Stop libav compatibility links from being considered as genuine.

[youtubedl] / youtube_dl / extractor / prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index e4c4ad7145395c1805fa97c6f8e83024fa28fde8..c6eee3b72a6e428012644ee7c10caad9be78fa86 100644 (file)
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -5,8 +5,12 @@ import re
 
 from hashlib import sha1
 from .common import InfoExtractor
 
 from hashlib import sha1
 from .common import InfoExtractor

+from ..compat import compat_str

 from ..utils import (
 from ..utils import (

-    compat_urllib_parse,
+    ExtractorError,
+    determine_ext,
+    float_or_none,
+    int_or_none,

     unified_strdate,
 )
 
     unified_strdate,
 )
 


@@ -14,15 +18,20 @@ from ..utils import (
 class ProSiebenSat1IE(InfoExtractor):
     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'
 class ProSiebenSat1IE(InfoExtractor):
     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'

-    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany|7tv)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'

 
     _TESTS = [
         {
 
     _TESTS = [
         {

+            # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
+            # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
+            # - malformed f4m manifest support
+            # - proper handling of URLs starting with `https?://` in 2.0 manifests
+            # - recursive child f4m manifests extraction

             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
             'info_dict': {
                 'id': '2104602',
             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
             'info_dict': {
                 'id': '2104602',

-                'ext': 'mp4',
-                'title': 'Staffel 2, Episode 18 - Jahresrückblick',
+                'ext': 'flv',
+                'title': 'Episode 18 - Staffel 2',

                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,
                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,


@@ -62,6 +71,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
                 # rtmp download
                 'skip_download': True,
             },

+            'skip': 'This video is unavailable',

         },
         {
             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
         },
         {
             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',


@@ -77,6 +87,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
                 # rtmp download
                 'skip_download': True,
             },

+            'skip': 'This video is unavailable',

         },
         {
             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
         },
         {
             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',


@@ -85,13 +96,14 @@ class ProSiebenSat1IE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Im Interview: Kai Wiesinger',
                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
                 'ext': 'mp4',
                 'title': 'Im Interview: Kai Wiesinger',
                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',

-                'upload_date': '20140225',
+                'upload_date': '20140203',

                 'duration': 522.56,
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },
                 'duration': 522.56,
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },

+            'skip': 'This video is unavailable',

         },
         {
             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
         },
         {
             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',


@@ -100,13 +112,14 @@ class ProSiebenSat1IE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
                 'ext': 'mp4',
                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',

-                'upload_date': '20140225',
+                'upload_date': '20141014',

                 'duration': 2410.44,
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },
                 'duration': 2410.44,
             },
             'params': {
                 # rtmp download
                 'skip_download': True,
             },

+            'skip': 'This video is unavailable',

         },
         {
             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
         },
         {
             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',


@@ -122,19 +135,19 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
                 # rtmp download
                 'skip_download': True,
             },

+            'skip': 'This video is unavailable',

         },
         {
             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
             'info_dict': {
                 'id': '2572814',
         },
         {
             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
             'info_dict': {
                 'id': '2572814',

-                'ext': 'mp4',
+                'ext': 'flv',

                 'title': 'Andreas Kümmert: Rocket Man',
                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
                 'upload_date': '20131017',
                 'duration': 469.88,
             },
             'params': {
                 'title': 'Andreas Kümmert: Rocket Man',
                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
                 'upload_date': '20131017',
                 'duration': 469.88,
             },
             'params': {

-                # rtmp download

                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },


@@ -142,14 +155,34 @@ class ProSiebenSat1IE(InfoExtractor):
             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
             'info_dict': {
                 'id': '2156342',
             'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
             'info_dict': {
                 'id': '2156342',

-                'ext': 'mp4',
+                'ext': 'flv',

                 'title': 'Kurztrips zum Valentinstag',
                 'title': 'Kurztrips zum Valentinstag',

-                'description': 'md5:8ba6301e70351ae0bedf8da00f7ba528',
-                'upload_date': '20130206',
+                'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',

                 'duration': 307.24,
             },
             'params': {
                 'duration': 307.24,
             },
             'params': {

-                # rtmp download
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
+            'info_dict': {
+                'id': '439664',
+                'title': 'Episode 8 - Ganze Folge - Playlist',
+                'description': 'md5:63b8963e71f481782aeea877658dec84',
+            },
+            'playlist_count': 2,
+        },
+        {
+            'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
+            'info_dict': {
+                'id': '4187506',
+                'ext': 'flv',
+                'title': 'Best of Circus HalliGalli',
+                'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
+                'upload_date': '20151229',
+            },
+            'params': {

                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },


@@ -158,19 +191,26 @@ class ProSiebenSat1IE(InfoExtractor):
     _CLIPID_REGEXES = [
         r'"clip_id"\s*:\s+"(\d+)"',
         r'clipid: "(\d+)"',
     _CLIPID_REGEXES = [
         r'"clip_id"\s*:\s+"(\d+)"',
         r'clipid: "(\d+)"',

