X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgeneric.py;h=123a21296d5d48cd475bf15388f33715610d6972;hb=cea931a9e5cf682eafe5b7cdacbbe22630b7a9e7;hp=6a34c2491ebcbef18b3c7017a0567c433acaa842;hpb=72950c4dce413d4274323173dc3e7a2c17d93392;p=youtube-dl

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 6a34c2491..123a21296 100644
--- 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,
@@ -35,6 +36,7 @@ from .brightcove import (
     BrightcoveLegacyIE,
     BrightcoveNewIE,
 )
+from .nexx import NexxIE
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
@@ -56,6 +58,7 @@ from .dailymotion import (
     DailymotionIE,
     DailymotionCloudIE,
 )
+from .dailymail import DailyMailIE
 from .onionstudios import OnionStudiosIE
 from .viewlift import ViewLiftEmbedIE
 from .mtv import MTVServicesEmbeddedIE
@@ -85,6 +88,12 @@ from .ustream import UstreamIE
 from .openload import OpenloadIE
 from .videopress import VideoPressIE
 from .rutube import RutubeIE
+from .limelight import LimelightBaseIE
+from .anvato import AnvatoIE
+from .washingtonpost import WashingtonPostIE
+from .wistia import WistiaIE
+from .mediaset import MediasetIE
+from .joj import JojIE
 
 
 class GenericIE(InfoExtractor):
@@ -429,6 +438,22 @@ 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',
@@ -737,6 +762,20 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Dailymotion'],
         },
+        # DailyMail embed
+        {
+            'url': 'http://www.bumm.sk/krimi/2017/07/05/biztonsagi-kamera-buktatta-le-az-agg-ferfit-utlegelo-apolot',
+            'info_dict': {
+                'id': '1495629',
+                'ext': 'mp4',
+                'title': 'Care worker punches elderly dementia patient in head 11 times',
+                'description': 'md5:3a743dee84e57e48ec68bf67113199a5',
+            },
+            'add_ie': ['DailyMail'],
+            'params': {
+                'skip_download': True,
+            },
+        },
         # YouTube embed
         {
             'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
@@ -1163,7 +1202,7 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
-        # Eagle.Platform embed (generic URL)
+        # EaglePlatform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1177,8 +1216,26 @@ class GenericIE(InfoExtractor):
                 'view_count': int,
                 'age_limit': 0,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
-        # ClipYou (Eagle.Platform) embed (custom URL)
+        # referrer protected EaglePlatform embed
+        {
+            'url': 'https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/',
+            'info_dict': {
+                'id': '582306',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+                'thumbnail': r're:^https?://.*\.jpg$',
+                'duration': 3382,
+                'view_count': int,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        # ClipYou (EaglePlatform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
             # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
@@ -1190,6 +1247,9 @@ class GenericIE(InfoExtractor):
                 'duration': 216,
                 'view_count': int,
             },
+            'params': {
+                'skip_download': True,
+            },
         },
         # Pladform embed
         {
@@ -1410,6 +1470,22 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            # Brightcove embed with whitespace around attribute names
+            'url': 'http://www.stack.com/video/3167554373001/learn-to-hit-open-three-pointers-with-damian-lillard-s-baseline-drift-drill',
+            'info_dict': {
+                'id': '3167554373001',
+                'ext': 'mp4',
+                'title': "Learn to Hit Open Three-Pointers With Damian Lillard's Baseline Drift Drill",
+                'description': 'md5:57bacb0e0f29349de4972bfda3191713',
+                'uploader_id': '1079349493',
+                'upload_date': '20140207',
+                'timestamp': 1391810548,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
         # Another form of arte.tv embed
         {
             'url': 'http://www.tv-replay.fr/redirection/09-04-16/arte-reportage-arte-11508975.html',
@@ -1474,6 +1550,22 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['BrightcoveLegacy'],
         },
+        # Nexx embed
+        {
+            'url': 'https://www.funk.net/serien/5940e15073f6120001657956/items/593efbb173f6120001657503',
+            'info_dict': {
+                'id': '247746',
+                'ext': 'mp4',
+                'title': "Yesterday's Jam (OV)",
+                'description': 'md5:09bc0984723fed34e2581624a84e05f0',
+                'timestamp': 1492594816,
+                'upload_date': '20170419',
+            },
+            'params': {
+                'format': 'bestvideo',
+                'skip_download': True,
+            },
+        },
         # Facebook <iframe> embed
         {
             'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
@@ -1484,6 +1576,21 @@ class GenericIE(InfoExtractor):
                 'title': 'Facebook video #599637780109885',
             },
         },
+        # Facebook <iframe> embed, plugin video
+        {
+            'url': 'http://5pillarsuk.com/2017/06/07/tariq-ramadan-disagrees-with-pr-exercise-by-imams-refusing-funeral-prayers-for-london-attackers/',
+            'info_dict': {
+                'id': '1754168231264132',
+                'ext': 'mp4',
+                'title': 'About the Imams and Religious leaders refusing to perform funeral prayers for...',
+                'uploader': 'Tariq Ramadan (official)',
+                'timestamp': 1496758379,
+                'upload_date': '20170606',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
         # Facebook API embed
         {
             'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
@@ -1651,6 +1758,71 @@ class GenericIE(InfoExtractor):
             },
             '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,
+        },
+        {
+            'url': 'http://kron4.com/2017/04/28/standoff-with-walnut-creek-murder-suspect-ends-with-arrest/',
+            'info_dict': {
+                'id': 'standoff-with-walnut-creek-murder-suspect-ends-with-arrest',
+                'title': 'Standoff with Walnut Creek murder suspect ends',
+                'description': 'md5:3ccc48a60fc9441eeccfc9c469ebf788',
+            },
+            'playlist_mincount': 4,
+        },
+        {
+            # WashingtonPost embed
+            'url': 'http://www.vanityfair.com/hollywood/2017/04/donald-trump-tv-pitches',
+            'info_dict': {
+                'id': '8caf6e88-d0ec-11e5-90d3-34c2c42653ac',
+                'ext': 'mp4',
+                'title': "No one has seen the drama series based on Trump's life \u2014 until now",
+                'description': 'Donald Trump wanted a weekly TV drama based on his life. It never aired. But The Washington Post recently obtained a scene from the pilot script — and enlisted actors.',
+                'timestamp': 1455216756,
+                'uploader': 'The Washington Post',
+                'upload_date': '20160211',
+            },
+            '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()],
+        },
+        {
+            # JOJ.sk embeds
+            'url': 'https://www.noviny.sk/slovensko/238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+            'info_dict': {
+                'id': '238543-slovenskom-sa-prehnala-vlna-silnych-burok',
+                'title': 'Slovenskom sa prehnala vlna silných búrok',
+            },
+            'playlist_mincount': 5,
+            'add_ie': [JojIE.ie_key()],
+        },
+        {
+            # AMP embed (see https://www.ampproject.org/docs/reference/components/amp-video)
+            'url': 'https://tvrain.ru/amp/418921/',
+            'md5': 'cc00413936695987e8de148b67d14f1d',
+            'info_dict': {
+                'id': '418921',
+                'ext': 'mp4',
+                'title': 'Стас Намин: «Мы нарушили девственность Кремля»',
+            },
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -1825,14 +1997,14 @@ 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
                 }]
@@ -1950,6 +2122,13 @@ class GenericIE(InfoExtractor):
         video_description = self._og_search_description(webpage, default=None)
         video_thumbnail = self._og_search_thumbnail(webpage, default=None)
 
