X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fbbc.py;h=50c1da185b74695f9197826e7e217fbaffe495c8;hb=0385aa6199206e4ba7745efec73be26c5826286a;hp=7fb80aa38fc39825fd7338b40fc994ccc0c8a185;hpb=3047121c639428235191ff5f7afbda7ecda38779;p=youtube-dl

diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py
index 7fb80aa38..50c1da185 100644
--- a/youtube_dl/extractor/bbc.py
+++ b/youtube_dl/extractor/bbc.py
@@ -10,7 +10,6 @@ from ..utils import (
     int_or_none,
     parse_duration,
     parse_iso8601,
-    remove_end,
     unescapeHTML,
 )
 from ..compat import (
@@ -23,7 +22,17 @@ class BBCCoUkIE(InfoExtractor):
     IE_NAME = 'bbc.co.uk'
     IE_DESC = 'BBC iPlayer'
     _ID_REGEX = r'[pb][\da-z]{7}'
-    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:programmes/(?!articles/)|iplayer(?:/[^/]+)?/(?:episode/|playlist/))|music/clips[/#])(?P<id>%s)' % _ID_REGEX
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?bbc\.co\.uk/
+                        (?:
+                            programmes/(?!articles/)|
+                            iplayer(?:/[^/]+)?/(?:episode/|playlist/)|
+                            music/clips[/#]|
+                            radio/player/
+                        )
+                        (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
+                    ''' % _ID_REGEX
 
     _MEDIASELECTOR_URLS = [
         # Provides HQ HLS streams with even better quality that pc mediaset but fails
@@ -35,6 +44,8 @@ class BBCCoUkIE(InfoExtractor):
 
     _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
     _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
+    # Unified Streaming Platform
+    _USP_RE = r'/([^/]+)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
 
     _NAMESPACES = (
         _MEDIASELECTION_NS,
@@ -46,13 +57,11 @@ class BBCCoUkIE(InfoExtractor):
             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
             'info_dict': {
                 'id': 'b039d07m',
-                'ext': 'flv',
-                'title': 'Kaleidoscope, Leonard Cohen',
+                'ext': 'mp4',
+                'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
-                'duration': 1740,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         },
@@ -77,14 +86,14 @@ class BBCCoUkIE(InfoExtractor):
                 'id': 'b00yng1d',
                 'ext': 'flv',
                 'title': 'The Voice UK: Series 3: Blind Auditions 5',
-                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'description': 'Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.',
                 'duration': 5100,
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+            'skip': 'this episode is not currently available',
         },
         {
             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
@@ -99,7 +108,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+            'skip': 'this episode is not currently available',
         }, {
             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
             'info_dict': {
@@ -112,19 +121,19 @@ class BBCCoUkIE(InfoExtractor):
             'params': {
                 # rtmp download
                 'skip_download': True,
-            }
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
         }, {
-            'url': 'http://www.bbc.co.uk/music/clips/p02frcc3',
+            'url': 'http://www.bbc.co.uk/music/clips/p022h44b',
             'note': 'Audio',
             'info_dict': {
-                'id': 'p02frcch',
-                'ext': 'flv',
-                'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix',
-                'description': 'French house superstar Madeon takes us out of the club and onto the after party.',
-                'duration': 3507,
+                'id': 'p022h44j',
+                'ext': 'mp4',
+                'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
+                'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
+                'duration': 227,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         }, {
@@ -132,13 +141,12 @@ class BBCCoUkIE(InfoExtractor):
             'note': 'Video',
             'info_dict': {
                 'id': 'p025c103',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
                 'duration': 226,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         }, {
@@ -154,7 +162,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'geolocation',
+            'skip': 'this episode is not currently available',
         }, {
             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
             'info_dict': {
@@ -168,22 +176,34 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'geolocation',
+            'skip': 'this episode is not currently available',
         }, {
             # iptv-all mediaset fails with geolocation however there is no geo restriction
             # for this programme at all
-            'url': 'http://www.bbc.co.uk/programmes/b06bp7lf',
+            'url': 'http://www.bbc.co.uk/programmes/b06rkn85',
             'info_dict': {
-                'id': 'b06bp7kf',
+                'id': 'b06rkms3',
                 'ext': 'flv',
-                'title': "Annie Mac's Friday Night, B.Traits sits in for Annie",
-                'description': 'B.Traits sits in for Annie Mac with a Mini-Mix from Disclosure.',
-                'duration': 10800,
+                'title': "Best of the Mini-Mixes 2015: Part 3, Annie Mac's Friday Night - BBC Radio 1",
+                'description': "Annie has part three in the Best of the Mini-Mixes 2015, plus the year's Most Played!",
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'this episode is not currently available on BBC iPlayer Radio',
+        }, {
+            # compact player (https://github.com/rg3/youtube-dl/issues/8147)
+            'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
+            'info_dict': {
+                'id': 'p028bfkj',
+                'ext': 'mp4',
+                'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+                'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
+            },
+            'params': {
+                'skip_download': True,
+            },
         }, {
             'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4',
             'only_matching': True,
@@ -193,6 +213,9 @@ class BBCCoUkIE(InfoExtractor):
         }, {
             'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo',
             'only_matching': True,
+        }, {
+            'url': 'http://www.bbc.co.uk/radio/player/p03cchwf',
+            'only_matching': True,
         }
     ]
 
