Update upstream source from tag 'upstream/2018.04.25'

[youtubedl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index ad2e933ee4e34c9ebdb982ca66278e6e4c4a06b0..e7bd1f18fb504c8541d0fbdc301980cab07d1f54 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -87,7 +87,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         (username, password) = self._get_login_info()
         # No authentication to be performed
         if username is None:
         (username, password) = self._get_login_info()
         # No authentication to be performed
         if username is None:

-            if self._LOGIN_REQUIRED:
+            if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:

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


@@ -332,6 +332,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                          (?:(?:(?:(?:\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/|

+                            (?:www\.)?hooktube\.com/|

                             (?:www\.)?yourepeat\.com/|
                             tube\.majestyc\.net/|
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                             (?:www\.)?yourepeat\.com/|
                             tube\.majestyc\.net/|
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains


@@ -1390,7 +1391,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             )
             (["\'])
                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
             )
             (["\'])
                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/

-                (?:embed|v|p)/.+?)
+                (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)

             \1''', webpage)]
 
         # lazyYT YouTube embed
             \1''', webpage)]
 
         # lazyYT YouTube embed


@@ -1595,6 +1596,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         if 'token' not in video_info:
                             video_info = get_video_info
                         break
                         if 'token' not in video_info:
                             video_info = get_video_info
                         break

+
+        def extract_unavailable_message():
+            return self._html_search_regex(
+                r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
+                video_webpage, 'unavailable message', default=None)
+

         if 'token' not in video_info:
             if 'reason' in video_info:
                 if 'The uploader has not made this video available in your country.' in video_info['reason']:
         if 'token' not in video_info:
             if 'reason' in video_info:
                 if 'The uploader has not made this video available in your country.' in video_info['reason']:


@@ -1603,8 +1610,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     countries = regions_allowed.split(',') if regions_allowed else None
                     self.raise_geo_restricted(
                         msg=video_info['reason'][0], countries=countries)
                     countries = regions_allowed.split(',') if regions_allowed else None
                     self.raise_geo_restricted(
                         msg=video_info['reason'][0], countries=countries)

+                reason = video_info['reason'][0]
+                if 'Invalid parameters' in reason:
+                    unavailable_message = extract_unavailable_message()
+                    if unavailable_message:
+                        reason = unavailable_message

                 raise ExtractorError(
                 raise ExtractorError(

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

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


@@ -1621,6 +1633,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         # description
         description_original = video_description = get_element_by_id("eow-description", video_webpage)
         if video_description:
         # description
         description_original = video_description = get_element_by_id("eow-description", video_webpage)
         if video_description:

+
+            def replace_url(m):
+                redir_url = compat_urlparse.urljoin(url, m.group(1))
+                parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
+                if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
+                    qs = compat_parse_qs(parsed_redir_url.query)
+                    q = qs.get('q')
+                    if q and q[0]:
+                        return q[0]
+                return redir_url
+

             description_original = video_description = re.sub(r'''(?x)
                 <a\s+
                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
             description_original = video_description = re.sub(r'''(?x)
                 <a\s+
                     (?:[a-zA-Z-]+="[^"]*"\s+)*?


@@ -1629,7 +1652,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     class="[^"]*"[^>]*>
                 [^<]+\.{3}\s*
                 </a>
                     class="[^"]*"[^>]*>
                 [^<]+\.{3}\s*
                 </a>

-            ''', r'\1', video_description)
+            ''', replace_url, video_description)

             video_description = clean_html(video_description)
         else:
             fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
             video_description = clean_html(video_description)
         else:
             fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)


@@ -1682,7 +1705,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         video_uploader_id = None
         video_uploader_url = None
         mobj = re.search(
         video_uploader_id = None
         video_uploader_url = None
         mobj = re.search(

-            r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
+            r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',

             video_webpage)
         if mobj is not None:
             video_uploader_id = mobj.group('uploader_id')
             video_webpage)
         if mobj is not None:
             video_uploader_id = mobj.group('uploader_id')


@@ -1798,7 +1821,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 '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', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
+        elif not is_live and (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)
             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)


@@ -1921,6 +1944,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                     break
                             if codecs:
                                 dct.update(parse_codecs(codecs))
                                     break
                             if codecs:
                                 dct.update(parse_codecs(codecs))

+                if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
+                    dct['downloader_options'] = {
+                        # Youtube throttles chunks >~10M
+                        'http_chunk_size': 10485760,
+                    }

                 formats.append(dct)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]
                 formats.append(dct)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]


@@ -1941,9 +1969,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
                 formats.append(a_format)
         else:
                 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
                 formats.append(a_format)
         else:

-            unavailable_message = self._html_search_regex(
-                r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
-                video_webpage, 'unavailable message', default=None)
+            unavailable_message = extract_unavailable_message()

             if unavailable_message:
                 raise ExtractorError(unavailable_message, expected=True)
             raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
             if unavailable_message:
                 raise ExtractorError(unavailable_message, expected=True)
             raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')


@@ -2039,39 +2065,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         }
 
 
         }
 
 

-class YoutubeSharedVideoIE(InfoExtractor):
-    _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
-    IE_NAME = 'youtube:shared'
-
-    _TEST = {
-        'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
-        'info_dict': {
-            'id': 'uPDB5I9wfp8',
-            'ext': 'webm',
-            'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
-            'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
-            'upload_date': '20160219',
-            'uploader': 'Pocoyo - Português (BR)',
-            'uploader_id': 'PocoyoBrazil',
-        },
-        'add_ie': ['Youtube'],
-        'params': {
-            # There are already too many Youtube downloads
-            'skip_download': True,
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-
-        real_video_id = self._html_search_meta(
-            'videoId', webpage, 'YouTube video id', fatal=True)
-
-        return self.url_result(real_video_id, YoutubeIE.ie_key())
-
-

