Merge branch 'ted_subtitles'

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

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index d7c9b38f9da7054a43827a5e44572986717714a7..9053f3ead8ee81b490fdc9982359465e29d965db 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -74,14 +74,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
             return False
 
-        galx = None
-        dsh = None
-        match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
-        if match:
-          galx = match.group(1)
-        match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
-        if match:
-          dsh = match.group(1)
+        galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
+                                  login_page, u'Login GALX parameter')
 
         # Log in
         login_form_strs = {
@@ -95,7 +89,6 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 u'checkConnection': u'',
                 u'checkedDomains': u'youtube',
                 u'dnConn': u'',
-                u'dsh': dsh,
                 u'pstMsg': u'0',
                 u'rmShown': u'1',
                 u'secTok': u'',
@@ -236,11 +229,13 @@ 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',
@@ -346,7 +341,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
         },
         {
             u"url":  u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
-            u"file":  u"1ltcDfZMA3U.flv",
+            u"file":  u"1ltcDfZMA3U.mp4",
             u"note": u"Test VEVO video (#897)",
             u"info_dict": {
                 u"upload_date": u"20070518",
@@ -1116,7 +1111,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'lang': lang,
                 'v': video_id,
                 'fmt': self._downloader.params.get('subtitlesformat'),
-                'name': l[0],
+                'name': l[0].encode('utf-8'),
             })
             url = u'http://www.youtube.com/api/timedtext?' + params
             sub_lang_list[lang] = url
@@ -1150,7 +1145,7 @@ 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']
@@ -1403,32 +1398,29 @@ 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]
@@ -1481,13 +1473,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
 
         results = []
-        for format_param, video_real_url in video_url_list:
+        for itag, video_real_url in video_url_list:
             # Extension
-            video_extension = self._video_extensions.get(format_param, 'flv')
+            video_extension = self._video_extensions.get(itag, 'flv')
 
-            video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
-                                              self._video_dimensions.get(format_param, '???'),
-                                              ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
+            video_format = '{0} - {1}{2}'.format(itag if itag else video_extension,
+                                              self._video_dimensions.get(itag, '???'),
+                                              ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '')
 
             results.append({
                 'id':       video_id,
@@ -1498,6 +1490,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 'title':    video_title,
                 'ext':      video_extension,
                 'format':   video_format,
+                'format_id': itag,
                 'thumbnail':    video_thumbnail,
                 'description':  video_description,
                 'player_url':   player_url,