@@ -223,11 +246,15 @@ class BBCCoUkIE(InfoExtractor):
             elif transfer_format == 'dash':
                 pass
             elif transfer_format == 'hls':
+                is_unified_streaming = re.search(self._USP_RE, href)
+                if is_unified_streaming:
+                    href = re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href)
                 m3u8_formats = self._extract_m3u8_formats(
                     href, programme_id, ext='mp4', entry_protocol='m3u8_native',
                     m3u8_id=supplier, fatal=False)
-                if m3u8_formats:
-                    formats.extend(m3u8_formats)
+                if is_unified_streaming:
+                    self._check_formats(m3u8_formats, programme_id)
+                formats.extend(m3u8_formats)
             # Direct link
             else:
                 formats.append({
@@ -282,13 +309,14 @@ class BBCCoUkIE(InfoExtractor):
         for connection in self._extract_connections(media):
             conn_formats = self._extract_connection(connection, programme_id)
             for format in conn_formats:
-                format.update({
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                    'vcodec': vcodec,
-                    'filesize': file_size,
-                })
+                if format.get('protocol') != 'm3u8_native':
+                    format.update({
+                        'width': width,
+                        'height': height,
+                        'vbr': vbr,
+                        'vcodec': vcodec,
+                        'filesize': file_size,
+                    })
                 if service:
                     format['format_id'] = '%s_%s' % (service, format['format_id'])
             formats.extend(conn_formats)
@@ -306,6 +334,7 @@ class BBCCoUkIE(InfoExtractor):
                     'format_id': '%s_%s' % (service, format['format_id']),
                     'abr': abr,
                     'acodec': acodec,
+                    'vcodec': 'none',
                 })
             formats.extend(conn_formats)
         return formats
@@ -454,6 +483,7 @@ class BBCCoUkIE(InfoExtractor):
         webpage = self._download_webpage(url, group_id, 'Downloading video page')
 
         programme_id = None
+        duration = None
 
         tviplayer = self._search_regex(
             r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById',
@@ -470,10 +500,15 @@ class BBCCoUkIE(InfoExtractor):
 
         if programme_id:
             formats, subtitles = self._download_media_selector(programme_id)
-            title = self._og_search_title(webpage)
+            title = self._og_search_title(webpage, default=None) or self._html_search_regex(
+                (r'<h2[^>]+id="parent-title"[^>]*>(.+?)</h2>',
+                 r'<div[^>]+class="info"[^>]*>\s*<h1>(.+?)</h1>'), webpage, 'title')
             description = self._search_regex(
-                r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
-                webpage, 'description', fatal=False)
+                (r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>',
+                 r'<div[^>]+class="info_+synopsis"[^>]*>([^<]+)</div>'),
+                webpage, 'description', default=None)
+            if not description:
+                description = self._html_search_meta('description', webpage)
         else:
             programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
 
@@ -532,7 +567,7 @@ class BBCIE(BBCCoUkIE):
         'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460',
         'info_dict': {
             'id': '3662a707-0af9-3149-963f-47bea720b460',
-            'title': 'BBC Blogs - Adam Curtis - BUGGER',
+            'title': 'BUGGER',
         },
         'playlist_count': 18,
     }, {
@@ -587,6 +622,7 @@ class BBCIE(BBCCoUkIE):
             'ext': 'mp4',
             'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
             'duration': 56,
+            'description': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''',
         },
         'params': {
             'skip_download': True,
@@ -640,9 +676,18 @@ class BBCIE(BBCCoUkIE):
         'url': 'http://www.bbc.com/sport/0/football/34475836',
         'info_dict': {
             'id': '34475836',
-            'title': 'What Liverpool can expect from Klopp',
+            'title': 'Jurgen Klopp: Furious football from a witty and winning coach',
+            'description': 'Fast-paced football, wit, wisdom and a ready smile - why Liverpool fans should come to love new boss Jurgen Klopp.',
         },
         'playlist_count': 3,
+    }, {
+        # school report article with single video
+        'url': 'http://www.bbc.co.uk/schoolreport/35744779',
+        'info_dict': {
+            'id': '35744779',
+            'title': 'School which breaks down barriers in Jerusalem',
+        },
+        'playlist_count': 1,
     }, {
         # single video with playlist URL from weather section
         'url': 'http://www.bbc.com/weather/features/33601775',
@@ -651,11 +696,17 @@ class BBCIE(BBCCoUkIE):
         # custom redirection to www.bbc.com
         'url': 'http://www.bbc.co.uk/news/science-environment-33661876',
         'only_matching': True,
+    }, {
+        # single video article embedded with data-media-vpid
+        'url': 'http://www.bbc.co.uk/sport/rowing/35908187',
+        'only_matching': True,
     }]
 
