X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=67184bc5d81f605450809b165dfef8be0d950202;hb=ea0c2f219c219de8f59f1ae82e106cec5911c56c;hp=bc22421aec27265265b8c61b4266db4337bbf4d7;hpb=b68a812ea839e44148516a34a15193189e58ba77;p=youtube-dl

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index bc22421ae..67184bc5d 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -85,6 +85,7 @@ from .ustream import UstreamIE
 from .openload import OpenloadIE
 from .videopress import VideoPressIE
 from .rutube import RutubeIE
+from .limelight import LimelightBaseIE
 
 
 class GenericIE(InfoExtractor):
@@ -429,6 +430,22 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,  # m3u8 download
             },
         },
+        {
+            # Brightcove video in <iframe>
+            'url': 'http://www.un.org/chinese/News/story.asp?NewsID=27724',
+            'md5': '36d74ef5e37c8b4a2ce92880d208b968',
+            'info_dict': {
+                'id': '5360463607001',
+                'ext': 'mp4',
+                'title': '叙利亚失明儿童在废墟上演唱《心跳》  呼吁获得正常童年生活',
+                'description': '联合国儿童基金会中东和北非区域大使、作曲家扎德·迪拉尼（Zade Dirani）在3月15日叙利亚冲突爆发7周年纪念日之际发布了为叙利亚谱写的歌曲《心跳》（HEARTBEAT），为受到六年冲突影响的叙利亚儿童发出强烈呐喊，呼吁世界做出共同努力，使叙利亚儿童重新获得享有正常童年生活的权利。',
+                'uploader': 'United Nations',
+                'uploader_id': '1362235914001',
+                'timestamp': 1489593889,
+                'upload_date': '20170315',
+            },
+            'add_ie': ['BrightcoveLegacy'],
+        },
         {
             # Brightcove with alternative playerID key
             'url': 'http://www.nature.com/nmeth/journal/v9/n7/fig_tab/nmeth.2062_SV1.html',
@@ -449,6 +466,59 @@ class GenericIE(InfoExtractor):
                 },
             }],
         },
+        {
+            # Brightcove with UUID in videoPlayer
+            'url': 'http://www8.hp.com/cn/zh/home.html',
+            'info_dict': {
+                'id': '5255815316001',
+                'ext': 'mp4',
+                'title': 'Sprocket Video - China',
+                'description': 'Sprocket Video - China',
+                'uploader': 'HP-Video Gallery',
+                'timestamp': 1482263210,
+                'upload_date': '20161220',
+                'uploader_id': '1107601872001',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 download
+            },
+            'skip': 'video rotates...weekly?',
+        },
+        {
+            # Brightcove:new type [2].
+            'url': 'http://www.delawaresportszone.com/video-st-thomas-more-earns-first-trip-to-basketball-semis',
+            'md5': '2b35148fcf48da41c9fb4591650784f3',
+            'info_dict': {
+                'id': '5348741021001',
+                'ext': 'mp4',
+                'upload_date': '20170306',
+                'uploader_id': '4191638492001',
+                'timestamp': 1488769918,
+                'title': 'VIDEO:  St. Thomas More earns first trip to basketball semis',
+
+            },
+        },
+        {
+            # Alternative brightcove <video> attributes
+            'url': 'http://www.programme-tv.net/videos/extraits/81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche/',
+            'info_dict': {
+                'id': '81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche',
+                'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche, Extraits : toutes les vidéos avec Télé-Loisirs",
+            },
+            'playlist': [{
+                'md5': '732d22ba3d33f2f3fc253c39f8f36523',
+                'info_dict': {
+                    'id': '5311302538001',
+                    'ext': 'mp4',
+                    'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche",
+                    'description': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche (France 2, 5 février 2017)",
+                    'timestamp': 1486321708,
+                    'upload_date': '20170205',
+                    'uploader_id': '800000640001',
+                },
+                'only_matching': True,
+            }],
+        },
         {
             # Brightcove with UUID in videoPlayer
             'url': 'http://www8.hp.com/cn/zh/home.html',
@@ -730,6 +800,21 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             }
         },
+        # YouTube <object> embed
+        {
