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,
  10     update_url_query,
  11     url_basename,
  12     urlencode_postdata,
  13 )
  14
  15
  16 class DropboxIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?dropbox\.com/(?:(?:e/)?scl/fi|sh?)/(?P<id>\w+)'
  18     _TESTS = [
  19         {
  20             'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh/youtube-dl%20test%20video%20%27%C3%A4%22BaW_jenozKc.mp4?dl=0',
  21             'info_dict': {
  22                 'id': 'nelirfsxnmcfbfh',
  23                 'ext': 'mp4',
  24                 'title': 'youtube-dl test video \'ä"BaW_jenozKc',
  25             },
  26         }, {
  27             'url': 'https://www.dropbox.com/s/nelirfsxnmcfbfh',
  28             'only_matching': True,
  29         }, {
  30             'url': 'https://www.dropbox.com/sh/2mgpiuq7kv8nqdf/AABy-fW4dkydT4GmWi2mdOUDa?dl=0&preview=Drone+Shot.mp4',
  31             'only_matching': True,
  32         }, {
  33             'url': 'https://www.dropbox.com/scl/fi/r2kd2skcy5ylbbta5y1pz/DJI_0003.MP4?dl=0&rlkey=wcdgqangn7t3lnmmv6li9mu9h',
  34             'only_matching': True,
  35         }, {
  36             'url': 'https://www.dropbox.com/e/scl/fi/r2kd2skcy5ylbbta5y1pz/DJI_0003.MP4?dl=0&rlkey=wcdgqangn7t3lnmmv6li9mu9h',
  37             'only_matching': True,
  38         },
  39     ]
  40
  41     def _yield_decoded_parts(self, webpage):
  42         for encoded in reversed(re.findall(r'registerStreamedPrefetch\s*\(\s*"[\w/+=]+"\s*,\s*"([\w/+=]+)"', webpage)):
  43             yield base64.b64decode(encoded).decode('utf-8', 'ignore')
  44
  45     def _real_extract(self, url):
  46         mobj = self._match_valid_url(url)
  47         video_id = mobj.group('id')
  48         webpage = self._download_webpage(url, video_id)
  49         fn = urllib.parse.unquote(url_basename(url))
  50         title = os.path.splitext(fn)[0]
  51         content_id = None
  52
  53         for part in self._yield_decoded_parts(webpage):
  54             if '/sm/password' in part:
  55                 content_id = self._search_regex(r'content_id=([\w.+=/-]+)', part, 'content ID')
  56                 break
  57
  58         if content_id:
  59             password = self.get_param('videopassword')
  60             if not password:
  61                 raise ExtractorError('Password protected video, use --video-password <password>', expected=True)
  62
  63             response = self._download_json(
  64                 'https://www.dropbox.com/sm/auth', video_id, 'POSTing video password',
  65                 data=urlencode_postdata({
  66                     'is_xhr': 'true',
  67                     't': self._get_cookies('https://www.dropbox.com')['t'].value,
  68                     'content_id': content_id,
  69                     'password': password,
  70                     'url': update_url(url, scheme='', netloc=''),
  71                 }))
  72             if response.get('status') != 'authed':
  73                 raise ExtractorError('Invalid password', expected=True)
  74
  75             webpage = self._download_webpage(url, video_id)
  76
  77         formats, subtitles = [], {}
  78         has_anonymous_download = False
  79         thumbnail = None
  80         for part in self._yield_decoded_parts(webpage):
  81             if not has_anonymous_download:
  82                 has_anonymous_download = self._search_regex(
  83                     r'(anonymous:\tanonymous)', part, 'anonymous', default=False)
  84             transcode_url = self._search_regex(
  85                 r'\n.(https://[^\x03\x08\x12\n]+\.m3u8)', part, 'transcode url', default=None)
  86             if not transcode_url:
  87                 continue
  88             formats, subtitles = self._extract_m3u8_formats_and_subtitles(transcode_url, video_id, 'mp4')
  89             thumbnail = self._search_regex(
  90                 r'(https://www\.dropbox\.com/temp_thumb_from_token/[\w/?&=]+)', part, 'thumbnail', default=None)
  91             break
  92
  93         # downloads enabled we can get the original file
  94         if has_anonymous_download:
  95             formats.append({
  96                 'url': update_url_query(url, {'dl': '1'}),
  97                 'format_id': 'original',
  98                 'format_note': 'Original',
  99                 'quality': 1,
 100             })
 101
 102         return {
 103             'id': video_id,
 104             'title': title,
 105             'formats': formats,
 106             'subtitles': subtitles,
 107             'thumbnail': thumbnail,
 108         }