     @classmethod
     def suitable(cls, url):
-        return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url)
+        EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
+        return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
+                else super(BBCIE, cls).suitable(url))
 
     def _extract_from_media_meta(self, media_meta, video_id):
         # Direct links to media in media metadata (e.g.
@@ -703,19 +754,19 @@ class BBCIE(BBCCoUkIE):
 
         webpage = self._download_webpage(url, playlist_id)
 
-        timestamp = None
-        playlist_title = None
-        playlist_description = None
+        json_ld_info = self._search_json_ld(webpage, playlist_id, default=None)
+        timestamp = json_ld_info.get('timestamp')
 
-        ld = self._parse_json(
-            self._search_regex(
-                r'(?s)<script type="application/ld\+json">(.+?)</script>',
-                webpage, 'ld json', default='{}'),
-            playlist_id, fatal=False)
-        if ld:
-            timestamp = parse_iso8601(ld.get('datePublished'))
-            playlist_title = ld.get('headline')
-            playlist_description = ld.get('articleBody')
+        playlist_title = json_ld_info.get('title')
+        if not playlist_title:
+            playlist_title = self._og_search_title(
+                webpage, default=None) or self._html_search_regex(
+                r'<title>(.+?)</title>', webpage, 'playlist title', default=None)
+            if playlist_title:
+                playlist_title = re.sub(r'(.+)\s*-\s*BBC.*?$', r'\1', playlist_title).strip()
+
+        playlist_description = json_ld_info.get(
+            'description') or self._og_search_description(webpage, default=None)
 
         if not timestamp:
             timestamp = parse_iso8601(self._search_regex(
@@ -729,6 +780,7 @@ class BBCIE(BBCCoUkIE):
         # article with multiple videos embedded with playlist.sxml (e.g.
         # http://www.bbc.com/sport/0/football/34475836)
         playlists = re.findall(r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', webpage)
+        playlists.extend(re.findall(r'data-media-id="([^"]+/playlist\.sxml)"', webpage))
         if playlists:
             entries = [
                 self._extract_from_playlist_sxml(playlist_url, playlist_id, timestamp)
@@ -775,13 +827,11 @@ class BBCIE(BBCCoUkIE):
                                 playlist.get('progressiveDownloadUrl'), playlist_id, timestamp))
 
         if entries:
-            playlist_title = playlist_title or remove_end(self._og_search_title(webpage), ' - BBC News')
-            playlist_description = playlist_description or self._og_search_description(webpage, default=None)
             return self.playlist_result(entries, playlist_id, playlist_title, playlist_description)
 
         # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret)
         programme_id = self._search_regex(
-            [r'data-video-player-vpid="(%s)"' % self._ID_REGEX,
+            [r'data-(?:video-player|media)-vpid="(%s)"' % self._ID_REGEX,
              r'<param[^>]+name="externalIdentifier"[^>]+value="(%s)"' % self._ID_REGEX,
              r'videoId\s*:\s*["\'](%s)["\']' % self._ID_REGEX],
             webpage, 'vpid', default=None)
@@ -807,10 +857,6 @@ class BBCIE(BBCCoUkIE):
                 'subtitles': subtitles,
             }
 
-        playlist_title = self._html_search_regex(
-            r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title')
-        playlist_description = self._og_search_description(webpage, default=None)
-
         def extract_all(pattern):
             return list(filter(None, map(
                 lambda s: self._parse_json(s, playlist_id, fatal=False),
@@ -910,7 +956,7 @@ class BBCIE(BBCCoUkIE):
 
 
 class BBCCoUkArticleIE(InfoExtractor):
-    _VALID_URL = 'http://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
+    _VALID_URL = r'https?://www.bbc.co.uk/programmes/articles/(?P<id>[a-zA-Z0-9]+)'
     IE_NAME = 'bbc.co.uk:article'
     IE_DESC = 'BBC articles'
 
@@ -937,3 +983,72 @@ class BBCCoUkArticleIE(InfoExtractor):
             r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
 
         return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkPlaylistBaseIE(InfoExtractor):
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result(self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
+            for video_id in re.findall(
+                self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage)]
+
+        title, description = self._extract_title_and_description(webpage)
+
+        return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:iplayer:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/episodes/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
+    _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
+    _TEST = {
+        'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 6,
+    }
+
+    def _extract_title_and_description(self, webpage):
+        title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
+        description = self._search_regex(
+            r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
+            webpage, 'description', fatal=False, group='value')
+        return title, description
+
+
+class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
+    _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
+    _TESTS = [{
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance - Clips - BBC Four',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 7,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
+        'only_matching': True,
+    }]
+
+    def _extract_title_and_description(self, webpage):
+        title = self._og_search_title(webpage, fatal=False)
+        description = self._og_search_description(webpage)
+        return title, description