 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com playlists'
     _VALID_URL = r"""(?x)(?:
 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com playlists'
     _VALID_URL = r"""(?x)(?:


@@ -2291,6 +2284,19 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
             page, 'title', default=None)
 
             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
             page, 'title', default=None)
 

+        _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
+        uploader = self._search_regex(
+            r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
+            page, 'uploader', default=None)
+        mobj = re.search(
+            r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
+            page)
+        if mobj:
+            uploader_id = mobj.group('uploader_id')
+            uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
+        else:
+            uploader_id = uploader_url = None
+

         has_videos = True
 
         if not playlist_title:
         has_videos = True
 
         if not playlist_title:


@@ -2301,8 +2307,15 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
             except StopIteration:
                 has_videos = False
 
             except StopIteration:
                 has_videos = False
 

-        return has_videos, self.playlist_result(
+        playlist = self.playlist_result(

             self._entries(page, playlist_id), playlist_id, playlist_title)
             self._entries(page, playlist_id), playlist_id, playlist_title)

+        playlist.update({
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'uploader_url': uploader_url,
+        })
+
+        return has_videos, playlist

 
     def _check_download_just_video(self, url, playlist_id):
         # Check if it's a video-specific URL
 
     def _check_download_just_video(self, url, playlist_id):
         # Check if it's a video-specific URL


@@ -2438,7 +2451,7 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
 
 class YoutubeUserIE(YoutubeChannelIE):
     IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
 
 class YoutubeUserIE(YoutubeChannelIE):
     IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'

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

     _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
     IE_NAME = 'youtube:user'
 
     _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
     IE_NAME = 'youtube:user'
 


@@ -2531,10 +2544,11 @@ class YoutubeLiveIE(YoutubeBaseInfoExtractor):
         webpage = self._download_webpage(url, channel_id, fatal=False)
         if webpage:
             page_type = self._og_search_property(
         webpage = self._download_webpage(url, channel_id, fatal=False)
         if webpage:
             page_type = self._og_search_property(

-                'type', webpage, 'page type', default=None)
+                'type', webpage, 'page type', default='')

             video_id = self._html_search_meta(
                 'videoId', webpage, 'video id', default=None)
             video_id = self._html_search_meta(
                 'videoId', webpage, 'video id', default=None)

-            if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
+            if page_type.startswith('video') and video_id and re.match(
+                    r'^[0-9A-Za-z_-]{11}$', video_id):

                 return self.url_result(video_id, YoutubeIE.ie_key())
         return self.url_result(base_url)
 
                 return self.url_result(video_id, YoutubeIE.ie_key())
         return self.url_result(base_url)
 


@@ -2569,7 +2583,11 @@ class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
     }]
 
 
     }]
 
 

-class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
+class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
+    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
+
+
+class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):

     IE_DESC = 'YouTube.com searches'
     # there doesn't appear to be a real limit, for example if you search for
     # 'python' you get more than 8.000.000 results
     IE_DESC = 'YouTube.com searches'
     # there doesn't appear to be a real limit, for example if you search for
     # 'python' you get more than 8.000.000 results


@@ -2603,8 +2621,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
                 raise ExtractorError(
                     '[youtube] No video results', expected=True)
 
                 raise ExtractorError(
                     '[youtube] No video results', expected=True)
 

-            new_videos = self._ids_to_results(orderedSet(re.findall(
-                r'href="/watch\?v=(.{11})', html_content)))
+            new_videos = list(self._process_page(html_content))

             videos += new_videos
             if not new_videos or len(videos) > limit:
                 break
             videos += new_videos
             if not new_videos or len(videos) > limit:
                 break


@@ -2627,11 +2644,10 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
 
 
     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
 
 

-class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
+class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):

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

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

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


@@ -2683,10 +2699,7 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()
 

-    def _real_extract(self, url):
-        page = self._download_webpage(
-            'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
-
+    def _entries(self, page):

         # The extraction process is the same as for playlists, but the regex
         # for the video ids doesn't contain an index
         ids = []
         # The extraction process is the same as for playlists, but the regex
         # for the video ids doesn't contain an index
         ids = []


@@ -2697,12 +2710,15 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
             # 'recommended' feed has infinite 'load more' and each new portion spins
             # the same videos in (sometimes) slightly different order, so we'll check
             # for unicity and break when portion has no new videos
             # 'recommended' feed has infinite 'load more' and each new portion spins
             # the same videos in (sometimes) slightly different order, so we'll check
             # for unicity and break when portion has no new videos

-            new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
+            new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))

             if not new_ids:
                 break
 
             ids.extend(new_ids)
 
             if not new_ids:
                 break
 
             ids.extend(new_ids)
 

+            for entry in self._ids_to_results(new_ids):
+                yield entry
+

             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
             if not mobj:
                 break
             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
             if not mobj:
                 break


@@ -2714,8 +2730,12 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
             content_html = more['content_html']
             more_widget_html = more['load_more_widget_html']
 
             content_html = more['content_html']
             more_widget_html = more['load_more_widget_html']
 

+    def _real_extract(self, url):
+        page = self._download_webpage(
+            'https://www.youtube.com/feed/%s' % self._FEED_NAME,
+            self._PLAYLIST_TITLE)

         return self.playlist_result(
         return self.playlist_result(

-            self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
+            self._entries(page), playlist_title=self._PLAYLIST_TITLE)

 
 
 class YoutubeWatchLaterIE(YoutubePlaylistIE):
 
 
 class YoutubeWatchLaterIE(YoutubePlaylistIE):