+            'url': 'http://www.improbable.com/2017/04/03/untrained-modern-youths-and-ancient-masters-in-selfie-portraits/',
+            'md5': '516718101ec834f74318df76259fb3cc',
+            'info_dict': {
+                'id': 'msN87y-iEx0',
+                'ext': 'webm',
+                'title': 'Feynman: Mirrors FUN TO IMAGINE 6',
+                'upload_date': '20080526',
+                'description': 'md5:0ffc78ea3f01b2e2c247d5f8d1d3c18d',
+                'uploader': 'Christopher Sykes',
+                'uploader_id': 'ChristopherJSykes',
+            },
+            'add_ie': ['Youtube'],
+        },
         # Camtasia studio
         {
             'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
@@ -902,12 +987,13 @@ class GenericIE(InfoExtractor):
         },
         # LazyYT
         {
-            'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
+            'url': 'https://skiplagged.com/',
             'info_dict': {
-                'id': '1986',
-                'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
+                'id': 'skiplagged',
+                'title': 'Skiplagged: The smart way to find cheap flights',
             },
-            'playlist_mincount': 2,
+            'playlist_mincount': 1,
+            'add_ie': ['Youtube'],
         },
         # Cinchcast embed
         {
@@ -990,6 +1076,20 @@ class GenericIE(InfoExtractor):
                 'thumbnail': r're:^https?://.*\.jpg$',
             },
         },
+        {
+            # JWPlayer config passed as variable
+            'url': 'http://www.txxx.com/videos/3326530/ariele/',
+            'info_dict': {
+                'id': '3326530_hq',
+                'ext': 'mp4',
+                'title': 'ARIELE | Tube Cup',
+                'uploader': 'www.txxx.com',
+                'age_limit': 18,
+            },
+            'params': {
+                'skip_download': True,
+            }
+        },
         # rtl.nl embed
         {
             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
@@ -1065,6 +1165,21 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
+        {
+            # Kaltura iframe embed
+            'url': 'http://www.gsd.harvard.edu/event/i-m-pei-a-centennial-celebration/',
+            'md5': 'ae5ace8eb09dc1a35d03b579a9c2cc44',
+            'info_dict': {
+                'id': '0_f2cfbpwy',
+                'ext': 'mp4',
+                'title': 'I. M. Pei: A Centennial Celebration',
+                'description': 'md5:1db8f40c69edc46ca180ba30c567f37c',
+                'upload_date': '20170403',
+                'uploader_id': 'batchUser',
+                'timestamp': 1491232186,
+            },
+            'add_ie': ['Kaltura'],
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -1542,6 +1657,26 @@ class GenericIE(InfoExtractor):
             'url': 'http://www.golfchannel.com/topics/shows/golftalkcentral.htm',
             'only_matching': True,
         },
+        {
+            # Senate ISVP iframe https
+            'url': 'https://www.hsgac.senate.gov/hearings/canadas-fast-track-refugee-plan-unanswered-questions-and-implications-for-us-national-security',
+            'md5': 'fb8c70b0b515e5037981a2492099aab8',
+            'info_dict': {
+                'id': 'govtaff020316',
+                'ext': 'mp4',
+                'title': 'Integrated Senate Video Player',
+            },
+            'add_ie': [SenateISVPIE.ie_key()],
+        },
+        {
+            # Limelight embeds (1 channel embed + 4 media embeds)
+            'url': 'http://www.sedona.com/FacilitatorTraining2017',
+            'info_dict': {
+                'id': 'FacilitatorTraining2017',
+                'title': 'Facilitator Training 2017',
+            },
+            'playlist_mincount': 5,
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -1584,7 +1719,7 @@ class GenericIE(InfoExtractor):
                 continue
 
             entries.append({
-                '_type': 'url',
+                '_type': 'url_transparent',
                 'url': next_url,
                 'title': it.find('title').text,
             })
@@ -1841,18 +1976,9 @@ class GenericIE(InfoExtractor):
         video_description = self._og_search_description(webpage, default=None)
         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
 
