Merge branch 'master' of github.com:rg3/youtube-dl

* 'master' of github.com:rg3/youtube-dl:
  [safari] Fix authentication (closes #21090)
  [extractor/common] Move workaround for applying first Set-Cookie header into a separate method

youtube_dl/extractor/common.py

@@ -2817,6 +2817,29 @@ class InfoExtractor(object):
         self._downloader.cookiejar.add_cookie_header(req)
         return compat_cookies.SimpleCookie(req.get_header('Cookie'))
 
+    def _apply_first_set_cookie_header(self, url_handle, cookie):
+        # Some sites (e.g. [1-3]) may serve two cookies under the same name
+        # in Set-Cookie header and expect the first (old) one to be set rather
+        # than second (new). However, as of RFC6265 the newer one cookie
+        # should be set into cookie store what actually happens.
+        # We will workaround this issue by resetting the cookie to
+        # the first one manually.
+        # 1. https://new.vk.com/
+        # 2. https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201
+        # 3. https://learning.oreilly.com/
+        for header, cookies in url_handle.headers.items():
+            if header.lower() != 'set-cookie':
+                continue
+            if sys.version_info[0] >= 3:
+                cookies = cookies.encode('iso-8859-1')
+            cookies = cookies.decode('utf-8')
+            cookie_value = re.search(
+                r'%s=(.+?);.*?\b[Dd]omain=(.+?)(?:[,;]|$)' % cookie, cookies)
+            if cookie_value:
+                value, domain = cookie_value.groups()
+                self._set_cookie(domain, cookie, value)
+                break
+
     def get_testcases(self, include_onlymatching=False):
         t = getattr(self, '_TEST', None)
         if t:

youtube_dl/extractor/safari.py

@@ -1,15 +1,18 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
 
+from ..compat import (
+    compat_parse_qs,
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
     ExtractorError,
-    sanitized_Request,
-    std_headers,
-    urlencode_postdata,
     update_url_query,
 )
 
@@ -31,44 +34,52 @@ class SafariBaseIE(InfoExtractor):
         if username is None:
             return
 
-        headers = std_headers.copy()
+        _, urlh = self._download_webpage_handle(
-        if 'Referer' not in headers:
+            'https://learning.oreilly.com/accounts/login-check/', None,
-            headers['Referer'] = self._LOGIN_URL
+            'Downloading login page')
 
-        login_page = self._download_webpage(
+        def is_logged(urlh):
-            self._LOGIN_URL, None, 'Downloading login form', headers=headers)
+            return 'learning.oreilly.com/home/' in compat_str(urlh.geturl())
 
-        def is_logged(webpage):
+        if is_logged(urlh):
-            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(
+        redirect_url = compat_str(urlh.geturl())
-            r"name='csrfmiddlewaretoken'\s+value='([^']+)'",
+        parsed_url = compat_urlparse.urlparse(redirect_url)
-            login_page, 'csrf token')
+        qs = compat_parse_qs(parsed_url.query)
+        next_uri = compat_urlparse.urljoin(
+            'https://api.oreilly.com', qs['next'][0])
 
-        login_form = {
+        auth, urlh = self._download_json_handle(
-            'csrfmiddlewaretoken': csrf,
+            'https://www.oreilly.com/member/auth/login/', None, 'Logging in',
-            'email': username,
+            data=json.dumps({
-            'password1': password,
+                'email': username,
-            'login': 'Sign In',
+                'password': password,
-            'next': '',
+                'redirect_uri': next_uri,
-        }
+            }).encode(), headers={
+                'Content-Type': 'application/json',
+                'Referer': redirect_url,
+            }, expected_status=400)
 
-        request = sanitized_Request(
+        credentials = auth.get('credentials')
-            self._LOGIN_URL, urlencode_postdata(login_form), headers=headers)
+        if (not auth.get('logged_in') and not auth.get('redirect_uri')
-        login_page = self._download_webpage(
+                and credentials):
-            request, None, 'Logging in')
-
-        if not is_logged(login_page):
             raise ExtractorError(
-                'Login failed; make sure your credentials are correct and try again.',
+                'Unable to login: %s' % credentials, expected=True)
-                expected=True)
 
