midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[safari] Improve authentication detection (closes #13319)

Browse source
This commit is contained in:
Sergey M․ 2017-06-08 23:20:48 +07:00
parent 931adf8cc1
commit 4244a13a1d
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 10 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

21
youtube_dl/extractor/safari.py
View file

@ -16,7 +16,6 @@ from ..utils import (
class SafariBaseIE(InfoExtractor): class SafariBaseIE(InfoExtractor):
_LOGIN_URL = 'https://www.safaribooksonline.com/accounts/login/' _LOGIN_URL = 'https://www.safaribooksonline.com/accounts/login/'
_SUCCESSFUL_LOGIN_REGEX = r'<a href="/accounts/logout/"[^>]*>Sign Out</a>'
_NETRC_MACHINE = 'safari' _NETRC_MACHINE = 'safari'
_API_BASE = 'https://www.safaribooksonline.com/api/v1' _API_BASE = 'https://www.safaribooksonline.com/api/v1'
@ -28,10 +27,6 @@ class SafariBaseIE(InfoExtractor):
self._login() self._login()
def _login(self): def _login(self):
# We only need to log in once for courses or individual videos
if self.LOGGED_IN:
return
(username, password) = self._get_login_info() (username, password) = self._get_login_info()
if username is None: if username is None:
return return
@ -39,11 +34,17 @@ class SafariBaseIE(InfoExtractor):
headers = std_headers.copy() headers = std_headers.copy()
if 'Referer' not in headers: if 'Referer' not in headers:
headers['Referer'] = self._LOGIN_URL headers['Referer'] = self._LOGIN_URL
login_page_request = sanitized_Request(self._LOGIN_URL, headers=headers)
login_page = self._download_webpage( login_page = self._download_webpage(
login_page_request, None, self._LOGIN_URL, None, 'Downloading login form', headers=headers)
'Downloading login form')
def is_logged(webpage):
return any(re.search(p, webpage) for p in (
r'href=["\']/accounts/logout/', r'>Sign Out<'))
if is_logged(login_page):
self.LOGGED_IN = True
return
csrf = self._html_search_regex( csrf = self._html_search_regex(
r"name='csrfmiddlewaretoken'\s+value='([^']+)'", r"name='csrfmiddlewaretoken'\s+value='([^']+)'",
@ -62,15 +63,13 @@ class SafariBaseIE(InfoExtractor):
login_page = self._download_webpage( login_page = self._download_webpage(
request, None, 'Logging in as %s' % username) request, None, 'Logging in as %s' % username)
if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None: if not is_logged(login_page):
raise ExtractorError( raise ExtractorError(
'Login failed; make sure your credentials are correct and try again.', 'Login failed; make sure your credentials are correct and try again.',
expected=True) expected=True)
self.LOGGED_IN = True self.LOGGED_IN = True
self.to_screen('Login successful')
class SafariIE(SafariBaseIE): class SafariIE(SafariBaseIE):
IE_NAME = 'safari' IE_NAME = 'safari'