-        r'clipId=(\d+)',
+        r'clip[iI]d=(\d+)',
+        r'clip[iI]d\s*=\s*["\'](\d+)',
+        r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",

     ]
     _TITLE_REGEXES = [
         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
         r'<!-- start video -->\s*<h1>(.+?)</h1>',
     ]
     _TITLE_REGEXES = [
         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
         r'<!-- start video -->\s*<h1>(.+?)</h1>',

-        r'<div class="ep-femvideos-pi4-video-txt">\s*<h2>(.+?)</h2>',
+        r'<h1 class="att-name">\s*(.+?)</h1>',
+        r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
+        r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
+        r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',

     ]
     _DESCRIPTION_REGEXES = [
         r'<p itemprop="description">\s*(.+?)</p>',
         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
     ]
     _DESCRIPTION_REGEXES = [
         r'<p itemprop="description">\s*(.+?)</p>',
         r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
         r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',

-        r'<p>(.+?)</p>\s*<div class="ep-femvideos-pi4-video-footer">',
+        r'<p class="att-description">\s*(.+?)\s*</p>',
+        r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
+        r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',

     ]
     _UPLOAD_DATE_REGEXES = [
         r'<meta property="og:published_time" content="(.+?)">',
     ]
     _UPLOAD_DATE_REGEXES = [
         r'<meta property="og:published_time" content="(.+?)">',


@@ -179,102 +219,129 @@ class ProSiebenSat1IE(InfoExtractor):
         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
     ]
         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
         r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
     ]

+    _PAGE_TYPE_REGEXES = [
+        r'<meta name="page_type" content="([^"]+)">',
+        r"'itemType'\s*:\s*'([^']*)'",
+    ]
+    _PLAYLIST_ID_REGEXES = [
+        r'content[iI]d=(\d+)',
+        r"'itemId'\s*:\s*'([^']*)'",
+    ]
+    _PLAYLIST_CLIP_REGEXES = [
+        r'(?s)data-qvt=.+?<a href="([^"]+)"',
+    ]

 
 

-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
-        page = self._download_webpage(url, video_id, 'Downloading page')
-
-        clip_id = self._html_search_regex(self._CLIPID_REGEXES, page, 'clip id')
+    def _extract_clip(self, url, webpage):
+        clip_id = self._html_search_regex(
+            self._CLIPID_REGEXES, webpage, 'clip id')

 
 

-        access_token = 'testclient'
-        client_name = 'kolibri-1.2.5'
+        access_token = 'prosieben'
+        client_name = 'kolibri-2.0.19-splec4'

         client_location = url
 
         client_location = url
 

-        videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
-            'access_token': access_token,
-            'client_location': client_location,
-            'client_name': client_name,
-            'ids': clip_id,
-        })
+        video = self._download_json(
+            'http://vas.sim-technik.de/vas/live/v2/videos',
+            clip_id, 'Downloading videos JSON', query={
+                'access_token': access_token,
+                'client_location': client_location,
+                'client_name': client_name,
+                'ids': clip_id,
+            })[0]

 
 

-        videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
+        if video.get('is_protected') is True:
+            raise ExtractorError('This video is DRM protected.', expected=True)

 
 

-        duration = float(videos[0]['duration'])
-        source_ids = [source['id'] for source in videos[0]['sources']]
-        source_ids_str = ','.join(map(str, source_ids))
+        duration = float_or_none(video.get('duration'))
+        source_ids = [compat_str(source['id']) for source in video['sources']]

 
         g = '01!8d8F_)r9]4s[qeuXfP%'
 
         g = '01!8d8F_)r9]4s[qeuXfP%'

+        client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name]).encode('utf-8')).hexdigest()

 
 

-        client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
-                                 .encode('utf-8')).hexdigest()
-
-        sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
-            'access_token': access_token,
-            'client_id': client_id,
-            'client_location': client_location,
-            'client_name': client_name,
-        }))
-
-        sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
+        sources = self._download_json(
+            'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
+            clip_id, 'Downloading sources JSON', query={
+                'access_token': access_token,
+                'client_id': client_id,
+                'client_location': client_location,
+                'client_name': client_name,
+            })

         server_id = sources['server_id']
 
         server_id = sources['server_id']
 

