Merge tag 'upstream/2015.01.16'

[youtubedl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 3c9b0b58402189338b5480d1d20f589b751e6dfe..2a1f8be0a7e6d6946b5e1801a4b3cd6dda338efe 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -1,137 +1,194 @@
 # coding: utf-8
 
 # coding: utf-8
 

-import errno
-import io
+from __future__ import unicode_literals
+
+

 import itertools
 import json
 import os.path
 import re
 import itertools
 import json
 import os.path
 import re

+import time

 import traceback
 
 from .common import InfoExtractor, SearchInfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..jsinterp import JSInterpreter
 from ..swfinterp import SWFInterpreter
 import traceback
 
 from .common import InfoExtractor, SearchInfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..jsinterp import JSInterpreter
 from ..swfinterp import SWFInterpreter

-from ..utils import (
+from ..compat import (

     compat_chr,
     compat_parse_qs,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
     compat_str,
     compat_chr,
     compat_parse_qs,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
     compat_str,

-
+)
+from ..utils import (

     clean_html,
     clean_html,

-    get_cachedir,
-    get_element_by_id,
-    get_element_by_attribute,

     ExtractorError,
     ExtractorError,

+    get_element_by_attribute,
+    get_element_by_id,

     int_or_none,
     int_or_none,

-    PagedList,
+    OnDemandPagedList,
+    orderedSet,

     unescapeHTML,
     unified_strdate,
     unescapeHTML,
     unified_strdate,

-    orderedSet,
-    write_json_file,

     uppercase_escape,
 )
 
     uppercase_escape,
 )
 

+

 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'

-    _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
-    _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
+    _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'

     _NETRC_MACHINE = 'youtube'
     # If True it will raise an error if no login info is provided
     _LOGIN_REQUIRED = False
 
     def _set_language(self):
     _NETRC_MACHINE = 'youtube'
     # If True it will raise an error if no login info is provided
     _LOGIN_REQUIRED = False
 
     def _set_language(self):

-        return bool(self._download_webpage(
-            self._LANG_URL, None,
-            note=u'Setting language', errnote='unable to set language',
-            fatal=False))
+        self._set_cookie(
+            '.youtube.com', 'PREF', 'f1=50000000&hl=en',
+            # YouTube sets the expire time to about two months
+            expire_time=time.time() + 2 * 30 * 24 * 3600)

 
     def _login(self):
 
     def _login(self):

+        """
+        Attempt to log in to YouTube.
+        True is returned if successful or skipped.
+        False is returned if login failed.
+
+        If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
+        """

         (username, password) = self._get_login_info()
         # No authentication to be performed
         if username is None:
             if self._LOGIN_REQUIRED:
         (username, password) = self._get_login_info()
         # No authentication to be performed
         if username is None:
             if self._LOGIN_REQUIRED:

-                raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
-            return False
+                raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
+            return True

 
         login_page = self._download_webpage(
             self._LOGIN_URL, None,
 
         login_page = self._download_webpage(
             self._LOGIN_URL, None,

-            note=u'Downloading login page',
-            errnote=u'unable to fetch login page', fatal=False)
+            note='Downloading login page',
+            errnote='unable to fetch login page', fatal=False)

         if login_page is False:
             return
 
         galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
         if login_page is False:
             return
 
         galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',

-                                  login_page, u'Login GALX parameter')
+                                  login_page, 'Login GALX parameter')

 
         # Log in
         login_form_strs = {
 
         # Log in
         login_form_strs = {

-                u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
-                u'Email': username,
-                u'GALX': galx,
-                u'Passwd': password,
-                u'PersistentCookie': u'yes',
-                u'_utf8': u'霱',
-                u'bgresponse': u'js_disabled',
-                u'checkConnection': u'',
-                u'checkedDomains': u'youtube',
-                u'dnConn': u'',
-                u'pstMsg': u'0',
-                u'rmShown': u'1',
-                u'secTok': u'',
-                u'signIn': u'Sign in',
-                u'timeStmp': u'',
-                u'service': u'youtube',
-                u'uilel': u'3',
-                u'hl': u'en_US',
+            'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
+            'Email': username,
+            'GALX': galx,
+            'Passwd': password,
+
+            'PersistentCookie': 'yes',
+            '_utf8': '霱',
+            'bgresponse': 'js_disabled',
+            'checkConnection': '',
+            'checkedDomains': 'youtube',
+            'dnConn': '',
+            'pstMsg': '0',
+            'rmShown': '1',
+            'secTok': '',
+            'signIn': 'Sign in',
+            'timeStmp': '',
+            'service': 'youtube',
+            'uilel': '3',
+            'hl': 'en_US',

         }
         }

+

         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
         # chokes on unicode
         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
         # chokes on unicode

-        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
+        login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())

         login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
 
         req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
         login_results = self._download_webpage(
             req, None,
         login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
 
         req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
         login_results = self._download_webpage(
             req, None,

-            note=u'Logging in', errnote=u'unable to log in', fatal=False)
+            note='Logging in', errnote='unable to log in', fatal=False)

         if login_results is False:
             return False
         if login_results is False:
             return False

-        if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
-            self._downloader.report_warning(u'unable to log in: bad username or password')
-            return False
-        return True

 
 

-    def _confirm_age(self):
-        age_form = {
-            'next_url': '/',
-            'action_confirm': 'Confirm',
-        }
-        req = compat_urllib_request.Request(self._AGE_URL,
-            compat_urllib_parse.urlencode(age_form).encode('ascii'))
+        if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
+            raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
+
+        # Two-Factor
+        # TODO add SMS and phone call support - these require making a request and then prompting the user
+
+        if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
+            tfa_code = self._get_tfa_info()
+
+            if tfa_code is None:
+                self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
+                self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
+                return False
+
+            # Unlike the first login form, secTok and timeStmp are both required for the TFA form
+
+            match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
+            if match is None:
+                self._downloader.report_warning('Failed to get secTok - did the page structure change?')
+            secTok = match.group(1)
+            match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
+            if match is None:
+                self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
+            timeStmp = match.group(1)
+
+            tfa_form_strs = {
+                'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
+                'smsToken': '',
+                'smsUserPin': tfa_code,
+                'smsVerifyPin': 'Verify',
+
+                'PersistentCookie': 'yes',
+                'checkConnection': '',
+                'checkedDomains': 'youtube',
+                'pstMsg': '1',
+                'secTok': secTok,
+                'timeStmp': timeStmp,
+                'service': 'youtube',
+                'hl': 'en_US',
+            }
+            tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
+            tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
+
+            tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
+            tfa_results = self._download_webpage(
+                tfa_req, None,
+                note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
+
+            if tfa_results is False:
+                return False
+
+            if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
+                self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
+                return False
+            if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
+                self._downloader.report_warning('unable to log in - did the page structure change?')
+                return False
+            if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
+                self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
+                return False

 
 

-        self._download_webpage(
-            req, None,
-            note=u'Confirming age', errnote=u'Unable to confirm age')
+        if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
+            self._downloader.report_warning('unable to log in: bad username or password')
+            return False

         return True
 
     def _real_initialize(self):
         if self._downloader is None:
             return
         return True
 
     def _real_initialize(self):
         if self._downloader is None:
             return

-        if not self._set_language():
-            return
+        self._set_language()

         if not self._login():
             return
         if not self._login():
             return

-        self._confirm_age()

 
 
 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
 
 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):

-    IE_DESC = u'YouTube.com'
+    IE_DESC = 'YouTube.com'

     _VALID_URL = r"""(?x)^
                      (
     _VALID_URL = r"""(?x)^
                      (

-                         (?:https?://|//)?                                    # http(s):// or protocol-independent URL (optional)
+                         (?:https?://|//)                                    # http(s):// or protocol-independent URL

                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
                             (?:www\.)?deturl\.com/www\.youtube\.com/|
                             (?:www\.)?pwnyoutube\.com/|
                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
                             (?:www\.)?deturl\.com/www\.youtube\.com/|
                             (?:www\.)?pwnyoutube\.com/|


@@ -140,7 +197,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
                          (?:                                                  # the various things that can precede the ID:
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
                          (?:                                                  # the various things that can precede the ID:

-                             (?:(?:v|embed|e)/)                               # v/ or embed/ or e/
+                             (?:(?:v|embed|e)/(?!videoseries))                # v/ or embed/ or e/

                              |(?:                                             # or the v= param in all its forms
                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)?  # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
                              |(?:                                             # or the v= param in all its forms
                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)?  # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!


