From 63da31b3b29af90062d8a72a905ffe4b5e499042 Mon Sep 17 00:00:00 2001 From: ndyanx <114776171+ndyanx@users.noreply.github.com> Date: Fri, 27 Sep 2024 17:05:22 -0500 Subject: [PATCH] [ie/dropbox] Fix password-protected video support (#10735) Also adds thumbnail extraction Closes #9864 Authored by: ndyanx --- yt_dlp/extractor/dropbox.py | 52 +++++++++++++++++++++++++------------ 1 file changed, 35 insertions(+), 17 deletions(-) diff --git a/yt_dlp/extractor/dropbox.py b/yt_dlp/extractor/dropbox.py index 51b40df42..c12209623 100644 --- a/yt_dlp/extractor/dropbox.py +++ b/yt_dlp/extractor/dropbox.py @@ -6,8 +6,10 @@ import urllib.parse from .common import InfoExtractor from ..utils import ( ExtractorError, + update_url, update_url_query, url_basename, + urlencode_postdata, ) @@ -36,43 +38,58 @@ class DropboxIE(InfoExtractor): }, ] + def _yield_decoded_parts(self, webpage): + for encoded in reversed(re.findall(r'registerStreamedPrefetch\s*\(\s*"[\w/+=]+"\s*,\s*"([\w/+=]+)"', webpage)): + yield base64.b64decode(encoded).decode('utf-8', 'ignore') + def _real_extract(self, url): mobj = self._match_valid_url(url) video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) fn = urllib.parse.unquote(url_basename(url)) title = os.path.splitext(fn)[0] - password = self.get_param('videopassword') - if (self._og_search_title(webpage) == 'Dropbox - Password Required' - or 'Enter the password for this link' in webpage): + for part in self._yield_decoded_parts(webpage): + if '/sm/password' in part: + webpage = self._download_webpage( + update_url('https://www.dropbox.com/sm/password', query=part.partition('?')[2]), video_id) + break + + if (self._og_search_title(webpage, default=None) == 'Dropbox - Password Required' + or 'Enter the password for this link' in webpage): if password: - content_id = self._search_regex(r'content_id=(.*?)["\']', webpage, 'content_id') - payload = f'is_xhr=true&t={self._get_cookies("https://www.dropbox.com").get("t").value}&content_id={content_id}&password={password}&url={url}' response = self._download_json( - 'https://www.dropbox.com/sm/auth', video_id, 'POSTing video password', data=payload.encode(), - headers={'content-type': 'application/x-www-form-urlencoded; charset=UTF-8'}) + 'https://www.dropbox.com/sm/auth', video_id, 'POSTing video password', + headers={'content-type': 'application/x-www-form-urlencoded; charset=UTF-8'}, + data=urlencode_postdata({ + 'is_xhr': 'true', + 't': self._get_cookies('https://www.dropbox.com')['t'].value, + 'content_id': self._search_regex(r'content_id=([\w.+=/-]+)["\']', webpage, 'content id'), + 'password': password, + 'url': url, + })) if response.get('status') != 'authed': - raise ExtractorError('Authentication failed!', expected=True) - webpage = self._download_webpage(url, video_id) - elif self._get_cookies('https://dropbox.com').get('sm_auth'): - webpage = self._download_webpage(url, video_id) - else: + raise ExtractorError('Invalid password', expected=True) + elif not self._get_cookies('https://dropbox.com').get('sm_auth'): raise ExtractorError('Password protected video, use --video-password ', expected=True) + webpage = self._download_webpage(url, video_id) - formats, subtitles, has_anonymous_download = [], {}, False - for encoded in reversed(re.findall(r'registerStreamedPrefetch\s*\(\s*"[\w/+=]+"\s*,\s*"([\w/+=]+)"', webpage)): - decoded = base64.b64decode(encoded).decode('utf-8', 'ignore') + formats, subtitles = [], {} + has_anonymous_download = False + thumbnail = None + for part in self._yield_decoded_parts(webpage): if not has_anonymous_download: has_anonymous_download = self._search_regex( - r'(anonymous:\tanonymous)', decoded, 'anonymous', default=False) + r'(anonymous:\tanonymous)', part, 'anonymous', default=False) transcode_url = self._search_regex( - r'\n.(https://[^\x03\x08\x12\n]+\.m3u8)', decoded, 'transcode url', default=None) + r'\n.(https://[^\x03\x08\x12\n]+\.m3u8)', part, 'transcode url', default=None) if not transcode_url: continue formats, subtitles = self._extract_m3u8_formats_and_subtitles(transcode_url, video_id, 'mp4') + thumbnail = self._search_regex( + r'(https://www\.dropbox\.com/temp_thumb_from_token/[\w/?&=]+)', part, 'thumbnail', default=None) break # downloads enabled we can get the original file @@ -89,4 +106,5 @@ class DropboxIE(InfoExtractor): 'title': title, 'formats': formats, 'subtitles': subtitles, + 'thumbnail': thumbnail, }