url, url_type, video_id = self._match_valid_url(url).group('base_url', 'type', 'id')
  75
  76         if url_type == 'share':
  77             webpage = self._get_real_webpage(url, base_url, video_id, 'share')
  78             meeting_id = self._get_page_data(webpage, video_id)['meetingId']
  79             redirect_path = self._download_json(
  80                 f'{base_url}nws/recording/1.0/play/share-info/{meeting_id}',
  81                 video_id, note='Downloading share info JSON')['result']['redirectUrl']
  82             url = urljoin(base_url, redirect_path)
  83
  84         webpage = self._get_real_webpage(url, base_url, video_id, 'play')
  85         file_id = self._get_page_data(webpage, video_id)['fileId']
  86         if not file_id:
  87             # When things go wrong, file_id can be empty string
  88             raise ExtractorError('Unable to extract file ID')
  89
  90         data = self._download_json(
  91             f'{base_url}nws/recording/1.0/play/info/{file_id}', video_id,
  92             note='Downloading play info JSON')['result']
  93
  94         subtitles = {}
  95         for _type in ('transcript', 'cc', 'chapter'):
  96             if data.get('%sUrl' % _type):
  97                 subtitles[_type] = [{
  98                     'url': urljoin(base_url, data['%sUrl' % _type]),
  99                     'ext': 'vtt',
 100                 }]
 101
 102         formats = []
 103
 104         if data.get('viewMp4Url'):
 105             formats.append({
 106                 'format_note': 'Camera stream',
 107                 'url': str_or_none(data.get('viewMp4Url')),
 108                 'width': int_or_none(traverse_obj(data, ('viewResolvtions', 0))),
 109                 'height': int_or_none(traverse_obj(data, ('viewResolvtions', 1))),
 110                 'format_id': str_or_none(traverse_obj(data, ('recording', 'id'))),
 111                 'ext': 'mp4',
 112                 'filesize_approx': parse_filesize(str_or_none(traverse_obj(data, ('recording', 'fileSizeInMB')))),
 113                 'preference': 0
 114             })
 115
 116         if data.get('shareMp4Url'):
 117             formats.append({
 118                 'format_note': 'Screen share stream',
 119                 'url': str_or_none(data.get('shareMp4Url')),
 120                 'width': int_or_none(traverse_obj(data, ('shareResolvtions', 0))),
 121                 'height': int_or_none(traverse_obj(data, ('shareResolvtions', 1))),
 122                 'format_id': str_or_none(traverse_obj(data, ('shareVideo', 'id'))),
 123                 'ext': 'mp4',
 124                 'preference': -1
 125             })
 126
 127         return {
 128             'id': video_id,
 129             'title': str_or_none(traverse_obj(data, ('meet', 'topic'))),
 130             'duration': int_or_none(data.get('duration')),
 131             'subtitles': subtitles,
 132             'formats': formats,
 133             'http_headers': {
 134                 'Referer': base_url,
 135             },
 136         }