@@ -149,10 +206,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                              )
                          ))
                          |youtu\.be/                                          # just youtu.be/xxxx
                              )
                          ))
                          |youtu\.be/                                          # just youtu.be/xxxx

-                         |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
+                         |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=

                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID

+                     (?!.*?&list=)                                            # combined list/video URLs are handled by the playlist IE

                      (?(1).+)?                                                # if we found the ID, everything can follow
                      $"""
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
                      (?(1).+)?                                                # if we found the ID, everything can follow
                      $"""
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'


@@ -198,14 +256,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},

-        '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
+        '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},  # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)

         '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},

+        '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
+        '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
+        '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},

 
         # Dash mp4 audio
 
         # Dash mp4 audio

-        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
-        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
-        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
+        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50},

 
         # Dash webm
         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
 
         # Dash webm
         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},


@@ -214,6 +275,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
         '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
         '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
         '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
         '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
         '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},

+        '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},

         '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},


@@ -223,134 +285,246 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
         '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},

+        '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
+        '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
+        '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
+        '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
+        '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},

 
         # Dash webm audio
 
         # Dash webm audio

-        '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 48, 'preference': -50},
+        '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},

         '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
 
         '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
 

+        # Dash webm audio with opus inside
+        '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
+        '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
+        '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
+

         # RTMP (unnamed)
         '_rtmp': {'protocol': 'rtmp'},
     }
 
         # RTMP (unnamed)
         '_rtmp': {'protocol': 'rtmp'},
     }
 

-    IE_NAME = u'youtube'
+    IE_NAME = 'youtube'

     _TESTS = [
         {
     _TESTS = [
         {

-            u"url":  u"http://www.youtube.com/watch?v=BaW_jenozKc",
-            u"file":  u"BaW_jenozKc.mp4",
-            u"info_dict": {
-                u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
-                u"uploader": u"Philipp Hagemeister",
-                u"uploader_id": u"phihag",
-                u"upload_date": u"20121002",
-                u"description": u"test chars:  \"'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .",
-                u"categories": [u'Science & Technology'],
+            'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
+            'info_dict': {
+                'id': 'BaW_jenozKc',
+                'ext': 'mp4',
+                'title': 'youtube-dl test video "\'/\\ä↭𝕐',
+                'uploader': 'Philipp Hagemeister',
+                'uploader_id': 'phihag',
+                'upload_date': '20121002',
+                'description': 'test chars:  "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
+                'categories': ['Science & Technology'],
+                'like_count': int,
+                'dislike_count': int,

             }
         },
         {
             }
         },
         {

-            u"url":  u"http://www.youtube.com/watch?v=UxxajLWwzqY",
-            u"file":  u"UxxajLWwzqY.mp4",
-            u"note": u"Test generic use_cipher_signature video (#897)",
-            u"info_dict": {
-                u"upload_date": u"20120506",
-                u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
-                u"description": u"md5:fea86fda2d5a5784273df5c7cc994d9f",
-                u"uploader": u"Icona Pop",
-                u"uploader_id": u"IconaPop"
+            'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
+            'note': 'Test generic use_cipher_signature video (#897)',
+            'info_dict': {
+                'id': 'UxxajLWwzqY',
+                'ext': 'mp4',
+                'upload_date': '20120506',
+                'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
+                'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
+                'uploader': 'Icona Pop',
+                'uploader_id': 'IconaPop',

             }
         },
         {
             }
         },
         {

-            u"url":  u"https://www.youtube.com/watch?v=07FYdnEawAQ",
-            u"file":  u"07FYdnEawAQ.mp4",
-            u"note": u"Test VEVO video with age protection (#956)",
-            u"info_dict": {
-                u"upload_date": u"20130703",
-                u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
-                u"description": u"md5:64249768eec3bc4276236606ea996373",
-                u"uploader": u"justintimberlakeVEVO",
-                u"uploader_id": u"justintimberlakeVEVO"
+            'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
+            'note': 'Test VEVO video with age protection (#956)',
+            'info_dict': {
+                'id': '07FYdnEawAQ',
+                'ext': 'mp4',
+                'upload_date': '20130703',
+                'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
+                'description': 'md5:64249768eec3bc4276236606ea996373',
+                'uploader': 'justintimberlakeVEVO',
+                'uploader_id': 'justintimberlakeVEVO',

             }
         },
         {
             }
         },
         {

-            u"url":  u"//www.YouTube.com/watch?v=yZIXLfi8CZQ",
-            u"file":  u"yZIXLfi8CZQ.mp4",
-            u"note": u"Embed-only video (#1746)",
-            u"info_dict": {
-                u"upload_date": u"20120608",
-                u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012",
-                u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7",
-                u"uploader": u"SET India",
-                u"uploader_id": u"setindia"
+            'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
+            'note': 'Embed-only video (#1746)',
+            'info_dict': {
+                'id': 'yZIXLfi8CZQ',
+                'ext': 'mp4',
+                'upload_date': '20120608',
+                'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
+                'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
+                'uploader': 'SET India',
+                'uploader_id': 'setindia'

             }
         },
         {
             }
         },
         {

-            u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I",
-            u"file": u"a9LDPn-MO4I.m4a",
-            u"note": u"256k DASH audio (format 141) via DASH manifest",
-            u"info_dict": {
-                u"upload_date": "20121002",
-                u"uploader_id": "8KVIDEO",
-                u"description": "No description available.",
-                u"uploader": "8KVIDEO",
-                u"title": "UHDTV TEST 8K VIDEO.mp4"
+            'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
+            'note': '256k DASH audio (format 141) via DASH manifest',
+            'info_dict': {
+                'id': 'a9LDPn-MO4I',
+                'ext': 'm4a',
+                'upload_date': '20121002',
+                'uploader_id': '8KVIDEO',
+                'description': '',
+                'uploader': '8KVIDEO',
+                'title': 'UHDTV TEST 8K VIDEO.mp4'

             },
             },

-            u"params": {
-                u"youtube_include_dash_manifest": True,
-                u"format": "141",
+            'params': {
+                'youtube_include_dash_manifest': True,
+                'format': '141',

             },
         },
         # DASH manifest with encrypted signature
         {
             },
         },
         # DASH manifest with encrypted signature
         {

-            u'url': u'https://www.youtube.com/watch?v=IB3lcPjvWLA',
-            u'info_dict': {
-                u'id': u'IB3lcPjvWLA',
-                u'ext': u'm4a',
-                u'title': u'Afrojack - The Spark ft. Spree Wilson',
-                u'description': u'md5:9717375db5a9a3992be4668bbf3bc0a8',
-                u'uploader': u'AfrojackVEVO',
-                u'uploader_id': u'AfrojackVEVO',
-                u'upload_date': u'20131011',
+            'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
+            'info_dict': {
+                'id': 'IB3lcPjvWLA',
+                'ext': 'm4a',
+                'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
+                'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
+                'uploader': 'AfrojackVEVO',
+                'uploader_id': 'AfrojackVEVO',
+                'upload_date': '20131011',
+            },
+            'params': {
+                'youtube_include_dash_manifest': True,
+                'format': '141',
+            },
+        },
+        # JS player signature function name containing $
+        {
+            'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
+            'info_dict': {
+                'id': 'nfWlot6h_JM',
+                'ext': 'm4a',
+                'title': 'Taylor Swift - Shake It Off',
+                'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
+                'uploader': 'TaylorSwiftVEVO',
+                'uploader_id': 'TaylorSwiftVEVO',
+                'upload_date': '20140818',
+            },
+            'params': {
+                'youtube_include_dash_manifest': True,
+                'format': '141',
+            },
+        },
+        # Controversy video
+        {
+            'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
+            'info_dict': {
+                'id': 'T4XJQO3qol8',
+                'ext': 'mp4',
+                'upload_date': '20100909',
+                'uploader': 'The Amazing Atheist',
+                'uploader_id': 'TheAmazingAtheist',
+                'title': 'Burning Everyone\'s Koran',
+                'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
+            }
+        },
+        # Normal age-gate video (No vevo, embed allowed)
+        {
+            'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
+            'info_dict': {
+                'id': 'HtVdAasjOgU',
+                'ext': 'mp4',
+                'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
+                'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
+                'uploader': 'The Witcher',
+                'uploader_id': 'WitcherGame',
+                'upload_date': '20140605',
+            },
+        },
+        # Age-gate video with encrypted signature
+        {
+            'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
+            'info_dict': {
+                'id': '6kLq3WMV1nU',
+                'ext': 'mp4',
+                'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
+                'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
+                'uploader': 'LloydVEVO',
+                'uploader_id': 'LloydVEVO',
+                'upload_date': '20110629',
+            },
+        },
+        # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
+        {
+            'url': '__2ABJjxzNo',
+            'info_dict': {
+                'id': '__2ABJjxzNo',
+                'ext': 'mp4',
+                'upload_date': '20100430',
+                'uploader_id': 'deadmau5',
+                'description': 'md5:12c56784b8032162bb936a5f76d55360',
+                'uploader': 'deadmau5',
+                'title': 'Deadmau5 - Some Chords (HD)',

             },
             },

-            u"params": {
-                u'youtube_include_dash_manifest': True,
-                u'format': '141',
+            'expected_warnings': [
+                'DASH manifest missing',
+            ]
+        },
+        # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
+        {
+            'url': 'lqQg6PlCWgI',
+            'info_dict': {
+                'id': 'lqQg6PlCWgI',
+                'ext': 'mp4',
+                'upload_date': '20120731',
+                'uploader_id': 'olympic',
+                'description': 'HO09  - Women -  GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
+                'uploader': 'Olympics',
+                'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',

             },
             },

+            'params': {
+                'skip_download': 'requires avconv',
+            }

         },
         },

+        # Non-square pixels
+        {
+            'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
+            'info_dict': {
+                'id': '_b-2C3KPAM0',
+                'ext': 'mp4',
+                'stretched_ratio': 16 / 9.,
+                'upload_date': '20110310',
+                'uploader_id': 'AllenMeow',
+                'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
+                'uploader': '孫艾倫',
+                'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
+            },
+        }

     ]
 
     ]
 

-
-    @classmethod
-    def suitable(cls, url):
-        """Receives a URL and returns True if suitable for this IE."""
-        if YoutubePlaylistIE.suitable(url): return False
-        return re.match(cls._VALID_URL, url) is not None
-

     def __init__(self, *args, **kwargs):
         super(YoutubeIE, self).__init__(*args, **kwargs)
         self._player_cache = {}
 
     def report_video_info_webpage_download(self, video_id):
         """Report attempt to download video info webpage."""
     def __init__(self, *args, **kwargs):
         super(YoutubeIE, self).__init__(*args, **kwargs)
         self._player_cache = {}
 
     def report_video_info_webpage_download(self, video_id):
         """Report attempt to download video info webpage."""

