Merge branch 'subtitles-rework'

[youtube-dl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 823d6aaf3a4f38ad5696e67a23b14e6fe690fabd..22db896b16066bff193bc1ef7eddab214b9440a9 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -11,7 +11,6 @@ import time
 import traceback
 
 from .common import InfoExtractor, SearchInfoExtractor
 import traceback
 
 from .common import InfoExtractor, SearchInfoExtractor

-from .subtitles import SubtitlesInfoExtractor

 from ..jsinterp import JSInterpreter
 from ..swfinterp import SWFInterpreter
 from ..compat import (
 from ..jsinterp import JSInterpreter
 from ..swfinterp import SWFInterpreter
 from ..compat import (


@@ -25,6 +24,7 @@ from ..compat import (
 from ..utils import (
     clean_html,
     ExtractorError,
 from ..utils import (
     clean_html,
     ExtractorError,

+    float_or_none,

     get_element_by_attribute,
     get_element_by_id,
     int_or_none,
     get_element_by_attribute,
     get_element_by_id,
     int_or_none,


@@ -184,7 +184,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             return
 
 
             return
 
 

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

     IE_DESC = 'YouTube.com'
     _VALID_URL = r"""(?x)^
                      (
     IE_DESC = 'YouTube.com'
     _VALID_URL = r"""(?x)^
                      (


@@ -256,7 +256,7 @@ 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},
         '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
         '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'},


@@ -264,9 +264,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
 
         # Dash mp4 audio
         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
 
         # 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, 'container': 'm4a_dash'},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},

 
         # 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},


@@ -287,7 +287,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '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'},
         '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'},
         '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
         '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
 
         # Dash webm audio
         '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},


@@ -392,6 +394,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'format': '141',
             },
         },
                 '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',
         # Controversy video
         {
             'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',


@@ -412,12 +431,25 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'id': 'HtVdAasjOgU',
                 'ext': 'mp4',
                 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
                 'id': 'HtVdAasjOgU',
                 'ext': 'mp4',
                 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',

-                'description': 'md5:eca57043abae25130f58f655ad9a7771',
+                'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',

                 'uploader': 'The Witcher',
                 'uploader_id': 'WitcherGame',
                 'upload_date': '20140605',
             },
         },
                 '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',
         # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
         {
             'url': '__2ABJjxzNo',


@@ -450,6 +482,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'skip_download': 'requires avconv',
             }
         },
                 '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',
+            },
+        }

     ]
 
     def __init__(self, *args, **kwargs):
     ]
 
     def __init__(self, *args, **kwargs):


@@ -478,7 +524,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     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)


@@ -494,26 +540,30 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         if cache_spec is not None:
             return lambda s: ''.join(s[i] for i in cache_spec)
 
         if cache_spec is not None:
             return lambda s: ''.join(s[i] for i in cache_spec)
 

+        download_note = (
+            'Downloading player %s' % player_url
+            if self._downloader.params.get('verbose') else
+            'Downloading %s player %s' % (player_type, player_id)
+        )

         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='Downloading %s player %s' % (player_type, player_id),
+                note=download_note,

                 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,
                 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,

-                note='Downloading %s player %s' % (player_type, player_id),
+                note=download_note,

                 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
 
                 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
 

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


@@ -527,8 +577,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 return 's[%s%s%s]' % (starts, ends, steps)
 
             step = None
                 return 's[%s%s%s]' % (starts, ends, steps)
 
             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:


@@ -559,7 +609,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(
 
     def _parse_sig_js(self, jscode):
         funcname = self._search_regex(

-            r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
+            r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,

             'Initial JS player signature function name')
 
         jsi = JSInterpreter(jscode)
             'Initial JS player signature function name')
 
         jsi = JSInterpreter(jscode)


@@ -597,38 +647,41 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             raise ExtractorError(
                 'Signature extraction failed: ' + tb, cause=e)
 
             raise ExtractorError(
                 'Signature extraction failed: ' + tb, cause=e)
 