+        info_dict.update({
+            'title': video_title,
+            'description': video_description,
+            'thumbnail': video_thumbnail,
+            'age_limit': age_limit,
+        })
+
         # Look for Brightcove Legacy Studio embeds
         bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
         if bc_urls:
@@ -1971,6 +2150,11 @@ class GenericIE(InfoExtractor):
         if bc_urls:
             return self.playlist_from_matches(bc_urls, video_id, video_title, ie='BrightcoveNew')
 
+        # Look for Nexx embeds
+        nexx_urls = NexxIE._extract_urls(webpage)
+        if nexx_urls:
+            return self.playlist_from_matches(nexx_urls, video_id, video_title, ie=NexxIE.ie_key())
+
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:
@@ -2043,58 +2227,27 @@ class GenericIE(InfoExtractor):
                 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(
-            r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
-        if match:
-            embed_url = self._proto_relative_url(
-                unescapeHTML(match.group('url')))
-            return {
-                '_type': 'url_transparent',
-                'url': embed_url,
-                'ie_key': 'Wistia',
-                'uploader': video_uploader,
-            }
+        # Look for DailyMail embeds
+        dailymail_urls = DailyMailIE._extract_urls(webpage)
+        if dailymail_urls:
+            return self.playlist_from_matches(
+                dailymail_urls, video_id, video_title, ie=DailyMailIE.ie_key())
 