-        client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
-                                          client_location, source_ids_str, g, client_name])
-                                 .encode('utf-8')).hexdigest()
-
-        url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
-            'access_token': access_token,
-            'client_id': client_id,
-            'client_location': client_location,
-            'client_name': client_name,
-            'server_id': server_id,
-            'source_ids': source_ids_str,
-        }))
-
-        urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
-
-        title = self._html_search_regex(self._TITLE_REGEXES, page, 'title')
-        description = self._html_search_regex(self._DESCRIPTION_REGEXES, page, 'description', fatal=False)
-        thumbnail = self._og_search_thumbnail(page)
-
-        upload_date = unified_strdate(self._html_search_regex(
-            self._UPLOAD_DATE_REGEXES, page, 'upload date', fatal=False))
-
-        formats = []
-
-        urls_sources = urls['sources']
-        if isinstance(urls_sources, dict):
-            urls_sources = urls_sources.values()
+        title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')

 
         def fix_bitrate(bitrate):
 
         def fix_bitrate(bitrate):

-            return bitrate / 1000 if bitrate % 1000 == 0 else bitrate
+            bitrate = int_or_none(bitrate)
+            if not bitrate:
+                return None
+            return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate

 
 

-        for source in urls_sources:
-            protocol = source['protocol']
-            if protocol == 'rtmp' or protocol == 'rtmpe':
-                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
-                if not mobj:
-                    continue
-                formats.append({
-                    'url': mobj.group('url'),
-                    'app': mobj.group('app'),
-                    'play_path': mobj.group('playpath'),
-                    'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
-                    'page_url': 'http://www.prosieben.de',
-                    'vbr': fix_bitrate(source['bitrate']),
-                    'ext': 'mp4',
-                    'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
-                })
-            else:
-                formats.append({
-                    'url': source['url'],
-                    'vbr': fix_bitrate(source['bitrate']),
+        formats = []
+        for source_id in source_ids:
+            client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, client_location, source_id, g, client_name]).encode('utf-8')).hexdigest()
+            urls = self._download_json(
+                'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
+                clip_id, 'Downloading urls JSON', fatal=False, query={
+                    'access_token': access_token,
+                    'client_id': client_id,
+                    'client_location': client_location,
+                    'client_name': client_name,
+                    'server_id': server_id,
+                    'source_ids': source_id,

                 })
                 })

-
+            if not urls:
+                continue
+            if urls.get('status_code') != 0:
+                raise ExtractorError('This video is unavailable', expected=True)
+            urls_sources = urls['sources']
+            if isinstance(urls_sources, dict):
+                urls_sources = urls_sources.values()
+            for source in urls_sources:
+                source_url = source.get('url')
+                if not source_url:
+                    continue
+                protocol = source.get('protocol')
+                mimetype = source.get('mimetype')
+                if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        source_url, clip_id, f4m_id='hds', fatal=False))
+                elif mimetype == 'application/x-mpegURL':
+                    formats.extend(self._extract_m3u8_formats(
+                        source_url, clip_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+                else:
+                    tbr = fix_bitrate(source['bitrate'])
+                    if protocol in ('rtmp', 'rtmpe'):
+                        mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
+                        if not mobj:
+                            continue
+                        path = mobj.group('path')
+                        mp4colon_index = path.rfind('mp4:')
+                        app = path[:mp4colon_index]
+                        play_path = path[mp4colon_index:]
+                        formats.append({
+                            'url': '%s/%s' % (mobj.group('url'), app),
+                            'app': app,
+                            'play_path': play_path,
+                            'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
+                            'page_url': 'http://www.prosieben.de',
+                            'tbr': tbr,
+                            'ext': 'flv',
+                            'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
+                        })
+                    else:
+                        formats.append({
+                            'url': source_url,
+                            'tbr': tbr,
+                            'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
+                        })

         self._sort_formats(formats)
 
         self._sort_formats(formats)
 

+        description = self._html_search_regex(
+            self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
+        thumbnail = self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._html_search_regex(
+            self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
+

         return {
             'id': clip_id,
             'title': title,
         return {
             'id': clip_id,
             'title': title,


@@ -283,4 +350,32 @@ class ProSiebenSat1IE(InfoExtractor):
             'upload_date': upload_date,
             'duration': duration,
             'formats': formats,
             'upload_date': upload_date,
             'duration': duration,
             'formats': formats,

-        }
\ No newline at end of file
+        }
+
+    def _extract_playlist(self, url, webpage):
+        playlist_id = self._html_search_regex(
+            self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
+        for regex in self._PLAYLIST_CLIP_REGEXES:
+            playlist_clips = re.findall(regex, webpage)
+            if playlist_clips:
+                title = self._html_search_regex(
+                    self._TITLE_REGEXES, webpage, 'title')
+                description = self._html_search_regex(
+                    self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
+                entries = [
+                    self.url_result(
+                        re.match('(.+?//.+?)/', url).group(1) + clip_path,
+                        'ProSiebenSat1')
+                    for clip_path in playlist_clips]
+                return self.playlist_result(entries, playlist_id, title, description)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        page_type = self._search_regex(
+            self._PAGE_TYPE_REGEXES, webpage,
+            'page type', default='clip').lower()
+        if page_type == 'clip':
+            return self._extract_clip(url, webpage)
+        elif page_type == 'playlist':
+            return self._extract_playlist(url, webpage)