[extractor/common] Add support for float durations in _parse_mpd_formats (closes...

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 459e7ffd6ced1685e802a642b5bc63b0c61e666c..e747258aa22378ddd25acbe3aeb5b096c32b2df3 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -940,7 +940,8 @@ class InfoExtractor(object):
 
     def _family_friendly_search(self, html):
         # See http://schema.org/VideoObject
-        family_friendly = self._html_search_meta('isFamilyFriendly', html)
+        family_friendly = self._html_search_meta(
+            'isFamilyFriendly', html, default=None)
 
         if not family_friendly:
             return None
@@ -1785,7 +1786,7 @@ class InfoExtractor(object):
                     ms_info['timescale'] = int(timescale)
                 segment_duration = source.get('duration')
                 if segment_duration:
-                    ms_info['segment_duration'] = int(segment_duration)
+                    ms_info['segment_duration'] = float(segment_duration)
 
             def extract_Initialization(source):
                 initialization = source.find(_add_ns('Initialization'))
@@ -2114,9 +2115,9 @@ class InfoExtractor(object):
                 return f
             return {}
 
-        def _media_formats(src, cur_media_type):
+        def _media_formats(src, cur_media_type, type_info={}):
             full_url = absolute_url(src)
-            ext = determine_ext(full_url)
+            ext = type_info.get('ext') or determine_ext(full_url)
             if ext == 'm3u8':
                 is_plain_url = False
                 formats = self._extract_m3u8_formats(
@@ -2165,9 +2166,9 @@ class InfoExtractor(object):
                     src = source_attributes.get('src')
                     if not src:
                         continue
-                    is_plain_url, formats = _media_formats(src, media_type)
+                    f = parse_content_type(source_attributes.get('type'))
+                    is_plain_url, formats = _media_formats(src, media_type, f)
                     if is_plain_url:
-                        f = parse_content_type(source_attributes.get('type'))
                         f.update(formats[0])
                         media_info['formats'].append(f)
                     else: