yt_dlp/extractor/zoom.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     int_or_none,
   9     str_or_none,
  10     js_to_json,
  11     parse_filesize,
  12     urlencode_postdata,
  13     urljoin,
  14 )
  15
  16
  17 class ZoomIE(InfoExtractor):
  18     IE_NAME = 'zoom'
  19     _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom.us/)rec(?:ording)?/(?:play|share)/(?P<id>[A-Za-z0-9_.-]+)'
  20     _TEST = {
  21         'url': 'https://economist.zoom.us/rec/play/dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  22         'md5': 'ab445e8c911fddc4f9adc842c2c5d434',
  23         'info_dict': {
  24             'id': 'dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
  25             'ext': 'mp4',
  26             'title': 'China\'s "two sessions" and the new five-year plan',
  27         },
  28         'skip': 'Recording requires email authentication to access',
  29     }
  30
  31     def _real_extract(self, url):
  32         base_url, play_id = self._match_valid_url(url).groups()
  33         webpage = self._download_webpage(url, play_id)
  34
  35         try:
  36             form = self._form_hidden_inputs('password_form', webpage)
  37         except ExtractorError:
  38             form = None
  39         if form:
  40             password = self.get_param('videopassword')
  41             if not password:
  42                 raise ExtractorError(
  43                     'This video is protected by a passcode, use the --video-password option', expected=True)
  44             is_meeting = form.get('useWhichPasswd') == 'meeting'
  45             validation = self._download_json(
  46                 base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''),
  47                 play_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({
  48                     'id': form[('meet' if is_meeting else 'file') + 'Id'],
  49                     'passwd': password,
  50                     'action': form.get('action'),
  51                 }))
  52             if not validation.get('status'):
  53                 raise ExtractorError(validation['errorMessage'], expected=True)
  54             webpage = self._download_webpage(url, play_id)
  55
  56         data = self._parse_json(self._search_regex(
  57             r'(?s)window\.__data__\s*=\s*({.+?});',
  58             webpage, 'data'), play_id, js_to_json)
  59
  60         subtitles = {}
  61         for _type in ('transcript', 'cc', 'chapter'):
  62             if data.get('%sUrl' % _type):
  63                 subtitles[_type] = [{
  64                     'url': urljoin(base_url, data['%sUrl' % _type]),
  65                     'ext': 'vtt',
  66                 }]
  67
  68         formats = []
  69
  70         if data.get('viewMp4Url'):
  71             formats.append({
  72                 'format_note': 'Camera stream',
  73                 'url': str_or_none(data.get('viewMp4Url')),
  74                 'width': int_or_none(data.get('viewResolvtionsWidth')),
  75                 'height': int_or_none(data.get('viewResolvtionsHeight')),
  76                 'format_id': str_or_none(data.get('recordingId')),
  77                 'ext': 'mp4',
  78                 'filesize_approx': parse_filesize(data.get('fileSize')),
  79                 'preference': 0
  80             })
  81
  82         if data.get('shareMp4Url'):
  83             formats.append({
  84                 'format_note': 'Screen share stream',
  85                 'url': str_or_none(data.get('shareMp4Url')),
  86                 'width': int_or_none(data.get('shareResolvtionsWidth')),
  87                 'height': int_or_none(data.get('shareResolvtionsHeight')),
  88                 'format_id': str_or_none(data.get('shareVideoId')),
  89                 'ext': 'mp4',
  90                 'preference': -1
  91             })
  92
  93         self._sort_formats(formats)
  94
  95         return {
  96             'id': play_id,
  97             'title': data.get('topic'),
  98             'subtitles': subtitles,
  99             'formats': formats,
 100             'http_headers': {
 101                 'Referer': base_url,
 102             },
 103         }