[youtube] Also use the 'adaptative_fmts' field from the /get_video_info page (fixes...

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

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index a7c5145134084cd81776ca01d2ee58a7119f0e05..8fb07d100ea221d3a9da058270466909b51461c7 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -236,11 +236,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         '136': 'mp4',
         '137': 'mp4',
         '138': 'mp4',
-        '139': 'mp4',
-        '140': 'mp4',
-        '141': 'mp4',
         '160': 'mp4',
 
+        # Dash mp4 audio
+        '139': 'm4a',
+        '140': 'm4a',
+        '141': 'm4a',
+
         # Dash webm
         '171': 'webm',
         '172': 'webm',
@@ -1150,7 +1152,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             list_page = self._download_webpage(list_url, video_id)
             caption_list = xml.etree.ElementTree.fromstring(list_page.encode('utf-8'))
             original_lang_node = caption_list.find('track')
-            if original_lang_node.attrib.get('kind') != 'asr' :
+            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')
                 return {}
             original_lang = original_lang_node.attrib['lang_code']
@@ -1250,6 +1252,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             url_map[itag] = format_url
         return url_map
 
+    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.')
+
     def _real_extract(self, url):
         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
         mobj = re.search(self._NEXT_URL_RE, url)
@@ -1382,6 +1388,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         else:
             video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
 
+        # annotations
+        video_annotations = None
+        if self._downloader.params.get('writeannotations', False):
+                video_annotations = self._extract_annotations(video_id)
+
         # Decide which formats to download
 
         try:
@@ -1394,32 +1405,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             # this signatures are encrypted
             if 'url_encoded_fmt_stream_map' not in args:
                 raise ValueError(u'No stream_map present')  # caught below
-            m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
+            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.search(r'[&,]s=', args.get('adaptive_fmts', u''))
+            m_s = re_signature.search(args.get('adaptive_fmts', u''))
             if m_s is not None:
-                if 'url_encoded_fmt_stream_map' in video_info:
-                    video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
-                else:
-                    video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
-            elif 'adaptive_fmts' in video_info:
-                if 'url_encoded_fmt_stream_map' in video_info:
-                    video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
+                if 'adaptive_fmts' in video_info:
+                    video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
                 else:
-                    video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
+                    video_info['adaptive_fmts'] = [args['adaptive_fmts']]
         except ValueError:
             pass
 
         if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
             self.report_rtmp_download()
             video_url_list = [(None, video_info['conn'][0])]
-        elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
-            if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
+        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]
+            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 = {}
-            for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
+            for url_data_str in encoded_url_map.split(','):
                 url_data = compat_parse_qs(url_data_str)
                 if 'itag' in url_data and 'url' in url_data:
                     url = url_data['url'][0]
@@ -1495,6 +1503,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'subtitles':    video_subtitles,
                 'duration':     video_duration,
                 'age_limit':    18 if age_gate else 0,
+                'annotations':  video_annotations
             })
         return results
 
@@ -1635,7 +1644,7 @@ class YoutubeChannelIE(InfoExtractor):
 
 class YoutubeUserIE(InfoExtractor):
     IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!watch(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
+    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
     _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
     _GDATA_PAGE_SIZE = 50
     _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'