[facebook] Add support for plugin video embeds and multiple embeds (closes #13493)

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 8f9097945be51147a91a4b8bb6c679c14d86164a..760a7f9c24d0c7de8f1c7e6e1a895effec67bb17 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -10,6 +10,7 @@ from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..compat import (
     compat_etree_fromstring,
+    compat_str,
     compat_urllib_parse_unquote,
     compat_urlparse,
     compat_xml_parse_error,
@@ -89,6 +90,7 @@ from .limelight import LimelightBaseIE
 from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
 from .wistia import WistiaIE
+from .mediaset import MediasetIE
 
 
 class GenericIE(InfoExtractor):
@@ -1719,6 +1721,19 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': [WashingtonPostIE.ie_key()],
         },
+        {
+            # Mediaset embed
+            'url': 'http://www.tgcom24.mediaset.it/politica/serracchiani-voglio-vivere-in-una-societa-aperta-reazioni-sproporzionate-_3071354-201702a.shtml',
+            'info_dict': {
+                'id': '720642',
+                'ext': 'mp4',
+                'title': 'Serracchiani: "Voglio vivere in una società aperta, con tutela del patto di fiducia"',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'add_ie': [MediasetIE.ie_key()],
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -1893,14 +1908,14 @@ class GenericIE(InfoExtractor):
         content_type = head_response.headers.get('Content-Type', '').lower()
         m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>[^;\s]+)', content_type)
         if m:
-            format_id = m.group('format_id')
+            format_id = compat_str(m.group('format_id'))
             if format_id.endswith('mpegurl'):
                 formats = self._extract_m3u8_formats(url, video_id, 'mp4')
             elif format_id == 'f4m':
                 formats = self._extract_f4m_formats(url, video_id)
             else:
                 formats = [{
-                    'format_id': m.group('format_id'),
+                    'format_id': format_id,
                     'url': url,
                     'vcodec': 'none' if m.group('type') == 'audio' else None
                 }]
@@ -2207,9 +2222,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
 
         # Look for embedded Facebook player
-        facebook_url = FacebookIE._extract_url(webpage)
-        if facebook_url is not None:
-            return self.url_result(facebook_url, 'Facebook')
+        facebook_urls = FacebookIE._extract_urls(webpage)
+        if facebook_urls:
+            return self.playlist_from_matches(facebook_urls, video_id, video_title)
 
         # Look for embedded VK player
         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
@@ -2519,29 +2534,6 @@ class GenericIE(InfoExtractor):
             return self.playlist_result(
                 limelight_urls, video_id, video_title, video_description)
 
-        mobj = re.search(r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})', webpage)
-        if mobj:
-            lm = {
-                'Media': 'media',
-                'Channel': 'channel',
-                'ChannelList': 'channel_list',
-            }
-            return self.url_result(smuggle_url('limelight:%s:%s' % (
-                lm[mobj.group(1)], mobj.group(2)), {'source_url': url}),
-                'Limelight%s' % mobj.group(1), mobj.group(2))
-
-        mobj = re.search(
-            r'''(?sx)
-                <object[^>]+class=(["\'])LimelightEmbeddedPlayerFlash\1[^>]*>.*?
-                    <param[^>]+
-                        name=(["\'])flashVars\2[^>]+
-                        value=(["\'])(?:(?!\3).)*mediaId=(?P<id>[a-z0-9]{32})
-            ''', webpage)
-        if mobj:
-            return self.url_result(smuggle_url(
-                'limelight:media:%s' % mobj.group('id'),
-                {'source_url': url}), 'LimelightMedia', mobj.group('id'))
-
         # Look for Anvato embeds
         anvato_urls = AnvatoIE._extract_urls(self, webpage, video_id)
         if anvato_urls:
@@ -2671,6 +2663,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 wapo_urls, video_id, video_title, ie=WashingtonPostIE.ie_key())
 
+        # Look for Mediaset embeds
+        mediaset_urls = MediasetIE._extract_urls(webpage)
+        if mediaset_urls:
+            return self.playlist_from_matches(
+                mediaset_urls, video_id, video_title, ie=MediasetIE.ie_key())
+
         # Looking for http://schema.org/VideoObject
         json_ld = self._search_json_ld(
             webpage, video_id, default={}, expected_type='VideoObject')