-    def _get_available_subtitles(self, video_id, webpage):
+    def _get_subtitles(self, video_id, webpage):

         try:
         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:
             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
             return {}
                 '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('unable to download video subtitles: %s' % compat_str(err))
             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
             if lang in sub_lang_list:
                 continue

-            params = compat_urllib_parse.urlencode({
-                'lang': lang,
-                'v': video_id,
-                'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
-                'name': unescapeHTML(l[0]).encode('utf-8'),
-            })
-            url = 'https://www.youtube.com/api/timedtext?' + params
-            sub_lang_list[lang] = url
+            sub_formats = []
+            for ext in ['sbv', 'vtt', 'srt']:
+                params = compat_urllib_parse.urlencode({
+                    'lang': lang,
+                    'v': video_id,
+                    'fmt': ext,
+                    'name': track.attrib['name'].encode('utf-8'),
+                })
+                sub_formats.append({
+                    'url': 'https://www.youtube.com/api/timedtext?' + params,
+                    'ext': ext,
+                })
+            sub_lang_list[lang] = sub_formats

         if not sub_lang_list:
             self._downloader.report_warning('video doesn\'t have subtitles')
             return {}
         return sub_lang_list
 
         if not sub_lang_list:
             self._downloader.report_warning('video doesn\'t have subtitles')
             return {}
         return sub_lang_list
 

-    def _get_available_automatic_caption(self, video_id, webpage):
+    def _get_automatic_captions(self, video_id, webpage):

         """We need the webpage for getting the captions url, pass it as an
            argument to speed up the process."""
         """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('%s: Looking for automatic captions' % video_id)
         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
         err_msg = 'Couldn\'t find automatic captions for %s' % video_id
         self.to_screen('%s: Looking for automatic captions' % video_id)
         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
         err_msg = 'Couldn\'t find automatic captions for %s' % video_id


@@ -649,22 +702,29 @@ 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':
+            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']
                 self._downloader.report_warning('Video doesn\'t have automatic captions')
                 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 = lang_node.attrib['lang_code']
 
             sub_lang_list = {}
             for lang_node in caption_list.findall('target'):
                 sub_lang = lang_node.attrib['lang_code']

-                params = compat_urllib_parse.urlencode({
-                    'lang': original_lang,
-                    'tlang': sub_lang,
-                    'fmt': sub_format,
-                    'ts': timestamp,
-                    'kind': 'asr',
-                })
-                sub_lang_list[sub_lang] = caption_url + '&' + params
+                sub_formats = []
+                for ext in ['sbv', 'vtt', 'srt']:
+                    params = compat_urllib_parse.urlencode({
+                        'lang': original_lang,
+                        'tlang': sub_lang,
+                        'fmt': ext,
+                        'ts': timestamp,
+                        'kind': caption_kind,
+                    })
+                    sub_formats.append({
+                        'url': caption_url + '&' + params,
+                        'ext': ext,
+                    })
+                sub_lang_list[sub_lang] = sub_formats

             return sub_lang_list
         # An extractor error can be raise by the download process if there are
         # no automatic captions but there are subtitles
             return sub_lang_list
         # An extractor error can be raise by the download process if there are
         # no automatic captions but there are subtitles


@@ -723,6 +783,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'format_id': format_id,
                 'url': video_url,
                 'width': int_or_none(r.attrib.get('width')),
                 '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,
                 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
                 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
                 'filesize': filesize,


@@ -733,8 +794,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     fo for fo in formats
                     if fo['format_id'] == format_id)
             except StopIteration:
                     fo for fo in formats
                     if fo['format_id'] == format_id)
             except StopIteration:

-                f.update(self._formats.get(format_id, {}))
-                formats.append(f)
+                full_info = self._formats.get(format_id, {}).copy()
+                full_info.update(f)
+                formats.append(full_info)

             else:
                 existing_format.update(f)
         return formats
             else:
                 existing_format.update(f)
         return formats


@@ -762,15 +824,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             player_url = None
 
         # Get video info
             player_url = None
 
         # Get video info

+        embed_webpage = None

         if re.search(r'player-age-gate-content">', video_webpage) is not None:
             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
         if re.search(r'player-age-gate-content">', video_webpage) is not None:
             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', default=''),
+                    r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),

