debian/patches/remove-autoupdate-mechanism.patch: Refresh.

[youtubedl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 43051512bc1013640bc99d98437b3cf2a7b258a3..617be8e96b6e305988b3ae4d0c5d3790074868e4 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -1944,6 +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]


@@ -2446,7 +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'
 


@@ -2578,7 +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


@@ -2612,8 +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


@@ -2636,11 +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,