From 75ca6bcee2466cb9ca3dc4d1ca35b56a59f6cc4d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Thu, 23 Jun 2016 04:17:13 +0700 Subject: [PATCH] [vk] Workaround buggy new.vk.com Set-Cookie headers --- youtube_dl/extractor/vk.py | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 79c819bc3..4e8ec0f86 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -3,6 +3,7 @@ import re import json +import sys from .common import InfoExtractor from ..compat import compat_str @@ -190,7 +191,7 @@ def _login(self): if username is None: return - login_page = self._download_webpage( + login_page, url_handle = self._download_webpage_handle( 'https://vk.com', None, 'Downloading login page') login_form = self._hidden_inputs(login_page) @@ -200,6 +201,22 @@ def _login(self): 'pass': password.encode('cp1251'), }) + # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header + # and expects the first one to be set rather than second (see + # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201). + # As of RFC6265 the newer one cookie should be set into cookie store + # what actually happens. + # We will workaround this VK issue by resetting the remixlhk cookie to + # the first one manually. + cookies = url_handle.headers.get('Set-Cookie') + if sys.version_info[0] >= 3: + cookies = cookies.encode('iso-8859-1') + cookies = cookies.decode('utf-8') + remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies) + if remixlhk: + value, domain = remixlhk.groups() + self._set_cookie(domain, 'remixlhk', value) + request = sanitized_Request( 'https://login.vk.com/?act=login', urlencode_postdata(login_form))