-        # Helper method
-        def _playlist_from_matches(matches, getter=None, ie=None):
-            urlrs = orderedSet(
-                self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
-                for m in matches)
-            return self.playlist_result(
-                urlrs, playlist_id=video_id, playlist_title=video_title)
-
         # Look for Brightcove Legacy Studio embeds
         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
         if bc_urls:
-            self.to_screen('Brightcove video detected.')
             entries = [{
                 '_type': 'url',
                 'url': smuggle_url(bc_url, {'Referer': url}),
@@ -1867,30 +1993,30 @@ class GenericIE(InfoExtractor):
             }
 
         # Look for Brightcove New Studio embeds
-        bc_urls = BrightcoveNewIE._extract_urls(webpage)
+        bc_urls = BrightcoveNewIE._extract_urls(self, webpage)
         if bc_urls:
-            return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
+            return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:
-            return _playlist_from_matches(tp_urls, ie='ThePlatform')
+            return self.playlist_from_matches(tp_urls, video_id, video_title, ie='ThePlatform')
 
         # Look for Vessel embeds
         vessel_urls = VesselIE._extract_urls(webpage)
         if vessel_urls:
-            return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
+            return self.playlist_from_matches(vessel_urls, video_id, video_title, ie=VesselIE.ie_key())
 
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
             webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='RtlNl')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='RtlNl')
 
         vimeo_urls = VimeoIE._extract_urls(url, webpage)
         if vimeo_urls:
-            return _playlist_from_matches(vimeo_urls, ie=VimeoIE.ie_key())
+            return self.playlist_from_matches(vimeo_urls, video_id, video_title, ie=VimeoIE.ie_key())
 
         vid_me_embed_url = self._search_regex(
             r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]',
@@ -1905,6 +2031,7 @@ class GenericIE(InfoExtractor):
                 data-video-url=|
                 <embed[^>]+?src=|
                 embedSWF\(?:\s*|
+                <object[^>]+data=|
                 new\s+SWFObject\(
             )
             (["\'])
@@ -1912,25 +2039,25 @@ class GenericIE(InfoExtractor):
                 (?:embed|v|p)/.+?)
             \1''', webpage)
         if matches:
-            return _playlist_from_matches(
-                matches, lambda m: unescapeHTML(m[1]))
+            return self.playlist_from_matches(
+                matches, video_id, video_title, lambda m: unescapeHTML(m[1]))
 
         # Look for lazyYT YouTube embed
         matches = re.findall(
             r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
 
         # Look for Wordpress "YouTube Video Importer" plugin
         matches = re.findall(r'''(?x)<div[^>]+
             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
         if matches:
-            return _playlist_from_matches(matches, lambda m: m[-1])
+            return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
 
         matches = DailymotionIE._extract_urls(webpage)
         if matches:
-            return _playlist_from_matches(matches)
+            return self.playlist_from_matches(matches, video_id, video_title)
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1939,8 +2066,8 @@ class GenericIE(InfoExtractor):
             playlists = re.findall(
                 r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
             if playlists:
-                return _playlist_from_matches(
-                    playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
+                return self.playlist_from_matches(
+                    playlists, video_id, video_title, lambda p: '//dailymotion.com/playlist/%s' % p)
 
         # Look for embedded Wistia player
         match = re.search(
@@ -2047,8 +2174,9 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             embeds = self._parse_json(mobj.group(1), video_id, fatal=False)
             if embeds:
-                return _playlist_from_matches(
-                    embeds, getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
+                return self.playlist_from_matches(
+                    embeds, video_id, video_title,
+                    getter=lambda v: OoyalaIE._url_for_embed_code(smuggle_url(v['provider_video_id'], {'domain': url})), ie='Ooyala')
 