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


@@ -914,10 +979,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, video_webpage)
 
         # subtitles
         video_subtitles = self.extract_subtitles(video_id, video_webpage)

-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, video_webpage)
-            return
+        automatic_captions = self.extract_automatic_captions(video_id, video_webpage)

 
         if 'length_seconds' not in video_info:
             self._downloader.report_warning('unable to extract video duration')
 
         if 'length_seconds' not in video_info:
             self._downloader.report_warning('unable to extract video duration')


@@ -967,12 +1029,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     url += '&signature=' + url_data['sig'][0]
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]
                     url += '&signature=' + url_data['sig'][0]
                 elif 's' in url_data:
                     encrypted_sig = url_data['s'][0]

-
-                    if not age_gate:
+                    ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
+
+                    jsplayer_url_json = self._search_regex(
+                        ASSETS_RE,
+                        embed_webpage if age_gate else video_webpage,
+                        'JS player URL (1)', default=None)
+                    if not jsplayer_url_json and not age_gate:
+                        # We need the embed website after all
+                        if embed_webpage is None:
+                            embed_url = proto + '://www.youtube.com/embed/%s' % video_id
+                            embed_webpage = self._download_webpage(
+                                embed_url, video_id, 'Downloading embed webpage')

                         jsplayer_url_json = self._search_regex(
                         jsplayer_url_json = self._search_regex(

-                            r'"assets":.+?"js":\s*("[^"]+")',
-                            video_webpage, 'JS player URL')
-                        player_url = json.loads(jsplayer_url_json)
+                            ASSETS_RE, embed_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*("[^"]+")',


@@ -1026,8 +1098,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     self.report_warning(
                         'Skipping DASH manifest: %r' % e, video_id)
                 else:
                     self.report_warning(
                         'Skipping DASH manifest: %r' % e, video_id)
                 else:

+                    # 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)
 
                     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 {


@@ -1040,6 +1128,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             'description': video_description,
             'categories': video_categories,
             'subtitles': video_subtitles,
             'description': video_description,
             'categories': video_categories,
             'subtitles': video_subtitles,

+            'automatic_captions': automatic_captions,

             'duration': video_duration,
             'age_limit': 18 if age_gate else 0,
             'annotations': video_annotations,
             'duration': video_duration,
             'age_limit': 18 if age_gate else 0,
             'annotations': video_annotations,


@@ -1047,6 +1136,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,

+            'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),

             'formats': formats,
         }
 
             'formats': formats,
         }
 


@@ -1084,6 +1174,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
     }, {
         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
         'info_dict': {
     }, {
         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
         'info_dict': {

+            'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',

             'title': 'YDL_Empty_List',
         },
         'playlist_count': 0,
             'title': 'YDL_Empty_List',
         },
         'playlist_count': 0,


@@ -1092,6 +1183,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         'info_dict': {
             'title': '29C3: Not my department',
         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
         'info_dict': {
             'title': '29C3: Not my department',

+            'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',

         },
         'playlist_count': 95,
     }, {
         },
         'playlist_count': 95,
     }, {


@@ -1099,6 +1191,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'PLBB231211A4F62143',
         'info_dict': {
             'title': '[OLD]Team Fortress 2 (Class-based LP)',
         'url': 'PLBB231211A4F62143',
         'info_dict': {
             'title': '[OLD]Team Fortress 2 (Class-based LP)',

+            'id': 'PLBB231211A4F62143',

         },
         'playlist_mincount': 26,
     }, {
         },
         'playlist_mincount': 26,
     }, {


@@ -1106,12 +1199,14 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
         'info_dict': {
             'title': 'Uploads from Cauchemar',
         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
         'info_dict': {
             'title': 'Uploads from Cauchemar',

+            'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',

         },
         'playlist_mincount': 799,
     }, {
         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         'info_dict': {
             'title': 'YDL_safe_search',
         },
         'playlist_mincount': 799,
     }, {
         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
         'info_dict': {
             'title': 'YDL_safe_search',

+            'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',

         },
         'playlist_count': 2,
     }, {
         },
         'playlist_count': 2,
     }, {


@@ -1120,6 +1215,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA15',
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA15',

+            'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',

         }
     }, {
         'note': 'Embedded SWF player',
         }
     }, {
         'note': 'Embedded SWF player',


@@ -1127,7 +1223,16 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA7',
         'playlist_count': 4,
         'info_dict': {
             'title': 'JODA7',

+            'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',

         }
         }

+    }, {
+        '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',
+            'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
+        },
+        'playlist_mincout': 21,

     }]
 
     def _real_initialize(self):
     }]
 
     def _real_initialize(self):


@@ -1178,9 +1283,6 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         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('For downloading YouTube.com top lists, use '
-                                 '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)


@@ -1212,6 +1314,10 @@ 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(
             more_widget_html = more['load_more_widget_html']
 
         playlist_title = self._html_search_regex(


@@ -1222,49 +1328,6 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
         return self.playlist_result(url_results, playlist_id, playlist_title)
 
 
         return self.playlist_result(url_results, playlist_id, playlist_title)
 
 

-class YoutubeTopListIE(YoutubePlaylistIE):
-    IE_NAME = 'youtube:toplist'
-    IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
-               ' (Example: "yttoplist:music:Top Tracks")')
-    _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
-    _TESTS = [{
-        'url': 'yttoplist:music:Trending',
-        'playlist_mincount': 5,
-        'skip': 'Only works for logged-in users',
-    }]
-
-    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})
-        channel_page = self._download_webpage(
-            'https://www.youtube.com/%s' % channel, title)
-        link = self._html_search_regex(
-            r'''(?x)
-                <a\s+href="([^"]+)".*?>\s*
-                <span\s+class="branded-page-module-title-text">\s*
-                <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
-            channel_page, '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 = '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):
     IE_DESC = 'YouTube.com channels'
     _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
 class YoutubeChannelIE(InfoExtractor):
     IE_DESC = 'YouTube.com channels'
     _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'


@@ -1273,6 +1336,9 @@ class YoutubeChannelIE(InfoExtractor):
         'note': 'paginated channel',
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,
         'note': 'paginated channel',
         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
         'playlist_mincount': 91,

+        'info_dict': {
+            'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
+        }

     }]
 
     def extract_videos_from_page(self, page):
     }]
 
     def extract_videos_from_page(self, page):


@@ -1555,9 +1621,11 @@ 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,
-                                       '%s feed' % self._FEED_NAME,
-                                       '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')
             load_more_widget_html = info.get('load_more_widget_html') or feed_html
             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_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)


@@ -1651,11 +1719,18 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)
     IE_NAME = 'youtube:truncated_url'
     IE_DESC = False  # Do not list
     _VALID_URL = r'''(?x)

-        (?:https?://)?[^/]+/watch\?(?:
+        (?:https?://)?
+        (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
+        (?:watch\?(?:

             feature=[a-z_]+|
             feature=[a-z_]+|

-            annotation_id=annotation_[^&]+
-        )?$|
-        (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
+            annotation_id=annotation_[^&]+|
+            x-yt-cl=[0-9]+|
+            hl=[^&]*|
+        )?
+        |
+            attribution_link\?a=[^&]+
+        )
+        $

     '''
 
     _TESTS = [{
     '''
 
     _TESTS = [{


@@ -1664,6 +1739,15 @@ class YoutubeTruncatedURLIE(InfoExtractor):
     }, {
         'url': 'http://www.youtube.com/watch?',
         'only_matching': True,
     }, {
         'url': 'http://www.youtube.com/watch?',
         'only_matching': True,

+    }, {
+        'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?feature=foo',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/watch?hl=en-GB',
+        'only_matching': True,

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


@@ -1674,3 +1758,20 @@ class YoutubeTruncatedURLIE(InfoExtractor):
             '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
             ' or simply  youtube-dl BaW_jenozKc  .',
             expected=True)
             '"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)