yt_dlp/extractor/dropbox.py

   1 import base64
   2 import os.path
   3 import re
   4 import urllib.parse
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     update_url_query,
  10     url_basename,
  11 )
  12
  13
  14 class DropboxIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?dropbox\.com/(?:(?:e/)?scl/fi|sh?)/(?P<id>\w+)'
  16     _TESTS = [
  17         {
  18             'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh/youtube-dl%20test%20video%20%27%C3%A4%22BaW_jenozKc.mp4?dl=0',
  19             'info_dict': {
  20                 'id': 'nelirfsxnmcfbfh',
  21                 'ext': 'mp4',
  22                 'title': 'youtube-dl test video \'ä"BaW_jenozKc',
  23             },
  24         }, {
  25             'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh',
  26             'only_matching': True,
  27         }, {
  28             'url': 'https://www.dropbox.com/sh/2mgpiuq7kv8nqdf/AABy-fW4dkydT4GmWi2mdOUDa?dl=0&preview=Drone+Shot.mp4',
  29             'only_matching': True,
  30         }, {
  31             'url': 'https://www.dropbox.com/scl/fi/r2kd2skcy5ylbbta5y1pz/DJI_0003.MP4?dl=0&rlkey=wcdgqangn7t3lnmmv6li9mu9h',
  32             'only_matching': True,
  33         }, {
  34             'url': 'https://www.dropbox.com/e/scl/fi/r2kd2skcy5ylbbta5y1pz/DJI_0003.MP4?dl=0&rlkey=wcdgqangn7t3lnmmv6li9mu9h',
  35             'only_matching': True,
  36         },
  37     ]
  38
  39     def _real_extract(self, url):
  40         mobj = self._match_valid_url(url)
  41         video_id = mobj.group('id')
  42         webpage = self._download_webpage(url, video_id)
  43         fn = urllib.parse.unquote(url_basename(url))
  44         title = os.path.splitext(fn)[0]
  45
  46         password = self.get_param('videopassword')
  47         if (self._og_search_title(webpage) == 'Dropbox - Password Required'
  48                 or 'Enter the password for this link' in webpage):
  49
  50             if password:
  51                 content_id = self._search_regex(r'content_id=(.*?)["\']', webpage, 'content_id')
  52                 payload = f'is_xhr=true&t={self._get_cookies("https://www.dropbox.com").get("t").value}&content_id={content_id}&password={password}&url={url}'
  53                 response = self._download_json(
  54                     'https://www.dropbox.com/sm/auth', video_id, 'POSTing video password', data=payload.encode(),
  55                     headers={'content-type': 'application/x-www-form-urlencoded; charset=UTF-8'})
  56
  57                 if response.get('status') != 'authed':
  58                     raise ExtractorError('Authentication failed!', expected=True)
  59                 webpage = self._download_webpage(url, video_id)
  60             elif self._get_cookies('https://dropbox.com').get('sm_auth'):
  61                 webpage = self._download_webpage(url, video_id)
  62             else:
  63                 raise ExtractorError('Password protected video, use --video-password <password>', expected=True)
  64
  65         formats, subtitles, has_anonymous_download = [], {}, False
  66         for encoded in reversed(re.findall(r'registerStreamedPrefetch\s*\(\s*"[\w/+=]+"\s*,\s*"([\w/+=]+)"', webpage)):
  67             decoded = base64.b64decode(encoded).decode('utf-8', 'ignore')
  68             if not has_anonymous_download:
  69                 has_anonymous_download = self._search_regex(
  70                     r'(anonymous:\tanonymous)', decoded, 'anonymous', default=False)
  71             transcode_url = self._search_regex(
  72                 r'\n.(https://[^\x03\x08\x12\n]+\.m3u8)', decoded, 'transcode url', default=None)
  73             if not transcode_url:
  74                 continue
  75             formats, subtitles = self._extract_m3u8_formats_and_subtitles(transcode_url, video_id, 'mp4')
  76             break
  77
  78         # downloads enabled we can get the original file
  79         if has_anonymous_download:
  80             formats.append({
  81                 'url': update_url_query(url, {'dl': '1'}),
  82                 'format_id': 'original',
  83                 'format_note': 'Original',
  84                 'quality': 1,
  85             })
  86
  87         return {
  88             'id': video_id,
  89             'title': title,
  90             'formats': formats,
  91             'subtitles': subtitles,
  92         }