[extractor/common] Skip html comment tags (Closes #6822)

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index b1af4587025e56c33b38ca294a8fd50beeea9ef5..d694e818e98dc29939118a6d9fccb8b942b03128 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -510,6 +510,12 @@ class InfoExtractor(object):
         """Report attempt to log in."""
         self.to_screen('Logging in')
 
+    @staticmethod
+    def raise_login_required(msg='This video is only available for registered users'):
+        raise ExtractorError(
+            '%s. Use --username and --password or --netrc to provide account credentials.' % msg,
+            expected=True)
+
     # Methods for following #608
     @staticmethod
     def url_result(url, ie=None, video_id=None, video_title=None):
@@ -725,9 +731,10 @@ class InfoExtractor(object):
 
     @staticmethod
     def _hidden_inputs(html):
+        html = re.sub(r'<!--(?:(?!<!--).)*-->', '', html)
         hidden_inputs = {}
-        for input in re.findall(r'<input([^>]+)>', html):
-            if not re.search(r'type=(["\'])hidden\1', input):
+        for input in re.findall(r'(?i)<input([^>]+)>', html):
+            if not re.search(r'type=(["\'])(?:hidden|submit)\1', input):
                 continue
             name = re.search(r'name=(["\'])(?P<value>.+?)\1', input)
             if not name:
@@ -740,7 +747,7 @@ class InfoExtractor(object):
 
     def _form_hidden_inputs(self, form_id, html):
         form = self._search_regex(
-            r'(?s)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id,
+            r'(?is)<form[^>]+?id=(["\'])%s\1[^>]*>(?P<form>.+?)</form>' % form_id,
             html, '%s form' % form_id, group='form')
         return self._hidden_inputs(form)
 
@@ -1144,9 +1151,14 @@ class InfoExtractor(object):
             ext = textstream.get('ext') or determine_ext(src)
             if not ext:
                 type_ = textstream.get('type')
-                if type_ == 'text/srt':
-                    ext = 'srt'
-            lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or subtitles_lang
+                SUBTITLES_TYPES = {
+                    'text/vtt': 'vtt',
+                    'text/srt': 'srt',
+                    'application/smptett+xml': 'tt',
+                }
+                if type_ in SUBTITLES_TYPES:
+                    ext = SUBTITLES_TYPES[type_]
+            lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') or textstream.get('lang') or subtitles_lang
             subtitles.setdefault(lang, []).append({
                 'url': src,
                 'ext': ext,
@@ -1274,6 +1286,23 @@ class InfoExtractor(object):
     def _get_subtitles(self, *args, **kwargs):
         raise NotImplementedError("This method must be implemented by subclasses")
 
+    @staticmethod
+    def _merge_subtitle_items(subtitle_list1, subtitle_list2):
+        """ Merge subtitle items for one language. Items with duplicated URLs
+        will be dropped. """
+        list1_urls = set([item['url'] for item in subtitle_list1])
+        ret = list(subtitle_list1)
+        ret.extend([item for item in subtitle_list2 if item['url'] not in list1_urls])
+        return ret
+
+    @classmethod
+    def _merge_subtitles(cls, subtitle_dict1, subtitle_dict2):
+        """ Merge two subtitle dictionaries, language by language. """
+        ret = dict(subtitle_dict1)
+        for lang in subtitle_dict2:
+            ret[lang] = cls._merge_subtitle_items(subtitle_dict1.get(lang, []), subtitle_dict2[lang])
+        return ret
+
     def extract_automatic_captions(self, *args, **kwargs):
         if (self._downloader.params.get('writeautomaticsub', False) or
                 self._downloader.params.get('listsubtitles')):