From aedaa455d9874f14662023f21b254168ecd55579 Mon Sep 17 00:00:00 2001 From: pukkandan Date: Wed, 18 May 2022 05:11:47 +0530 Subject: [PATCH] [vimeo] Fix extractors Closes #3037, Closes #2858, Closes #2880, Closes #3712 May also fix #3602, #3360 --- yt_dlp/extractor/dropout.py | 2 +- yt_dlp/extractor/generic.py | 2 +- yt_dlp/extractor/vimeo.py | 32 +++++++++++++++++++------------- 3 files changed, 21 insertions(+), 15 deletions(-) diff --git a/yt_dlp/extractor/dropout.py b/yt_dlp/extractor/dropout.py index 475825eb8..096216418 100644 --- a/yt_dlp/extractor/dropout.py +++ b/yt_dlp/extractor/dropout.py @@ -137,7 +137,7 @@ class DropoutIE(InfoExtractor): return { '_type': 'url_transparent', 'ie_key': VHXEmbedIE.ie_key(), - 'url': embed_url, + 'url': VHXEmbedIE._smuggle_referrer(embed_url, 'https://www.dropout.tv'), 'id': self._search_regex(r'embed\.vhx\.tv/videos/(.+?)\?', embed_url, 'id'), 'display_id': display_id, 'title': title, diff --git a/yt_dlp/extractor/generic.py b/yt_dlp/extractor/generic.py index 54d9f61c9..dda2b1eef 100644 --- a/yt_dlp/extractor/generic.py +++ b/yt_dlp/extractor/generic.py @@ -2954,7 +2954,7 @@ class GenericIE(InfoExtractor): if vimeo_urls: return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key()) - vhx_url = VHXEmbedIE._extract_url(webpage) + vhx_url = VHXEmbedIE._extract_url(url, webpage) if vhx_url: return self.url_result(vhx_url, VHXEmbedIE.ie_key()) diff --git a/yt_dlp/extractor/vimeo.py b/yt_dlp/extractor/vimeo.py index 59c5353ab..961734345 100644 --- a/yt_dlp/extractor/vimeo.py +++ b/yt_dlp/extractor/vimeo.py @@ -40,6 +40,18 @@ class VimeoBaseInfoExtractor(InfoExtractor): _LOGIN_REQUIRED = False _LOGIN_URL = 'https://vimeo.com/log_in' + @staticmethod + def _smuggle_referrer(url, referrer_url): + return smuggle_url(url, {'http_headers': {'Referer': referrer_url}}) + + def _unsmuggle_headers(self, url): + """@returns (url, smuggled_data, headers)""" + url, data = unsmuggle_url(url, {}) + headers = self.get_param('http_headers').copy() + if 'http_headers' in data: + headers.update(data['http_headers']) + return url, data, headers + def _perform_login(self, username, password): webpage = self._download_webpage( self._LOGIN_URL, None, 'Downloading login page') @@ -717,10 +729,6 @@ class VimeoIE(VimeoBaseInfoExtractor): # vimeo embed with check-password page protected by Referer header ] - @staticmethod - def _smuggle_referrer(url, referrer_url): - return smuggle_url(url, {'http_headers': {'Referer': referrer_url}}) - @staticmethod def _extract_urls(url, webpage): urls = [] @@ -754,8 +762,8 @@ class VimeoIE(VimeoBaseInfoExtractor): 'Content-Type': 'application/x-www-form-urlencoded', }) checked = self._download_json( - url + '/check-password', video_id, - 'Verifying the password', data=data, headers=headers) + f'{compat_urlparse.urlsplit(url)._replace(query=None).geturl()}/check-password', + video_id, 'Verifying the password', data=data, headers=headers) if checked is False: raise ExtractorError('Wrong video password', expected=True) return checked @@ -830,10 +838,7 @@ class VimeoIE(VimeoBaseInfoExtractor): raise def _real_extract(self, url): - url, data = unsmuggle_url(url, {}) - headers = self.get_param('http_headers').copy() - if 'http_headers' in data: - headers.update(data['http_headers']) + url, data, headers = self._unsmuggle_headers(url) if 'Referer' not in headers: headers['Referer'] = url @@ -1383,14 +1388,15 @@ class VHXEmbedIE(VimeoBaseInfoExtractor): _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P\d+)' @staticmethod - def _extract_url(webpage): + def _extract_url(url, webpage): mobj = re.search( r']+src="(https?://embed\.vhx\.tv/videos/\d+[^"]*)"', webpage) - return unescapeHTML(mobj.group(1)) if mobj else None + return VimeoIE._smuggle_referrer(unescapeHTML(mobj.group(1)), url) if mobj else None def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + url, _, headers = self._unsmuggle_headers(url) + webpage = self._download_webpage(url, video_id, headers=headers) config_url = self._parse_json(self._search_regex( r'window\.OTTData\s*=\s*({.+})', webpage, 'ott data'), video_id, js_to_json)['config_url']