-        self.LOGGED_IN = True
+        # oreilly serves two same groot_sessionid cookies in Set-Cookie header
+        # and expects first one to be actually set
+        self._apply_first_set_cookie_header(urlh, 'groot_sessionid')
+
+        _, urlh = self._download_webpage_handle(
+            auth.get('redirect_uri') or next_uri, None, 'Completing login',)
+
+        if is_logged(urlh):
+            self.LOGGED_IN = True
+            return
+
+        raise ExtractorError('Unable to log in')
 
 
 class SafariIE(SafariBaseIE):
@@ -76,7 +87,7 @@ class SafariIE(SafariBaseIE):
     IE_DESC = 'safaribooksonline.com online video'
     _VALID_URL = r'''(?x)
                         https?://
-                            (?:www\.)?(?:safaribooksonline|learning\.oreilly)\.com/
+                            (?:www\.)?(?:safaribooksonline|(?:learning\.)?oreilly)\.com/
                             (?:
                                 library/view/[^/]+/(?P<course_id>[^/]+)/(?P<part>[^/?\#&]+)\.html|
                                 videos/[^/]+/[^/]+/(?P<reference_id>[^-]+-[^/?\#&]+)
@@ -107,6 +118,9 @@ class SafariIE(SafariBaseIE):
     }, {
         'url': 'https://learning.oreilly.com/videos/hadoop-fundamentals-livelessons/9780133392838/9780133392838-00_SeriesIntro',
         'only_matching': True,
+    }, {
+        'url': 'https://www.oreilly.com/library/view/hadoop-fundamentals-livelessons/9780133392838/00_SeriesIntro.html',
+        'only_matching': True,
     }]
 
     _PARTNER_ID = '1926081'
@@ -163,7 +177,7 @@ class SafariIE(SafariBaseIE):
 
 class SafariApiIE(SafariBaseIE):
     IE_NAME = 'safari:api'
-    _VALID_URL = r'https?://(?:www\.)?(?:safaribooksonline|learning\.oreilly)\.com/api/v1/book/(?P<course_id>[^/]+)/chapter(?:-content)?/(?P<part>[^/?#&]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?(?:safaribooksonline|(?:learning\.)?oreilly)\.com/api/v1/book/(?P<course_id>[^/]+)/chapter(?:-content)?/(?P<part>[^/?#&]+)\.html'
 
     _TESTS = [{
         'url': 'https://www.safaribooksonline.com/api/v1/book/9780133392838/chapter/part00.html',
@@ -188,7 +202,7 @@ class SafariCourseIE(SafariBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:www\.)?(?:safaribooksonline|learning\.oreilly)\.com/
+                            (?:www\.)?(?:safaribooksonline|(?:learning\.)?oreilly)\.com/
                             (?:
                                 library/view/[^/]+|
                                 api/v1/book|
@@ -219,6 +233,9 @@ class SafariCourseIE(SafariBaseIE):
     }, {
         'url': 'https://learning.oreilly.com/videos/hadoop-fundamentals-livelessons/9780133392838',
         'only_matching': True,
+    }, {
+        'url': 'https://www.oreilly.com/library/view/hadoop-fundamentals-livelessons/9780133392838/',
+        'only_matching': True,
     }]
 
     @classmethod

youtube_dl/extractor/vk.py

@@ -3,7 +3,6 @@ from __future__ import unicode_literals
 
 import collections
 import re
-import sys
 
 from .common import InfoExtractor
 from ..compat import compat_urlparse
@@ -45,24 +44,9 @@ class VKBaseIE(InfoExtractor):
             'pass': password.encode('cp1251'),
         })
 
-        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
+        # vk serves two same remixlhk cookies in Set-Cookie header and expects
-        # and expects the first one to be set rather than second (see
+        # first one to be actually set
-        # https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201).
+        self._apply_first_set_cookie_header(url_handle, 'remixlhk')
-        # 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.
-        for header, cookies in url_handle.headers.items():
-            if header.lower() != 'set-cookie':
-                continue
-            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)
-                break
 
         login_page = self._download_webpage(
             'https://login.vk.com/?act=login', None,