-        self.to_screen(u'%s: Downloading video info webpage' % video_id)
+        self.to_screen('%s: Downloading video info webpage' % video_id)

 
     def report_information_extraction(self, video_id):
         """Report attempt to extract video information."""
 
     def report_information_extraction(self, video_id):
         """Report attempt to extract video information."""

-        self.to_screen(u'%s: Extracting video information' % video_id)
+        self.to_screen('%s: Extracting video information' % video_id)

 
     def report_unavailable_format(self, video_id, format):
         """Report extracted video URL."""
 
     def report_unavailable_format(self, video_id, format):
         """Report extracted video URL."""

-        self.to_screen(u'%s: Format %s not available' % (video_id, format))
+        self.to_screen('%s: Format %s not available' % (video_id, format))

 
     def report_rtmp_download(self):
         """Indicate the download will use the RTMP protocol."""
 
     def report_rtmp_download(self):
         """Indicate the download will use the RTMP protocol."""

-        self.to_screen(u'RTMP download detected')
+        self.to_screen('RTMP download detected')

 
     def _signature_cache_id(self, example_sig):
         """ Return a string representation of a signature """
 
     def _signature_cache_id(self, example_sig):
         """ Return a string representation of a signature """

-        return u'.'.join(compat_str(len(part)) for part in example_sig.split('.'))
+        return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))

 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(

-            r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',

             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)


@@ -361,65 +535,46 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         func_id = '%s_%s_%s' % (
             player_type, player_id, self._signature_cache_id(example_sig))
         assert os.path.basename(func_id) == func_id
         func_id = '%s_%s_%s' % (
             player_type, player_id, self._signature_cache_id(example_sig))
         assert os.path.basename(func_id) == func_id

-        cache_dir = get_cachedir(self._downloader.params)

 
 

-        cache_enabled = cache_dir is not None
-        if cache_enabled:
-            cache_fn = os.path.join(os.path.expanduser(cache_dir),
-                                    u'youtube-sigfuncs',
-                                    func_id + '.json')
-            try:
-                with io.open(cache_fn, 'r', encoding='utf-8') as cachef:
-                    cache_spec = json.load(cachef)
-                return lambda s: u''.join(s[i] for i in cache_spec)
-            except IOError:
-                pass  # No cache available
+        cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
+        if cache_spec is not None:
+            return lambda s: ''.join(s[i] for i in cache_spec)

 
         if player_type == 'js':
             code = self._download_webpage(
                 player_url, video_id,
 
         if player_type == 'js':
             code = self._download_webpage(
                 player_url, video_id,

-                note=u'Downloading %s player %s' % (player_type, player_id),
-                errnote=u'Download of %s failed' % player_url)
+                note='Downloading %s player %s' % (player_type, player_id),
+                errnote='Download of %s failed' % player_url)

             res = self._parse_sig_js(code)
         elif player_type == 'swf':
             urlh = self._request_webpage(
                 player_url, video_id,
             res = self._parse_sig_js(code)
         elif player_type == 'swf':
             urlh = self._request_webpage(
                 player_url, video_id,

-                note=u'Downloading %s player %s' % (player_type, player_id),
-                errnote=u'Download of %s failed' % player_url)
+                note='Downloading %s player %s' % (player_type, player_id),
+                errnote='Download of %s failed' % player_url)

             code = urlh.read()
             res = self._parse_sig_swf(code)
         else:
             assert False, 'Invalid player type %r' % player_type
 
             code = urlh.read()
             res = self._parse_sig_swf(code)
         else:
             assert False, 'Invalid player type %r' % player_type
 

-        if cache_enabled:
-            try:
-                test_string = u''.join(map(compat_chr, range(len(example_sig))))
-                cache_res = res(test_string)
-                cache_spec = [ord(c) for c in cache_res]
-                try:
-                    os.makedirs(os.path.dirname(cache_fn))
-                except OSError as ose:
-                    if ose.errno != errno.EEXIST:
-                        raise
-                write_json_file(cache_spec, cache_fn)
-            except Exception:
-                tb = traceback.format_exc()
-                self._downloader.report_warning(
-                    u'Writing cache to %r failed: %s' % (cache_fn, tb))
+        if cache_spec is None:
+            test_string = ''.join(map(compat_chr, range(len(example_sig))))
+            cache_res = res(test_string)
+            cache_spec = [ord(c) for c in cache_res]

 
 

+        self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)

         return res
 
     def _print_sig_code(self, func, example_sig):
         def gen_sig_code(idxs):
             def _genslice(start, end, step):
         return res
 
     def _print_sig_code(self, func, example_sig):
         def gen_sig_code(idxs):
             def _genslice(start, end, step):

-                starts = u'' if start == 0 else str(start)
-                ends = (u':%d' % (end+step)) if end + step >= 0 else u':'
-                steps = u'' if step == 1 else (u':%d' % step)
-                return u's[%s%s%s]' % (starts, ends, steps)
+                starts = '' if start == 0 else str(start)
+                ends = (':%d' % (end + step)) if end + step >= 0 else ':'
+                steps = '' if step == 1 else (':%d' % step)
+                return 's[%s%s%s]' % (starts, ends, steps)

 
             step = None
 
             step = None

-            start = '(Never used)'  # Quelch pyflakes warnings - start will be
-                                    # set as soon as step is set
+            # Quelch pyflakes warnings - start will be set when step is set
+            start = '(Never used)'

             for i, prev in zip(idxs[1:], idxs[:-1]):
                 if step is not None:
                     if i - prev == step:
             for i, prev in zip(idxs[1:], idxs[:-1]):
                 if step is not None:
                     if i - prev == step:


@@ -432,26 +587,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     start = prev
                     continue
                 else:
                     start = prev
                     continue
                 else:

-                    yield u's[%d]' % prev
+                    yield 's[%d]' % prev

             if step is None:
             if step is None:

-                yield u's[%d]' % i
+                yield 's[%d]' % i

             else:
                 yield _genslice(start, i, step)
 
             else:
                 yield _genslice(start, i, step)
 

-        test_string = u''.join(map(compat_chr, range(len(example_sig))))
+        test_string = ''.join(map(compat_chr, range(len(example_sig))))

         cache_res = func(test_string)
         cache_spec = [ord(c) for c in cache_res]
         cache_res = func(test_string)
         cache_spec = [ord(c) for c in cache_res]

-        expr_code = u' + '.join(gen_sig_code(cache_spec))
+        expr_code = ' + '.join(gen_sig_code(cache_spec))

         signature_id_tuple = '(%s)' % (
             ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
         signature_id_tuple = '(%s)' % (
             ', '.join(compat_str(len(p)) for p in example_sig.split('.')))

-        code = (u'if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
-                u'    return %s\n') % (signature_id_tuple, expr_code)
-        self.to_screen(u'Extracted signature function:\n' + code)
+        code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
+                '    return %s\n') % (signature_id_tuple, expr_code)
+        self.to_screen('Extracted signature function:\n' + code)

 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(
 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(

-            r'signature=([$a-zA-Z]+)', jscode,
-             u'Initial JS player signature function name')
+            r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
+            'Initial JS player signature function name')

 
         jsi = JSInterpreter(jscode)
         initial_function = jsi.extract_function(funcname)
 
         jsi = JSInterpreter(jscode)
         initial_function = jsi.extract_function(funcname)


@@ -459,19 +614,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     def _parse_sig_swf(self, file_contents):
         swfi = SWFInterpreter(file_contents)
 
     def _parse_sig_swf(self, file_contents):
         swfi = SWFInterpreter(file_contents)

-        TARGET_CLASSNAME = u'SignatureDecipher'
+        TARGET_CLASSNAME = 'SignatureDecipher'

         searched_class = swfi.extract_class(TARGET_CLASSNAME)
         searched_class = swfi.extract_class(TARGET_CLASSNAME)

-        initial_function = swfi.extract_function(searched_class, u'decipher')
+        initial_function = swfi.extract_function(searched_class, 'decipher')

         return lambda s: initial_function([s])
 
     def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
         """Turn the encrypted s field into a working signature"""
 
         if player_url is None:
         return lambda s: initial_function([s])
 
     def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
         """Turn the encrypted s field into a working signature"""
 
         if player_url is None:

-            raise ExtractorError(u'Cannot decrypt signature without player_url')
+            raise ExtractorError('Cannot decrypt signature without player_url')

 
 

-        if player_url.startswith(u'//'):
-            player_url = u'https:' + player_url
+        if player_url.startswith('//'):
+            player_url = 'https:' + player_url

         try:
             player_id = (player_url, self._signature_cache_id(s))
             if player_id not in self._player_cache:
         try:
             player_id = (player_url, self._signature_cache_id(s))
             if player_id not in self._player_cache:


@@ -486,31 +641,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         except Exception as e:
             tb = traceback.format_exc()
             raise ExtractorError(
         except Exception as e:
             tb = traceback.format_exc()
             raise ExtractorError(