         # Look for Aparat videos
         mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
@@ -2110,13 +2238,13 @@ class GenericIE(InfoExtractor):
         # Look for funnyordie embed
         matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
         if matches:
-            return _playlist_from_matches(
-                matches, getter=unescapeHTML, ie='FunnyOrDie')
+            return self.playlist_from_matches(
+                matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie')
 
         # Look for BBC iPlayer embed
         matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage)
         if matches:
-            return _playlist_from_matches(matches, ie='BBCCoUk')
+            return self.playlist_from_matches(matches, video_id, video_title, ie='BBCCoUk')
 
         # Look for embedded RUTV player
         rutv_url = RUTVIE._extract_url(webpage)
@@ -2131,32 +2259,32 @@ class GenericIE(InfoExtractor):
         # Look for embedded SportBox player
         sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
         if sportbox_urls:
-            return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
+            return self.playlist_from_matches(sportbox_urls, video_id, video_title, ie='SportBoxEmbed')
 
         # Look for embedded XHamster player
         xhamster_urls = XHamsterEmbedIE._extract_urls(webpage)
         if xhamster_urls:
-            return _playlist_from_matches(xhamster_urls, ie='XHamsterEmbed')
+            return self.playlist_from_matches(xhamster_urls, video_id, video_title, ie='XHamsterEmbed')
 
         # Look for embedded TNAFlixNetwork player
         tnaflix_urls = TNAFlixNetworkEmbedIE._extract_urls(webpage)
         if tnaflix_urls:
-            return _playlist_from_matches(tnaflix_urls, ie=TNAFlixNetworkEmbedIE.ie_key())
+            return self.playlist_from_matches(tnaflix_urls, video_id, video_title, ie=TNAFlixNetworkEmbedIE.ie_key())
 
         # Look for embedded PornHub player
         pornhub_urls = PornHubIE._extract_urls(webpage)
         if pornhub_urls:
-            return _playlist_from_matches(pornhub_urls, ie=PornHubIE.ie_key())
+            return self.playlist_from_matches(pornhub_urls, video_id, video_title, ie=PornHubIE.ie_key())
 
         # Look for embedded DrTuber player
         drtuber_urls = DrTuberIE._extract_urls(webpage)
         if drtuber_urls:
-            return _playlist_from_matches(drtuber_urls, ie=DrTuberIE.ie_key())
+            return self.playlist_from_matches(drtuber_urls, video_id, video_title, ie=DrTuberIE.ie_key())
 
         # Look for embedded RedTube player
         redtube_urls = RedTubeIE._extract_urls(webpage)
         if redtube_urls:
-            return _playlist_from_matches(redtube_urls, ie=RedTubeIE.ie_key())
+            return self.playlist_from_matches(redtube_urls, video_id, video_title, ie=RedTubeIE.ie_key())
 
         # Look for embedded Tvigle player
         mobj = re.search(
@@ -2202,12 +2330,12 @@ class GenericIE(InfoExtractor):
         # Look for embedded soundcloud player
         soundcloud_urls = SoundcloudIE._extract_urls(webpage)
         if soundcloud_urls:
-            return _playlist_from_matches(soundcloud_urls, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
+            return self.playlist_from_matches(soundcloud_urls, video_id, video_title, getter=unescapeHTML, ie=SoundcloudIE.ie_key())
 
         # Look for tunein player
         tunein_urls = TuneInBaseIE._extract_urls(webpage)
         if tunein_urls:
-            return _playlist_from_matches(tunein_urls)
+            return self.playlist_from_matches(tunein_urls, video_id, video_title)
 
         # Look for embedded mtvservices player
         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
@@ -2381,6 +2509,11 @@ class GenericIE(InfoExtractor):
             return self.url_result(piksel_url, PikselIE.ie_key())
 
