X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=25d056b3c21ea3f98c5c36384e86d243dfd4a913;hb=494d664e679c5b0f85e3c899579e7eb8a1cc8246;hp=0a0d2f41a45f80e0831a33fa7b01da16e5bd8df6;hpb=64b6a4e91ef735c8d07e93c4e670a01bcb10e7bb;p=youtube-dl

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 0a0d2f41a..25d056b3c 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -31,6 +31,7 @@ from ..utils import (
     clean_html,
     dict_get,
     error_to_compat_str,
+    extract_attributes,
     ExtractorError,
     float_or_none,
     get_element_by_attribute,
@@ -324,17 +325,18 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
         for video_id, video_title in self.extract_videos_from_page(content):
             yield self.url_result(video_id, 'Youtube', video_id, video_title)
 
-    def extract_videos_from_page(self, page):
-        ids_in_page = []
-        titles_in_page = []
-        for mobj in re.finditer(self._VIDEO_RE, page):
+    def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page):
+        for mobj in re.finditer(video_re, page):
             # The link with index 0 is not the first video of the playlist (not sure if still actual)
             if 'index' in mobj.groupdict() and mobj.group('id') == '0':
                 continue
             video_id = mobj.group('id')
-            video_title = unescapeHTML(mobj.group('title'))
+            video_title = unescapeHTML(
+                mobj.group('title')) if 'title' in mobj.groupdict() else None
             if video_title:
                 video_title = video_title.strip()
+            if video_title == '► Play all':
+                video_title = None
             try:
                 idx = ids_in_page.index(video_id)
                 if video_title and not titles_in_page[idx]:
@@ -342,6 +344,12 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
             except ValueError:
                 ids_in_page.append(video_id)
                 titles_in_page.append(video_title)
+
+    def extract_videos_from_page(self, page):
+        ids_in_page = []
+        titles_in_page = []
+        self.extract_videos_from_page_impl(
+            self._VIDEO_RE, page, ids_in_page, titles_in_page)
         return zip(ids_in_page, titles_in_page)
 
 
@@ -379,8 +387,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                             (?:www\.)?invidious\.enkirton\.net/|
                             (?:www\.)?invidious\.13ad\.de/|
                             (?:www\.)?invidious\.mastodon\.host/|
+                            (?:www\.)?invidious\.nixnet\.xyz/|
                             (?:www\.)?tube\.poal\.co/|
                             (?:www\.)?vid\.wxzm\.sx/|
+                            (?:www\.)?yt\.elukerio\.org/|
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
                          (?:                                                  # the various things that can precede the ID:
@@ -1801,10 +1811,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         break
 
         def extract_unavailable_message():
-            return self._html_search_regex(
-                (r'(?s)<div[^>]+id=["\']unavailable-submessage["\'][^>]+>(.+?)</div',
-                 r'(?s)<h1[^>]+id=["\']unavailable-message["\'][^>]*>(.+?)</h1>'),
-                video_webpage, 'unavailable message', default=None)
+            messages = []
+            for tag, kind in (('h1', 'message'), ('div', 'submessage')):
+                msg = self._html_search_regex(
+                    r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
+                    video_webpage, 'unavailable %s' % kind, default=None)
+                if msg:
+                    messages.append(msg)
+            if messages:
+                return '\n'.join(messages)
 
         if not video_info:
             unavailable_message = extract_unavailable_message()
@@ -2438,7 +2453,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
                         (%(playlist_id)s)
                      )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
-    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
+    _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
+    _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
     IE_NAME = 'youtube:playlist'
     _TESTS = [{
         'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
@@ -2603,6 +2619,34 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
     def _real_initialize(self):
         self._login()
 
+    def extract_videos_from_page(self, page):
+        ids_in_page = []
+        titles_in_page = []
+
+        for item in re.findall(
+                r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
+            attrs = extract_attributes(item)
+            video_id = attrs['data-video-id']
+            video_title = unescapeHTML(attrs.get('data-title'))
+            if video_title:
+                video_title = video_title.strip()
+            ids_in_page.append(video_id)
+            titles_in_page.append(video_title)
+
+        # Fallback with old _VIDEO_RE
+        self.extract_videos_from_page_impl(
+            self._VIDEO_RE, page, ids_in_page, titles_in_page)
+
+        # Relaxed fallbacks
+        self.extract_videos_from_page_impl(
+            r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
+            ids_in_page, titles_in_page)
+        self.extract_videos_from_page_impl(
+            r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
+            ids_in_page, titles_in_page)
+
+        return zip(ids_in_page, titles_in_page)
+
     def _extract_mix(self, playlist_id):
         # The mixes are generated from a single video
         # the id of the playlist is just 'RD' + video_id