-                u'Signature extraction failed: ' + tb, cause=e)
+                'Signature extraction failed: ' + tb, cause=e)

 
     def _get_available_subtitles(self, video_id, webpage):
         try:
 
     def _get_available_subtitles(self, video_id, webpage):
         try:

-            sub_list = self._download_webpage(
+            subs_doc = self._download_xml(

                 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
                 video_id, note=False)
         except ExtractorError as err:
                 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
                 video_id, note=False)
         except ExtractorError as err:

-            self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
+            self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))

             return {}
             return {}

-        lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)

 
         sub_lang_list = {}
 
         sub_lang_list = {}

-        for l in lang_list:
-            lang = l[1]
+        for track in subs_doc.findall('track'):
+            lang = track.attrib['lang_code']
+            if lang in sub_lang_list:
+                continue

             params = compat_urllib_parse.urlencode({
                 'lang': lang,
                 'v': video_id,
                 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
             params = compat_urllib_parse.urlencode({
                 'lang': lang,
                 'v': video_id,
                 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),

-                'name': unescapeHTML(l[0]).encode('utf-8'),
+                'name': track.attrib['name'].encode('utf-8'),

             })
             })

-            url = u'https://www.youtube.com/api/timedtext?' + params
+            url = 'https://www.youtube.com/api/timedtext?' + params

             sub_lang_list[lang] = url
         if not sub_lang_list:
             sub_lang_list[lang] = url
         if not sub_lang_list:

-            self._downloader.report_warning(u'video doesn\'t have subtitles')
+            self._downloader.report_warning('video doesn\'t have subtitles')

             return {}
         return sub_lang_list
 
             return {}
         return sub_lang_list
 


@@ -518,17 +674,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         """We need the webpage for getting the captions url, pass it as an
            argument to speed up the process."""
         sub_format = self._downloader.params.get('subtitlesformat', 'srt')
         """We need the webpage for getting the captions url, pass it as an
            argument to speed up the process."""
         sub_format = self._downloader.params.get('subtitlesformat', 'srt')

-        self.to_screen(u'%s: Looking for automatic captions' % video_id)
+        self.to_screen('%s: Looking for automatic captions' % video_id)

         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)

-        err_msg = u'Couldn\'t find automatic captions for %s' % video_id
+        err_msg = 'Couldn\'t find automatic captions for %s' % video_id

         if mobj is None:
             self._downloader.report_warning(err_msg)
             return {}
         player_config = json.loads(mobj.group(1))
         try:
         if mobj is None:
             self._downloader.report_warning(err_msg)
             return {}
         player_config = json.loads(mobj.group(1))
         try:

-            args = player_config[u'args']
-            caption_url = args[u'ttsurl']
-            timestamp = args[u'timestamp']
+            args = player_config['args']
+            caption_url = args['ttsurl']
+            timestamp = args['timestamp']

             # We get the available subtitles
             list_params = compat_urllib_parse.urlencode({
                 'type': 'list',
             # We get the available subtitles
             list_params = compat_urllib_parse.urlencode({
                 'type': 'list',


@@ -538,10 +694,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             list_url = caption_url + '&' + list_params
             caption_list = self._download_xml(list_url, video_id)
             original_lang_node = caption_list.find('track')
             list_url = caption_url + '&' + list_params
             caption_list = self._download_xml(list_url, video_id)
             original_lang_node = caption_list.find('track')

-            if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
-                self._downloader.report_warning(u'Video doesn\'t have automatic captions')
+            if original_lang_node is None:
+                self._downloader.report_warning('Video doesn\'t have automatic captions')

                 return {}
             original_lang = original_lang_node.attrib['lang_code']
                 return {}
             original_lang = original_lang_node.attrib['lang_code']

+            caption_kind = original_lang_node.attrib.get('kind', '')

 
             sub_lang_list = {}
             for lang_node in caption_list.findall('target'):
 
             sub_lang_list = {}
             for lang_node in caption_list.findall('target'):


@@ -551,7 +708,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     'tlang': sub_lang,
                     'fmt': sub_format,
                     'ts': timestamp,
                     'tlang': sub_lang,
                     'fmt': sub_format,
                     'ts': timestamp,

-                    'kind': 'asr',
+                    'kind': caption_kind,

                 })
                 sub_lang_list[sub_lang] = caption_url + '&' + params
             return sub_lang_list
                 })
                 sub_lang_list[sub_lang] = caption_url + '&' + params
             return sub_lang_list


@@ -565,18 +722,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
     def extract_id(cls, url):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:
     def extract_id(cls, url):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)

         video_id = mobj.group(2)
         return video_id
 
     def _extract_from_m3u8(self, manifest_url, video_id):
         url_map = {}
         video_id = mobj.group(2)
         return video_id
 
     def _extract_from_m3u8(self, manifest_url, video_id):
         url_map = {}

+

         def _get_urls(_manifest):
             lines = _manifest.split('\n')
             urls = filter(lambda l: l and not l.startswith('#'),
         def _get_urls(_manifest):
             lines = _manifest.split('\n')
             urls = filter(lambda l: l and not l.startswith('#'),

-                            lines)
+                          lines)

             return urls
             return urls

-        manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
+        manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')

         formats_urls = _get_urls(manifest)
         for format_url in formats_urls:
             itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
         formats_urls = _get_urls(manifest)
         for format_url in formats_urls:
             itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')


@@ -585,12 +743,53 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     def _extract_annotations(self, video_id):
         url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
 
     def _extract_annotations(self, video_id):
         url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id

-        return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.')
+        return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
+
+    def _parse_dash_manifest(
+            self, video_id, dash_manifest_url, player_url, age_gate):
+        def decrypt_sig(mobj):
+            s = mobj.group(1)
+            dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
+            return '/signature/%s' % dec_s
+        dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
+        dash_doc = self._download_xml(
+            dash_manifest_url, video_id,
+            note='Downloading DASH manifest',
+            errnote='Could not download DASH manifest')
+
+        formats = []
+        for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
+            url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
+            if url_el is None:
+                continue
+            format_id = r.attrib['id']
+            video_url = url_el.text
+            filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
+            f = {
+                'format_id': format_id,
+                'url': video_url,
+                'width': int_or_none(r.attrib.get('width')),
+                'height': int_or_none(r.attrib.get('height')),
+                'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+                'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+                'filesize': filesize,
+                'fps': int_or_none(r.attrib.get('frameRate')),
+            }
+            try:
+                existing_format = next(
+                    fo for fo in formats
+                    if fo['format_id'] == format_id)
+            except StopIteration:
+                f.update(self._formats.get(format_id, {}).items())
+                formats.append(f)
+            else:
+                existing_format.update(f)
+        return formats

 
     def _real_extract(self, url):
         proto = (
 
     def _real_extract(self, url):
         proto = (

-            u'http' if self._downloader.params.get('prefer_insecure', False)
-            else u'https')
+            'http' if self._downloader.params.get('prefer_insecure', False)
+            else 'https')

 
         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
         mobj = re.search(self._NEXT_URL_RE, url)
 
         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
         mobj = re.search(self._NEXT_URL_RE, url)


@@ -599,7 +798,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         video_id = self.extract_id(url)
 
         # Get video webpage
         video_id = self.extract_id(url)
 
         # Get video webpage

-        url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
+        url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id

         video_webpage = self._download_webpage(url, video_id)
 
         # Attempt to extract SWF player URL
         video_webpage = self._download_webpage(url, video_id)
 
         # Attempt to extract SWF player URL


@@ -610,42 +809,60 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             player_url = None
 
         # Get video info
             player_url = None
 
         # Get video info

-        self.report_video_info_webpage_download(video_id)

         if re.search(r'player-age-gate-content">', video_webpage) is not None:
         if re.search(r'player-age-gate-content">', video_webpage) is not None:

-            self.report_age_confirmation()

             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
             # this can be viewed without login into Youtube
             age_gate = True
             # We simulate the access to the video from www.youtube.com/v/{video_id}
             # this can be viewed without login into Youtube

+            url = proto + '://www.youtube.com/embed/%s' % video_id
+            embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')

             data = compat_urllib_parse.urlencode({
                 'video_id': video_id,
                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
                 'sts': self._search_regex(
             data = compat_urllib_parse.urlencode({
                 'video_id': video_id,
                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
                 'sts': self._search_regex(

-                    r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'),
+                    r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),

             })
             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
             })
             video_info_url = proto + '://www.youtube.com/get_video_info?' + data

-            video_info_webpage = self._download_webpage(video_info_url, video_id,
-                                    note=False,
-                                    errnote='unable to download video info webpage')
+            video_info_webpage = self._download_webpage(
+                video_info_url, video_id,
+                note='Refetching age-gated info webpage',
+                errnote='unable to download video info webpage')

             video_info = compat_parse_qs(video_info_webpage)
         else:
             age_gate = False
             video_info = compat_parse_qs(video_info_webpage)
         else:
             age_gate = False