-        match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
-        if match:
+        # Look for embedded Wistia player
+        wistia_url = WistiaIE._extract_url(webpage)
+        if wistia_url:
             return {
                 '_type': 'url_transparent',
-                'url': 'wistia:%s' % match.group('id'),
-                'ie_key': 'Wistia',
+                'url': self._proto_relative_url(wistia_url),
+                'ie_key': WistiaIE.ie_key(),
                 'uploader': video_uploader,
             }
 
-        match = re.search(
-            r'''(?sx)
-                <script[^>]+src=(["'])(?:https?:)?//fast\.wistia\.com/assets/external/E-v1\.js\1[^>]*>.*?
-                <div[^>]+class=(["']).*?\bwistia_async_(?P<id>[a-z0-9]+)\b.*?\2
-            ''', webpage)
-        if match:
-            return self.url_result(self._proto_relative_url(
-                'wistia:%s' % match.group('id')), 'Wistia')
-
         # Look for SVT player
         svt_url = SVTIE._extract_url(webpage)
         if svt_url:
             return self.url_result(svt_url, 'SVT')
 
-        # Look for embedded condenast player
-        matches = re.findall(
-            r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
-            webpage)
-        if matches:
-            return {
-                '_type': 'playlist',
-                'entries': [{
-                    '_type': 'url',
-                    'ie_key': 'CondeNast',
-                    'url': ma,
-                } for ma in matches],
-                'title': video_title,
-                'id': video_id,
-            }
-
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
         if mobj is not None:
@@ -2176,9 +2329,9 @@ 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)
@@ -2375,12 +2528,12 @@ class GenericIE(InfoExtractor):
         if kaltura_url:
             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
 
-        # Look for Eagle.Platform embeds
+        # Look for EaglePlatform embeds
         eagleplatform_url = EaglePlatformIE._extract_url(webpage)
         if eagleplatform_url:
-            return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
+            return self.url_result(smuggle_url(eagleplatform_url, {'referrer': url}), EaglePlatformIE.ie_key())
 
-        # Look for ClipYou (uses Eagle.Platform) embeds
+        # Look for ClipYou (uses EaglePlatform) embeds
         mobj = re.search(
             r'<iframe[^>]+src="https?://(?P<host>media\.clipyou\.ru)/index/player\?.*\brecord_id=(?P<id>\d+).*"', webpage)
         if mobj is not None:
@@ -2483,28 +2636,16 @@ class GenericIE(InfoExtractor):
             return self.url_result(piksel_url, PikselIE.ie_key())
 
         # Look for Limelight embeds
-        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))
+        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'''(?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:
+            return self.playlist_result(
+                anvato_urls, video_id, video_title, video_description)
 
         # Look for AdobeTVVideo embeds
         mobj = re.search(
@@ -2621,20 +2762,46 @@ class GenericIE(InfoExtractor):
         rutube_urls = RutubeIE._extract_urls(webpage)
         if rutube_urls:
             return self.playlist_from_matches(
-                rutube_urls, ie=RutubeIE.ie_key())
+                rutube_urls, video_id, video_title, ie=RutubeIE.ie_key())
+
+        # Look for WashingtonPost embeds
+        wapo_urls = WashingtonPostIE._extract_urls(webpage)
+        if wapo_urls:
+            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())
+
+        # Look for JOJ.sk embeds
+        joj_urls = JojIE._extract_urls(webpage)
+        if joj_urls:
+            return self.playlist_from_matches(
+                joj_urls, video_id, video_title, ie=JojIE.ie_key())
+
+        def merge_dicts(dict1, dict2):
+            merged = {}
+            for k, v in dict1.items():
+                if v is not None:
+                    merged[k] = v
+            for k, v in dict2.items():
+                if v is None:
+                    continue
+                if (k not in merged or
+                        (isinstance(v, compat_str) and v and
+                            isinstance(merged[k], compat_str) and
+                            not merged[k])):
+                    merged[k] = v
+            return merged
 
         # Looking for http://schema.org/VideoObject
         json_ld = self._search_json_ld(
             webpage, video_id, default={}, expected_type='VideoObject')
         if json_ld.get('url'):
-            info_dict.update({
-                'title': video_title or info_dict['title'],
-                'description': video_description,
-                'thumbnail': video_thumbnail,
-                'age_limit': age_limit
-            })
-            info_dict.update(json_ld)
-            return info_dict
+            return merge_dicts(json_ld, info_dict)
 
         # Look for HTML5 media
         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
@@ -2652,9 +2819,7 @@ class GenericIE(InfoExtractor):
         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
+            return merge_dicts(info, info_dict)
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):