[malltv] Add extractor (closes #18058)

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

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 8452125c8802b1398698c334084c85b673e80964..c3b0586a0da65ce42ce5f3ca6aee2e7119d3d98f 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -606,6 +606,11 @@ class InfoExtractor(object):
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             if isinstance(err, compat_urllib_error.HTTPError):
                 if self.__can_accept_status_code(err, expected_status):
+                    # Retain reference to error to prevent file object from
+                    # being closed before it can be read. Works around the
+                    # effects of <https://bugs.python.org/issue15002>
+                    # introduced in Python 3.4.1.
+                    err.fp._error = err
                     return err.fp
 
             if errnote is False:
@@ -1053,7 +1058,7 @@ class InfoExtractor(object):
     @staticmethod
     def _og_regexes(prop):
         content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))'
-        property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)'
+        property_re = (r'(?:name|property)=(?:\'og[:-]%(prop)s\'|"og[:-]%(prop)s"|\s*og[:-]%(prop)s\b)'
                        % {'prop': re.escape(prop)})
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
@@ -1234,17 +1239,30 @@ class InfoExtractor(object):
                 if expected_type is not None and expected_type != item_type:
                     return info
                 if item_type in ('TVEpisode', 'Episode'):
+                    episode_name = unescapeHTML(e.get('name'))
                     info.update({
-                        'episode': unescapeHTML(e.get('name')),
+                        'episode': episode_name,
                         'episode_number': int_or_none(e.get('episodeNumber')),
                         'description': unescapeHTML(e.get('description')),
                     })
+                    if not info.get('title') and episode_name:
+                        info['title'] = episode_name
                     part_of_season = e.get('partOfSeason')
                     if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'):
-                        info['season_number'] = int_or_none(part_of_season.get('seasonNumber'))
+                        info.update({
+                            'season': unescapeHTML(part_of_season.get('name')),
+                            'season_number': int_or_none(part_of_season.get('seasonNumber')),
+                        })
                     part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
                     if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'):
                         info['series'] = unescapeHTML(part_of_series.get('name'))
+                elif item_type == 'Movie':
+                    info.update({
+                        'title': unescapeHTML(e.get('name')),
+                        'description': unescapeHTML(e.get('description')),
+                        'duration': parse_duration(e.get('duration')),
+                        'timestamp': unified_timestamp(e.get('dateCreated')),
+                    })
                 elif item_type in ('Article', 'NewsArticle'):
                     info.update({
                         'timestamp': parse_iso8601(e.get('datePublished')),
@@ -1581,6 +1599,7 @@ class InfoExtractor(object):
         # References:
         # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-21
         # 2. https://github.com/rg3/youtube-dl/issues/12211
+        # 3. https://github.com/rg3/youtube-dl/issues/18923
 
         # We should try extracting formats only from master playlists [1, 4.3.4],
         # i.e. playlists that describe available qualities. On the other hand
@@ -1652,11 +1671,16 @@ class InfoExtractor(object):
             rendition = stream_group[0]
             return rendition.get('NAME') or stream_group_id
 
+        # parse EXT-X-MEDIA tags before EXT-X-STREAM-INF in order to have the
+        # chance to detect video only formats when EXT-X-STREAM-INF tags
+        # precede EXT-X-MEDIA tags in HLS manifest such as [3].
+        for line in m3u8_doc.splitlines():
+            if line.startswith('#EXT-X-MEDIA:'):
+                extract_media(line)
+
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
                 last_stream_inf = parse_m3u8_attributes(line)
-            elif line.startswith('#EXT-X-MEDIA:'):
-                extract_media(line)
             elif line.startswith('#') or not line.strip():
                 continue
             else:
@@ -2609,7 +2633,7 @@ class InfoExtractor(object):
                 'id': this_video_id,
                 'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')),
                 'description': video_data.get('description'),
-                'thumbnail': self._proto_relative_url(video_data.get('image')),
+                'thumbnail': urljoin(base_url, self._proto_relative_url(video_data.get('image'))),
                 'timestamp': int_or_none(video_data.get('pubdate')),
                 'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')),
                 'subtitles': subtitles,
@@ -2636,12 +2660,9 @@ class InfoExtractor(object):
         for source in jwplayer_sources_data:
             if not isinstance(source, dict):
                 continue
-            source_url = self._proto_relative_url(source.get('file'))
-            if not source_url:
-                continue
-            if base_url:
-                source_url = compat_urlparse.urljoin(base_url, source_url)
-            if source_url in urls:
+            source_url = urljoin(
+                base_url, self._proto_relative_url(source.get('file')))
+            if not source_url or source_url in urls:
                 continue
             urls.append(source_url)
             source_type = source.get('type') or ''