-            for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
-                video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
-                        % (video_id, el_type))
-                video_info_webpage = self._download_webpage(video_info_url, video_id,
-                                        note=False,
-                                        errnote='unable to download video info webpage')
-                video_info = compat_parse_qs(video_info_webpage)
-                if 'token' in video_info:
-                    break
+            try:
+                # Try looking directly into the video webpage
+                mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
+                if not mobj:
+                    raise ValueError('Could not find ytplayer.config')  # caught below
+                json_code = uppercase_escape(mobj.group(1))
+                ytplayer_config = json.loads(json_code)
+                args = ytplayer_config['args']
+                # Convert to the same format returned by compat_parse_qs
+                video_info = dict((k, [v]) for k, v in args.items())
+                if 'url_encoded_fmt_stream_map' not in args:
+                    raise ValueError('No stream_map present')  # caught below
+            except ValueError:
+                # We fallback to the get_video_info pages (used by the embed page)
+                self.report_video_info_webpage_download(video_id)
+                for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
+                    video_info_url = (
+                        '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
+                        % (proto, video_id, el_type))
+                    video_info_webpage = self._download_webpage(
+                        video_info_url,
+                        video_id, note=False,
+                        errnote='unable to download video info webpage')
+                    video_info = compat_parse_qs(video_info_webpage)
+                    if 'token' in video_info:
+                        break

         if 'token' not in video_info:
             if 'reason' in video_info:
                 raise ExtractorError(
         if 'token' not in video_info:
             if 'reason' in video_info:
                 raise ExtractorError(

-                    u'YouTube said: %s' % video_info['reason'][0],
+                    'YouTube said: %s' % video_info['reason'][0],

                     expected=True, video_id=video_id)
             else:
                 raise ExtractorError(
                     expected=True, video_id=video_id)
             else:
                 raise ExtractorError(

-                    u'"token" parameter not in video info for unknown reason',
+                    '"token" parameter not in video info for unknown reason',

                     video_id=video_id)
 
         if 'view_count' in video_info:
                     video_id=video_id)
 
         if 'view_count' in video_info:


@@ -655,14 +872,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
         # Check for "rental" videos
         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
 
         # Check for "rental" videos
         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:

-            raise ExtractorError(u'"rental" videos not supported')
+            raise ExtractorError('"rental" videos not supported')

 
         # Start extracting information
         self.report_information_extraction(video_id)
 
         # uploader
         if 'author' not in video_info:
 
         # Start extracting information
         self.report_information_extraction(video_id)
 
         # uploader
         if 'author' not in video_info:

-            raise ExtractorError(u'Unable to extract uploader name')
+            raise ExtractorError('Unable to extract uploader name')

         video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
 
         # uploader_id
         video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
 
         # uploader_id


@@ -671,14 +888,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         if mobj is not None:
             video_uploader_id = mobj.group(1)
         else:
         if mobj is not None:
             video_uploader_id = mobj.group(1)
         else:

-            self._downloader.report_warning(u'unable to extract uploader nickname')
+            self._downloader.report_warning('unable to extract uploader nickname')

 
         # title
         if 'title' in video_info:
             video_title = video_info['title'][0]
         else:
 
         # title
         if 'title' in video_info:
             video_title = video_info['title'][0]
         else:

-            self._downloader.report_warning(u'Unable to extract video title')
-            video_title = u'_'
+            self._downloader.report_warning('Unable to extract video title')
+            video_title = '_'

 
         # thumbnail image
         # We try first to get a high quality image:
 
         # thumbnail image
         # We try first to get a high quality image:


@@ -687,7 +904,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         if m_thumb is not None:
             video_thumbnail = m_thumb.group(1)
         elif 'thumbnail_url' not in video_info:
         if m_thumb is not None:
             video_thumbnail = m_thumb.group(1)
         elif 'thumbnail_url' not in video_info:

-            self._downloader.report_warning(u'unable to extract video thumbnail')
+            self._downloader.report_warning('unable to extract video thumbnail')

             video_thumbnail = None
         else:   # don't panic if we can't find it
             video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
             video_thumbnail = None
         else:   # don't panic if we can't find it
             video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])


@@ -703,7 +920,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
             upload_date = unified_strdate(upload_date)
 
             upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
             upload_date = unified_strdate(upload_date)
 

-        m_cat_container = get_element_by_id("eow-category", video_webpage)
+        m_cat_container = self._search_regex(
+            r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
+            video_webpage, 'categories', default=None)

         if m_cat_container:
             category = self._html_search_regex(
                 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
         if m_cat_container:
             category = self._html_search_regex(
                 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',


@@ -730,17 +949,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             if fd_mobj:
                 video_description = unescapeHTML(fd_mobj.group(1))
             else:
             if fd_mobj:
                 video_description = unescapeHTML(fd_mobj.group(1))
             else:

-                video_description = u''
+                video_description = ''

 
 

-        def _extract_count(klass):
+        def _extract_count(count_name):

             count = self._search_regex(
             count = self._search_regex(

-                r'class="%s">([\d,]+)</span>' % re.escape(klass),
-                video_webpage, klass, default=None)
+                r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
+                video_webpage, count_name, default=None)

             if count is not None:
                 return int(count.replace(',', ''))
             return None
             if count is not None:
                 return int(count.replace(',', ''))
             return None

-        like_count = _extract_count(u'likes-count')
-        dislike_count = _extract_count(u'dislikes-count')
+        like_count = _extract_count('like')
+        dislike_count = _extract_count('dislike')

 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, video_webpage)
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, video_webpage)


@@ -750,7 +969,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             return
 
         if 'length_seconds' not in video_info:
             return
 
         if 'length_seconds' not in video_info:

-            self._downloader.report_warning(u'unable to extract video duration')
+            self._downloader.report_warning('unable to extract video duration')

             video_duration = None
         else:
             video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
             video_duration = None
         else:
             video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))


@@ -758,33 +977,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         # annotations
         video_annotations = None
         if self._downloader.params.get('writeannotations', False):
         # annotations
         video_annotations = None
         if self._downloader.params.get('writeannotations', False):

-                video_annotations = self._extract_annotations(video_id)
-
-        # Decide which formats to download
-        try:
-            mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
-            if not mobj:
-                raise ValueError('Could not find vevo ID')
-            json_code = uppercase_escape(mobj.group(1))
-            ytplayer_config = json.loads(json_code)
-            args = ytplayer_config['args']
-            # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
-            # this signatures are encrypted
-            if 'url_encoded_fmt_stream_map' not in args:
-                raise ValueError(u'No stream_map present')  # caught below
-            re_signature = re.compile(r'[&,]s=')
-            m_s = re_signature.search(args['url_encoded_fmt_stream_map'])
-            if m_s is not None:
-                self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
-                video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
-            m_s = re_signature.search(args.get('adaptive_fmts', u''))
-            if m_s is not None:
-                if 'adaptive_fmts' in video_info:
-                    video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
-                else:
-                    video_info['adaptive_fmts'] = [args['adaptive_fmts']]
-        except ValueError:
-            pass
+            video_annotations = self._extract_annotations(video_id)

 
         def _map_to_format_list(urlmap):
             formats = []
 
         def _map_to_format_list(urlmap):
             formats = []


@@ -807,8 +1000,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'url': video_info['conn'][0],
                 'player_url': player_url,
             }]
                 'url': video_info['conn'][0],
                 'player_url': player_url,
             }]

-        elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
-            encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0]
+        elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
+            encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]

             if 'rtmpe%3Dyes' in encoded_url_map:
                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
             url_map = {}
             if 'rtmpe%3Dyes' in encoded_url_map:
                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
             url_map = {}


@@ -824,15 +1017,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
 
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
 

-                    if not age_gate:
-                        jsplayer_url_json = self._search_regex(
-                            r'"assets":.+?"js":\s*("[^"]+")',
-                            video_webpage, u'JS player URL')
-                        player_url = json.loads(jsplayer_url_json)
+                    jsplayer_url_json = self._search_regex(
+                        r'"assets":.+?"js":\s*("[^"]+")',
+                        embed_webpage if age_gate else video_webpage, 'JS player URL')
+                    player_url = json.loads(jsplayer_url_json)

                     if player_url is None:
                         player_url_json = self._search_regex(
                             r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
                     if player_url is None:
                         player_url_json = self._search_regex(
                             r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',

-                            video_webpage, u'age gate player URL')
+                            video_webpage, 'age gate player URL')

                         player_url = json.loads(player_url_json)
 
                     if self._downloader.params.get('verbose'):
                         player_url = json.loads(player_url_json)
 
                     if self._downloader.params.get('verbose'):


@@ -843,18 +1035,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                             if player_url.endswith('swf'):
                                 player_version = self._search_regex(
                                     r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
                             if player_url.endswith('swf'):
                                 player_version = self._search_regex(
                                     r'-(.+?)(?:/watch_as3)?\.swf$', player_url,

-                                    u'flash player', fatal=False)
+                                    'flash player', fatal=False)

                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
                                     r'html5player-([^/]+?)(?:/html5player)?\.js',
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
                                     r'html5player-([^/]+?)(?:/html5player)?\.js',
                                     player_url,
                                     'html5 player', fatal=False)

-                                player_desc = u'html5 player %s' % player_version
+                                player_desc = 'html5 player %s' % player_version

 
                         parts_sizes = self._signature_cache_id(encrypted_sig)
 
                         parts_sizes = self._signature_cache_id(encrypted_sig)

-                        self.to_screen(u'{%s} signature length %s, %s' %
-                            (format_id, parts_sizes, player_desc))
+                        self.to_screen('{%s} signature length %s, %s' %
+                                       (format_id, parts_sizes, player_desc))

 
                     signature = self._decrypt_signature(
                         encrypted_sig, video_id, player_url, age_gate)
 
                     signature = self._decrypt_signature(
                         encrypted_sig, video_id, player_url, age_gate)


@@ -868,93 +1060,75 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             url_map = self._extract_from_m3u8(manifest_url, video_id)
             formats = _map_to_format_list(url_map)
         else:
             url_map = self._extract_from_m3u8(manifest_url, video_id)
             formats = _map_to_format_list(url_map)
         else:

-            raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
+            raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')

 
         # Look for the DASH manifest
 
         # Look for the DASH manifest

-        if (self._downloader.params.get('youtube_include_dash_manifest', False)):
-            try:
-                # The DASH manifest used needs to be the one from the original video_webpage.
-                # The one found in get_video_info seems to be using different signatures.
-                # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
-                # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
-                # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
-                if age_gate:
-                    dash_manifest_url = video_info.get('dashmpd')[0]
+        if self._downloader.params.get('youtube_include_dash_manifest', True):
+            dash_mpd = video_info.get('dashmpd')
+            if dash_mpd:
+                dash_manifest_url = dash_mpd[0]
+                try:
+                    dash_formats = self._parse_dash_manifest(
+                        video_id, dash_manifest_url, player_url, age_gate)
+                except (ExtractorError, KeyError) as e:
+                    self.report_warning(
+                        'Skipping DASH manifest: %r' % e, video_id)

                 else:
                 else:

-                    dash_manifest_url = ytplayer_config['args']['dashmpd']
-                def decrypt_sig(mobj):
-                    s = mobj.group(1)
-                    dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
-                    return '/signature/%s' % dec_s
-                dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
-                dash_doc = self._download_xml(
-                    dash_manifest_url, video_id,
-                    note=u'Downloading DASH manifest',
-                    errnote=u'Could not download DASH manifest')
-                for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
-                    url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
-                    if url_el is None:
-                        continue
-                    format_id = r.attrib['id']
-                    video_url = url_el.text
-                    filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
-                    f = {
-                        'format_id': format_id,
-                        'url': video_url,
-                        'width': int_or_none(r.attrib.get('width')),
-                        'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
-                        'asr': int_or_none(r.attrib.get('audioSamplingRate')),
-                        'filesize': filesize,
-                    }
-                    try:
-                        existing_format = next(
-                            fo for fo in formats
-                            if fo['format_id'] == format_id)
-                    except StopIteration:
-                        f.update(self._formats.get(format_id, {}))
-                        formats.append(f)
-                    else:
-                        existing_format.update(f)
-
-            except (ExtractorError, KeyError) as e:
-                self.report_warning(u'Skipping DASH manifest: %s' % e, video_id)
+                    # Hide the formats we found through non-DASH
+                    dash_keys = set(df['format_id'] for df in dash_formats)
+                    for f in formats:
+                        if f['format_id'] in dash_keys:
+                            f['format_id'] = 'nondash-%s' % f['format_id']
+                            f['preference'] = f.get('preference', 0) - 10000
+                    formats.extend(dash_formats)
+
+        # Check for malformed aspect ratio
+        stretched_m = re.search(
+            r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
+            video_webpage)
+        if stretched_m:
+            ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
+            for f in formats:
+                if f.get('vcodec') != 'none':
+                    f['stretched_ratio'] = ratio

 
         self._sort_formats(formats)
 
         return {
 
         self._sort_formats(formats)
 
         return {

-            'id':           video_id,
-            'uploader':     video_uploader,
-            'uploader_id':  video_uploader_id,
-            'upload_date':  upload_date,
-            'title':        video_title,
-            'thumbnail':    video_thumbnail,
-            'description':  video_description,
-            'categories':   video_categories,
-            'subtitles':    video_subtitles,
-            'duration':     video_duration,
-            'age_limit':    18 if age_gate else 0,
-            'annotations':  video_annotations,
+            'id': video_id,
+            'uploader': video_uploader,
+            'uploader_id': video_uploader_id,
+            'upload_date': upload_date,
+            'title': video_title,
+            'thumbnail': video_thumbnail,
+            'description': video_description,
+            'categories': video_categories,
+            'subtitles': video_subtitles,
+            'duration': video_duration,
+            'age_limit': 18 if age_gate else 0,
+            'annotations': video_annotations,

             'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
             'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,

-            'view_count':   view_count,
+            'view_count': view_count,

             'like_count': like_count,
             'dislike_count': dislike_count,
             'like_count': like_count,
             'dislike_count': dislike_count,

-            'formats':      formats,
+            'formats': formats,

         }
 
         }
 

+

 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):

-    IE_DESC = u'YouTube.com playlists'
+    IE_DESC = 'YouTube.com playlists'

     _VALID_URL = r"""(?x)(?:
                         (?:https?://)?
                         (?:\w+\.)?
                         youtube\.com/
                         (?:
     _VALID_URL = r"""(?x)(?:
                         (?:https?://)?
                         (?:\w+\.)?
                         youtube\.com/
                         (?:

-                           (?:course|view_play_list|my_playlists|artist|playlist|watch)
+                           (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)

                            \? (?:.*?&)*? (?:p|a|list)=
                         |  p/
                         )
                         (
                             (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
                            \? (?:.*?&)*? (?:p|a|list)=
                         |  p/
                         )
                         (
                             (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}

-                            # Top tracks, they can also include dots 
+                            # Top tracks, they can also include dots

                             |(?:MC)[\w\.]*
                         )
                         .*
                             |(?:MC)[\w\.]*
                         )
                         .*


@@ -962,29 +1136,95 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                         ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
                      )"""
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
                         ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
                      )"""
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'

-    _MORE_PAGES_INDICATOR = r'data-link-type="next"'

     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'

-    IE_NAME = u'youtube:playlist'
+    IE_NAME = 'youtube:playlist'
+    _TESTS = [{
+        'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
+        'info_dict': {
+            'title': 'ytdl test PL',
+            'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
+        },
+        'playlist_count': 3,
+    }, {
+        'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
+        'info_dict': {
+            'title': 'YDL_Empty_List',
+        },
+        'playlist_count': 0,
+    }, {
+        'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
+        'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
+        'info_dict': {
+            'title': '29C3: Not my department',
+        },
+        'playlist_count': 95,
+    }, {
+        'note': 'issue #673',
+        'url': 'PLBB231211A4F62143',
+        'info_dict': {
+            'title': '[OLD]Team Fortress 2 (Class-based LP)',
+        },
+        'playlist_mincount': 26,
+    }, {
+        'note': 'Large playlist',
+        'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
+        'info_dict': {
+            'title': 'Uploads from Cauchemar',
+        },
+        'playlist_mincount': 799,
+    }, {
+        'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
+        'info_dict': {
+            'title': 'YDL_safe_search',
+        },
+        'playlist_count': 2,
+    }, {
+        'note': 'embedded',
+        'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
+        'playlist_count': 4,
+        'info_dict': {
+            'title': 'JODA15',
+        }
+    }, {
+        'note': 'Embedded SWF player',
+        'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
+        'playlist_count': 4,
+        'info_dict': {
+            'title': 'JODA7',
+        }
+    }, {
+        'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
+        'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
+        'info_dict': {
+                'title': 'Uploads from Interstellar Movie',
+        },
+        'playlist_mincout': 21,
+    }]

 
     def _real_initialize(self):
         self._login()
 
     def _ids_to_results(self, ids):
 
     def _real_initialize(self):
         self._login()
 
     def _ids_to_results(self, ids):

-        return [self.url_result(vid_id, 'Youtube', video_id=vid_id)
-                       for vid_id in ids]
+        return [
+            self.url_result(vid_id, 'Youtube', video_id=vid_id)
+            for vid_id in ids]

 
     def _extract_mix(self, playlist_id):
         # The mixes are generated from a a single video
         # the id of the playlist is just 'RD' + video_id
         url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
 
     def _extract_mix(self, playlist_id):
         # The mixes are generated from a a single video
         # the id of the playlist is just 'RD' + video_id
         url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)

-        webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix')
+        webpage = self._download_webpage(
+            url, playlist_id, 'Downloading Youtube mix')

         search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
         search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)

-        title_span = (search_title('playlist-title') or
-            search_title('title long-title') or search_title('title'))
+        title_span = (
+            search_title('playlist-title') or
+            search_title('title long-title') or
+            search_title('title'))

         title = clean_html(title_span)
         title = clean_html(title_span)

-        video_re = r'''(?x)data-video-username=".*?".*?
-                       href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id)
-        ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL))
+        ids = orderedSet(re.findall(
+            r'''(?xs)data-video-username=".*?".*?
+                       href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
+            webpage))

         url_results = self._ids_to_results(ids)
 
         return self.playlist_result(url_results, playlist_id, title)
         url_results = self._ids_to_results(ids)
 
         return self.playlist_result(url_results, playlist_id, title)


@@ -993,7 +1233,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         # Extract playlist id
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
         # Extract playlist id
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:

-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)

         playlist_id = mobj.group(1) or mobj.group(2)
 
         # Check if it's a video-specific URL
         playlist_id = mobj.group(1) or mobj.group(2)
 
         # Check if it's a video-specific URL


@@ -1001,17 +1241,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         if 'v' in query_dict:
             video_id = query_dict['v'][0]
             if self._downloader.params.get('noplaylist'):
         if 'v' in query_dict:
             video_id = query_dict['v'][0]
             if self._downloader.params.get('noplaylist'):

-                self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id)
+                self.to_screen('Downloading just video %s because of --no-playlist' % video_id)

                 return self.url_result(video_id, 'Youtube', video_id=video_id)
             else:
                 return self.url_result(video_id, 'Youtube', video_id=video_id)
             else:

-                self.to_screen(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
+                self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))

 
         if playlist_id.startswith('RD'):
             # Mixes require a custom extraction process
             return self._extract_mix(playlist_id)
 
         if playlist_id.startswith('RD'):
             # Mixes require a custom extraction process
             return self._extract_mix(playlist_id)

-        if playlist_id.startswith('TL'):
-            raise ExtractorError(u'For downloading YouTube.com top lists, use '
-                u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)

 
         url = self._TEMPLATE_URL % playlist_id
         page = self._download_webpage(url, playlist_id)
 
         url = self._TEMPLATE_URL % playlist_id
         page = self._download_webpage(url, playlist_id)


@@ -1020,7 +1257,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         # Check if the playlist exists or is private
         if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
             raise ExtractorError(
         # Check if the playlist exists or is private
         if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
             raise ExtractorError(

-                u'The playlist doesn\'t exist or is private, use --username or '
+                'The playlist doesn\'t exist or is private, use --username or '

                 '--netrc to access it.',
                 expected=True)
 
                 '--netrc to access it.',
                 expected=True)
 


@@ -1043,54 +1280,29 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
                 'Downloading page #%s' % page_num,
                 transform_source=uppercase_escape)
             content_html = more['content_html']
                 'Downloading page #%s' % page_num,
                 transform_source=uppercase_escape)
             content_html = more['content_html']

+            if not content_html.strip():
+                # Some webpages show a "Load more" button but they don't
+                # have more videos
+                break

             more_widget_html = more['load_more_widget_html']
 
         playlist_title = self._html_search_regex(
             r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
             more_widget_html = more['load_more_widget_html']
 
         playlist_title = self._html_search_regex(
             r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',

-            page, u'title')
+            page, 'title')

 
         url_results = self._ids_to_results(ids)
         return self.playlist_result(url_results, playlist_id, playlist_title)
 
 
 
         url_results = self._ids_to_results(ids)
         return self.playlist_result(url_results, playlist_id, playlist_title)
 
 

-class YoutubeTopListIE(YoutubePlaylistIE):
-    IE_NAME = u'youtube:toplist'
-    IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"'
-        u' (Example: "yttoplist:music:Top Tracks")')
-    _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        channel = mobj.group('chann')
-        title = mobj.group('title')
-        query = compat_urllib_parse.urlencode({'title': title})
-        playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query)
-        channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title)
-        link = self._html_search_regex(playlist_re, channel_page, u'list')
-        url = compat_urlparse.urljoin('https://www.youtube.com/', link)
-        
-        video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
-        ids = []
-        # sometimes the webpage doesn't contain the videos
-        # retry until we get them
-        for i in itertools.count(0):
-            msg = u'Downloading Youtube mix'
-            if i > 0:
-                msg += ', retry #%d' % i
-            webpage = self._download_webpage(url, title, msg)
-            ids = orderedSet(re.findall(video_re, webpage))
-            if ids:
-                break
-        url_results = self._ids_to_results(ids)
-        return self.playlist_result(url_results, playlist_title=title)
-
-

 class YoutubeChannelIE(InfoExtractor):
 class YoutubeChannelIE(InfoExtractor):

-    IE_DESC = u'YouTube.com channels'
-    _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
-    _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
-    _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
-    IE_NAME = u'youtube:channel'
+    IE_DESC = 'YouTube.com channels'
+    _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
+    IE_NAME = 'youtube:channel'
+    _TESTS = [{
+        'note': 'paginated channel',
+        'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
+        'playlist_mincount': 91,
+    }]

 
     def extract_videos_from_page(self, page):
         ids_in_page = []
 
     def extract_videos_from_page(self, page):
         ids_in_page = []


@@ -1100,13 +1312,8 @@ class YoutubeChannelIE(InfoExtractor):
         return ids_in_page
 
     def _real_extract(self, url):
         return ids_in_page
 
     def _real_extract(self, url):

-        # Extract channel id
-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+        channel_id = self._match_id(url)

 
 

-        # Download channel page
-        channel_id = mobj.group(1)

         video_ids = []
         url = 'https://www.youtube.com/channel/%s/videos' % channel_id
         channel_page = self._download_webpage(url, channel_id)
         video_ids = []
         url = 'https://www.youtube.com/channel/%s/videos' % channel_id
         channel_page = self._download_webpage(url, channel_id)


@@ -1120,50 +1327,67 @@ class YoutubeChannelIE(InfoExtractor):
             # The videos are contained in a single page
             # the ajax pages can't be used, they are empty
             video_ids = self.extract_videos_from_page(channel_page)
             # The videos are contained in a single page
             # the ajax pages can't be used, they are empty
             video_ids = self.extract_videos_from_page(channel_page)

-        else:
-            # Download all channel pages using the json-based channel_ajax query
+            entries = [
+                self.url_result(video_id, 'Youtube', video_id=video_id)
+                for video_id in video_ids]
+            return self.playlist_result(entries, channel_id)
+
+        def _entries():
+            more_widget_html = content_html = channel_page

             for pagenum in itertools.count(1):
             for pagenum in itertools.count(1):

-                url = self._MORE_PAGES_URL % (pagenum, channel_id)
-                page = self._download_json(
-                    url, channel_id, note=u'Downloading page #%s' % pagenum,
-                    transform_source=uppercase_escape)

 
 

-                ids_in_page = self.extract_videos_from_page(page['content_html'])
-                video_ids.extend(ids_in_page)
-    
-                if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
+                ids_in_page = self.extract_videos_from_page(content_html)
+                for video_id in ids_in_page:
+                    yield self.url_result(
+                        video_id, 'Youtube', video_id=video_id)
+
+                mobj = re.search(
+                    r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
+                    more_widget_html)
+                if not mobj:

                     break
 
                     break
 

-        self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
+                more = self._download_json(
+                    'https://youtube.com/%s' % mobj.group('more'), channel_id,
+                    'Downloading page #%s' % (pagenum + 1),
+                    transform_source=uppercase_escape)
+                content_html = more['content_html']
+                more_widget_html = more['load_more_widget_html']

 
 

-        url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
-                       for video_id in video_ids]
-        return self.playlist_result(url_entries, channel_id)
+        return self.playlist_result(_entries(), channel_id)

 
 
 class YoutubeUserIE(InfoExtractor):
 
 
 class YoutubeUserIE(InfoExtractor):

-    IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
+    IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
+    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'

     _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
     _GDATA_PAGE_SIZE = 50
     _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
     _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
     _GDATA_PAGE_SIZE = 50
     _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'

-    IE_NAME = u'youtube:user'
+    IE_NAME = 'youtube:user'
+
+    _TESTS = [{
+        'url': 'https://www.youtube.com/user/TheLinuxFoundation',
+        'playlist_mincount': 320,
+        'info_dict': {
+            'title': 'TheLinuxFoundation',
+        }
+    }, {
+        'url': 'ytuser:phihag',
+        'only_matching': True,
+    }]

 
     @classmethod
     def suitable(cls, url):
         # Don't return True if the url can be extracted with other youtube
         # extractor, the regex would is too permissive and it would match.
         other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
 
     @classmethod
     def suitable(cls, url):
         # Don't return True if the url can be extracted with other youtube
         # extractor, the regex would is too permissive and it would match.
         other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)

-        if any(ie.suitable(url) for ie in other_ies): return False
-        else: return super(YoutubeUserIE, cls).suitable(url)
+        if any(ie.suitable(url) for ie in other_ies):
+            return False
+        else:
+            return super(YoutubeUserIE, cls).suitable(url)

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        # Extract username
-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-
-        username = mobj.group(1)
+        username = self._match_id(url)

 
         # Download video ids using YouTube Data API. Result size per
         # query is limited (currently to 50 videos) so we need to query
 
         # Download video ids using YouTube Data API. Result size per
         # query is limited (currently to 50 videos) so we need to query


@@ -1176,13 +1400,13 @@ class YoutubeUserIE(InfoExtractor):
             gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
             page = self._download_webpage(
                 gdata_url, username,
             gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
             page = self._download_webpage(
                 gdata_url, username,

-                u'Downloading video ids from %d to %d' % (
+                'Downloading video ids from %d to %d' % (

                     start_index, start_index + self._GDATA_PAGE_SIZE))
 
             try:
                 response = json.loads(page)
             except ValueError as err:
                     start_index, start_index + self._GDATA_PAGE_SIZE))
 
             try:
                 response = json.loads(page)
             except ValueError as err:

-                raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
+                raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))

             if 'entry' not in response['feed']:
                 return
 
             if 'entry' not in response['feed']:
                 return
 


@@ -1198,16 +1422,16 @@ class YoutubeUserIE(InfoExtractor):
                     'id': video_id,
                     'title': title,
                 }
                     'id': video_id,
                     'title': title,
                 }

-        url_results = PagedList(download_page, self._GDATA_PAGE_SIZE)
+        url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)

 
         return self.playlist_result(url_results, playlist_title=username)
 
 
 class YoutubeSearchIE(SearchInfoExtractor):
 
         return self.playlist_result(url_results, playlist_title=username)
 
 
 class YoutubeSearchIE(SearchInfoExtractor):

-    IE_DESC = u'YouTube.com searches'
-    _API_URL = u'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
+    IE_DESC = 'YouTube.com searches'
+    _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'

     _MAX_RESULTS = 1000
     _MAX_RESULTS = 1000

-    IE_NAME = u'youtube:search'
+    IE_NAME = 'youtube:search'