         # Look for Limelight embeds
+        limelight_urls = LimelightBaseIE._extract_urls(webpage, url)
+        if limelight_urls:
+            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 = {
@@ -2490,35 +2623,35 @@ class GenericIE(InfoExtractor):
         # Look for DBTV embeds
         dbtv_urls = DBTVIE._extract_urls(webpage)
         if dbtv_urls:
-            return _playlist_from_matches(dbtv_urls, ie=DBTVIE.ie_key())
+            return self.playlist_from_matches(dbtv_urls, video_id, video_title, ie=DBTVIE.ie_key())
 
         # Look for Videa embeds
         videa_urls = VideaIE._extract_urls(webpage)
         if videa_urls:
-            return _playlist_from_matches(videa_urls, ie=VideaIE.ie_key())
+            return self.playlist_from_matches(videa_urls, video_id, video_title, ie=VideaIE.ie_key())
 
         # Look for 20 minuten embeds
         twentymin_urls = TwentyMinutenIE._extract_urls(webpage)
         if twentymin_urls:
-            return _playlist_from_matches(
-                twentymin_urls, ie=TwentyMinutenIE.ie_key())
+            return self.playlist_from_matches(
+                twentymin_urls, video_id, video_title, ie=TwentyMinutenIE.ie_key())
 
         # Look for Openload embeds
         openload_urls = OpenloadIE._extract_urls(webpage)
         if openload_urls:
-            return _playlist_from_matches(
-                openload_urls, ie=OpenloadIE.ie_key())
+            return self.playlist_from_matches(
+                openload_urls, video_id, video_title, ie=OpenloadIE.ie_key())
 
         # Look for VideoPress embeds
         videopress_urls = VideoPressIE._extract_urls(webpage)
         if videopress_urls:
-            return _playlist_from_matches(
-                videopress_urls, ie=VideoPressIE.ie_key())
+            return self.playlist_from_matches(
+                videopress_urls, video_id, video_title, ie=VideoPressIE.ie_key())
 
         # Look for Rutube embeds
         rutube_urls = RutubeIE._extract_urls(webpage)
         if rutube_urls:
-            return _playlist_from_matches(
+            return self.playlist_from_matches(
                 rutube_urls, ie=RutubeIE.ie_key())
 
         # Looking for http://schema.org/VideoObject
@@ -2545,14 +2678,14 @@ class GenericIE(InfoExtractor):
                 self._sort_formats(entry['formats'])
             return self.playlist_result(entries)
 
-        jwplayer_data_str = self._find_jwplayer_data(webpage)
-        if jwplayer_data_str:
-            try:
-                jwplayer_data = self._parse_json(
-                    jwplayer_data_str, video_id, transform_source=js_to_json)
-                return self._parse_jwplayer_data(jwplayer_data, video_id)
-            except ExtractorError:
-                pass
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
+        if jwplayer_data:
+            info = self._parse_jwplayer_data(
+                jwplayer_data, video_id, require_title=False, base_url=url)
+            if not info.get('title'):
+                info['title'] = video_title
+            return info
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
@@ -2561,7 +2694,7 @@ class GenericIE(InfoExtractor):
                 return True
             vpath = compat_urlparse.urlparse(vurl).path
             vext = determine_ext(vpath)
-            return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js')
+            return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml', 'js', 'xml')
 
         def filter_video(urls):
             return list(filter(check_video, urls))
@@ -2627,11 +2760,14 @@ class GenericIE(InfoExtractor):
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
-                self.report_following_redirect(new_url)
-                return {
-                    '_type': 'url',
-                    'url': new_url,
-                }
+                if new_url != url:
+                    self.report_following_redirect(new_url)
+                    return {
+                        '_type': 'url',
+                        'url': new_url,
+                    }
+                else:
+                    found = None
 
         if not found:
             # twitter:player is a https URL to iframe player that may or may not