     _SEARCH_KEY = 'ytsearch'
 
     def _get_n_results(self, query, n):
     _SEARCH_KEY = 'ytsearch'
 
     def _get_n_results(self, query, n):


@@ -1223,15 +1447,15 @@ class YoutubeSearchIE(SearchInfoExtractor):
                 compat_urllib_parse.quote_plus(query.encode('utf-8')),
                 (PAGE_SIZE * pagenum) + 1)
             data_json = self._download_webpage(
                 compat_urllib_parse.quote_plus(query.encode('utf-8')),
                 (PAGE_SIZE * pagenum) + 1)
             data_json = self._download_webpage(

-                result_url, video_id=u'query "%s"' % query,
-                note=u'Downloading page %s' % (pagenum + 1),
-                errnote=u'Unable to download API page')
+                result_url, video_id='query "%s"' % query,
+                note='Downloading page %s' % (pagenum + 1),
+                errnote='Unable to download API page')

             data = json.loads(data_json)
             api_response = data['data']
 
             if 'items' not in api_response:
                 raise ExtractorError(
             data = json.loads(data_json)
             api_response = data['data']
 
             if 'items' not in api_response:
                 raise ExtractorError(

-                    u'[youtube] No video results', expected=True)
+                    '[youtube] No video results', expected=True)

 
             new_ids = list(video['id'] for video in api_response['items'])
             video_ids += new_ids
 
             new_ids = list(video['id'] for video in api_response['items'])
             video_ids += new_ids


@@ -1250,13 +1474,20 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
     _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
     _SEARCH_KEY = 'ytsearchdate'
     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
     _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
     _SEARCH_KEY = 'ytsearchdate'

-    IE_DESC = u'YouTube.com searches, newest videos first'
+    IE_DESC = 'YouTube.com searches, newest videos first'

 
 
 class YoutubeSearchURLIE(InfoExtractor):
 
 
 class YoutubeSearchURLIE(InfoExtractor):

-    IE_DESC = u'YouTube.com search URLs'
-    IE_NAME = u'youtube:search_url'
+    IE_DESC = 'YouTube.com search URLs'
+    IE_NAME = 'youtube:search_url'

     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'

+    _TESTS = [{
+        'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
+        'playlist_mincount': 5,
+        'info_dict': {
+            'title': 'youtube-dl test video',
+        }
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)


@@ -1264,7 +1495,7 @@ class YoutubeSearchURLIE(InfoExtractor):
 
         webpage = self._download_webpage(url, query)
         result_code = self._search_regex(
 
         webpage = self._download_webpage(url, query)
         result_code = self._search_regex(

-            r'(?s)<ol class="item-section"(.*?)</ol>', webpage, u'result HTML')
+            r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')

 
         part_codes = re.findall(
             r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
 
         part_codes = re.findall(
             r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)


@@ -1290,18 +1521,39 @@ class YoutubeSearchURLIE(InfoExtractor):
 
 
 class YoutubeShowIE(InfoExtractor):
 
 
 class YoutubeShowIE(InfoExtractor):

-    IE_DESC = u'YouTube.com (multi-season) shows'
-    _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
-    IE_NAME = u'youtube:show'
+    IE_DESC = 'YouTube.com (multi-season) shows'
+    _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
+    IE_NAME = 'youtube:show'
+    _TESTS = [{
+        'url': 'http://www.youtube.com/show/airdisasters',
+        'playlist_mincount': 3,
+        'info_dict': {
+            'id': 'airdisasters',
+            'title': 'Air Disasters',
+        }
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        show_name = mobj.group(1)
-        webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
+        playlist_id = mobj.group('id')
+        webpage = self._download_webpage(
+            url, playlist_id, 'Downloading show webpage')

         # There's one playlist for each season of the show
         m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
         # There's one playlist for each season of the show
         m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))

-        self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
-        return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
+        self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
+        entries = [
+            self.url_result(
+                'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
+            for season in m_seasons
+        ]
+        title = self._og_search_title(webpage, fatal=False)
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': title,
+            'entries': entries,
+        }

 
 
 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
 
 
 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):


@@ -1323,7 +1575,7 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
 
     @property
     def IE_NAME(self):
 
     @property
     def IE_NAME(self):

-        return u'youtube:%s' % self._FEED_NAME
+        return 'youtube:%s' % self._FEED_NAME

 
     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()


@@ -1332,10 +1584,13 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
         feed_entries = []
         paging = 0
         for i in itertools.count(1):
         feed_entries = []
         paging = 0
         for i in itertools.count(1):

-            info = self._download_json(self._FEED_TEMPLATE % paging,
-                                          u'%s feed' % self._FEED_NAME,
-                                          u'Downloading page %s' % i)
+            info = self._download_json(
+                self._FEED_TEMPLATE % paging,
+                '%s feed' % self._FEED_NAME,
+                'Downloading page %s' % i,
+                transform_source=uppercase_escape)

             feed_html = info.get('feed_html') or info.get('content_html')
             feed_html = info.get('feed_html') or info.get('content_html')

+            load_more_widget_html = info.get('load_more_widget_html') or feed_html

             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
             ids = orderedSet(m.group(1) for m in m_ids)
             feed_entries.extend(
             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
             ids = orderedSet(m.group(1) for m in m_ids)
             feed_entries.extend(


@@ -1343,50 +1598,86 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
                 for video_id in ids)
             mobj = re.search(
                 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
                 for video_id in ids)
             mobj = re.search(
                 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',

-                feed_html)
+                load_more_widget_html)

             if mobj is None:
                 break
             paging = mobj.group('paging')
         return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
 
             if mobj is None:
                 break
             paging = mobj.group('paging')
         return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
 

-class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
-    IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
-    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
-    _FEED_NAME = 'subscriptions'
-    _PLAYLIST_TITLE = u'Youtube Subscriptions'

 
 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
 
 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):

-    IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
+    IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'

     _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
     _FEED_NAME = 'recommended'
     _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
     _FEED_NAME = 'recommended'

-    _PLAYLIST_TITLE = u'Youtube Recommended videos'
+    _PLAYLIST_TITLE = 'Youtube Recommended videos'
+

 
 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
 
 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):

-    IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
+    IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'

     _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
     _FEED_NAME = 'watch_later'
     _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
     _FEED_NAME = 'watch_later'

-    _PLAYLIST_TITLE = u'Youtube Watch Later'
+    _PLAYLIST_TITLE = 'Youtube Watch Later'

     _PERSONAL_FEED = True
 
     _PERSONAL_FEED = True
 

+

 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):

-    IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)'
-    _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory'
+    IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
+    _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'

     _FEED_NAME = 'history'
     _PERSONAL_FEED = True
     _FEED_NAME = 'history'
     _PERSONAL_FEED = True

-    _PLAYLIST_TITLE = u'Youtube Watch History'
+    _PLAYLIST_TITLE = 'Youtube Watch History'
+

 
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
 
 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):

-    IE_NAME = u'youtube:favorites'
-    IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
+    IE_NAME = 'youtube:favorites'
+    IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'

     _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
     _LOGIN_REQUIRED = True
 
     def _real_extract(self, url):
         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
     _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
     _LOGIN_REQUIRED = True
 
     def _real_extract(self, url):
         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')

-        playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
+        playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')

         return self.url_result(playlist_id, 'YoutubePlaylist')
 
 
         return self.url_result(playlist_id, 'YoutubePlaylist')
 
 

+class YoutubeSubscriptionsIE(YoutubePlaylistIE):
+    IE_NAME = 'youtube:subscriptions'
+    IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
+    _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
+    _TESTS = []
+
+    def _real_extract(self, url):
+        title = 'Youtube Subscriptions'
+        page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
+
+        # The extraction process is the same as for playlists, but the regex
+        # for the video ids doesn't contain an index
+        ids = []
+        more_widget_html = content_html = page
+
+        for page_num in itertools.count(1):
+            matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
+            new_ids = orderedSet(matches)
+            ids.extend(new_ids)
+
+            mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
+            if not mobj:
+                break
+
+            more = self._download_json(
+                'https://youtube.com/%s' % mobj.group('more'), title,
+                'Downloading page #%s' % page_num,
+                transform_source=uppercase_escape)
+            content_html = more['content_html']
+            more_widget_html = more['load_more_widget_html']
+
+        return {
+            '_type': 'playlist',
+            'title': title,
+            'entries': self._ids_to_results(ids),
+        }
+
+

 class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
 class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list


@@ -1408,9 +1699,26 @@ class YoutubeTruncatedURLIE(InfoExtractor):
 
     def _real_extract(self, url):
         raise ExtractorError(
 
     def _real_extract(self, url):
         raise ExtractorError(

-            u'Did you forget to quote the URL? Remember that & is a meta '
-            u'character in most shells, so you want to put the URL in quotes, '
-            u'like  youtube-dl '
-            u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
-            u' or simply  youtube-dl BaW_jenozKc  .',
+            'Did you forget to quote the URL? Remember that & is a meta '
+            'character in most shells, so you want to put the URL in quotes, '
+            'like  youtube-dl '
+            '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
+            ' or simply  youtube-dl BaW_jenozKc  .',
+            expected=True)
+
+
+class YoutubeTruncatedIDIE(InfoExtractor):
+    IE_NAME = 'youtube:truncated_id'
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'https?://(?:www\.)youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
+
+    _TESTS = [{
+        'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        raise ExtractorError(
+            'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),

             expected=True)
             expected=True)