Prepate to upload.

[youtubedl] / youtube_dl / extractor / prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index 7cc07a2ad5b88c51aa9f5d339839fd743727e17e..e19a470a5eee5efda48fe88a694c7e3a62010963 100644 (file)
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -16,6 +16,11 @@ from ..utils import (
 
 
 class ProSiebenSat1BaseIE(InfoExtractor):
 
 
 class ProSiebenSat1BaseIE(InfoExtractor):

+    _GEO_COUNTRIES = ['DE']
+    _ACCESS_ID = None
+    _SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
+    _V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'
+

     def _extract_video_info(self, url, clip_id):
         client_location = url
 
     def _extract_video_info(self, url, clip_id):
         client_location = url
 


@@ -31,90 +36,128 @@ class ProSiebenSat1BaseIE(InfoExtractor):
         if video.get('is_protected') is True:
             raise ExtractorError('This video is DRM protected.', expected=True)
 
         if video.get('is_protected') is True:
             raise ExtractorError('This video is DRM protected.', expected=True)
 

-        duration = float_or_none(video.get('duration'))
-        source_ids = [compat_str(source['id']) for source in video['sources']]
-
-        client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()
-
-        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': self._TOKEN,
-                'client_id': client_id,
-                'client_location': client_location,
-                'client_name': self._CLIENT_NAME,
-            })
-        server_id = sources['server_id']
+        formats = []
+        if self._ACCESS_ID:
+            raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
+            server_token = (self._download_json(
+                self._V4_BASE_URL + 'protocols', clip_id,
+                'Downloading protocols JSON',
+                headers=self.geo_verification_headers(), query={
+                    'access_id': self._ACCESS_ID,
+                    'client_token': sha1((raw_ct).encode()).hexdigest(),
+                    'video_id': clip_id,
+                }, fatal=False) or {}).get('server_token')
+            if server_token:
+                urls = (self._download_json(
+                    self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
+                        'access_id': self._ACCESS_ID,
+                        'client_token': sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
+                        'protocols': self._SUPPORTED_PROTOCOLS,
+                        'server_token': server_token,
+                        'video_id': clip_id,
+                    }, fatal=False) or {}).get('urls') or {}
+                for protocol, variant in urls.items():
+                    source_url = variant.get('clear', {}).get('url')
+                    if not source_url:
+                        continue
+                    if protocol == 'dash':
+                        formats.extend(self._extract_mpd_formats(
+                            source_url, clip_id, mpd_id=protocol, fatal=False))
+                    elif protocol == 'hls':
+                        formats.extend(self._extract_m3u8_formats(
+                            source_url, clip_id, 'mp4', 'm3u8_native',
+                            m3u8_id=protocol, fatal=False))
+                    else:
+                        formats.append({
+                            'url': source_url,
+                            'format_id': protocol,
+                        })
+        if not formats:
+            source_ids = [compat_str(source['id']) for source in video['sources']]

 
 

-        def fix_bitrate(bitrate):
-            bitrate = int_or_none(bitrate)
-            if not bitrate:
-                return None
-            return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
+            client_id = self._SALT[:2] + sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode('utf-8')).hexdigest()

 
 

-        formats = []
-        for source_id in source_ids:
-            client_id = self._SALT[:2] + sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._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={
+            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': self._TOKEN,
                     'client_id': client_id,
                     'client_location': client_location,
                     'client_name': self._CLIENT_NAME,
                     'access_token': self._TOKEN,
                     'client_id': client_id,
                     'client_location': client_location,
                     'client_name': self._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:
+            server_id = sources['server_id']
+
+            def fix_bitrate(bitrate):
+                bitrate = int_or_none(bitrate)
+                if not bitrate:
+                    return None
+                return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
+
+            for source_id in source_ids:
+                client_id = self._SALT[:2] + sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._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': self._TOKEN,
+                        'client_id': client_id,
+                        'client_location': client_location,
+                        'client_name': self._CLIENT_NAME,
+                        'server_id': server_id,
+                        'source_ids': source_id,
+                    })
+                if not urls:

                     continue
                     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 ''),
-                        })
+                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))
+                    elif mimetype == 'application/dash+xml':
+                        formats.extend(self._extract_mpd_formats(
+                            source_url, clip_id, mpd_id='dash', fatal=False))

                     else:
                     else:

-                        formats.append({
-                            'url': source_url,
-                            'tbr': tbr,
-                            'format_id': 'http%s' % ('-%d' % tbr if tbr 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)
 
         return {
         self._sort_formats(formats)
 
         return {

-            'duration': duration,
+            'duration': float_or_none(video.get('duration')),

             'formats': formats,
         }
 
             'formats': formats,
         }
 


@@ -126,34 +169,31 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
                     https?://
                         (?:www\.)?
                         (?:
                     https?://
                         (?:www\.)?
                         (?:

+                            (?:beta\.)?

                             (?:
                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
                             )\.(?:de|at|ch)|
                             (?:
                                 prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
                             )\.(?:de|at|ch)|

-                            ran\.de|fem\.com|advopedia\.de
+                            ran\.de|fem\.com|advopedia\.de|galileo\.tv/video

                         )
                         /(?P<id>.+)
                     '''
 
     _TESTS = [
         {
                         )
                         /(?P<id>.+)
                     '''
 
     _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:
+            # Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
+            # in response to fixing https://github.com/ytdl-org/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',
             # - 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',

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

                 'title': 'Episode 18 - Staffel 2',
                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,
             },
                 'title': 'Episode 18 - Staffel 2',
                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,
             },

-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },

         },
         {
             'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
         },
         {
             'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',


@@ -255,7 +295,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             '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': 'flv',
+                'ext': 'mp4',

                 'title': 'Andreas Kümmert: Rocket Man',
                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
                 'upload_date': '20131017',
                 'title': 'Andreas Kümmert: Rocket Man',
                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
                 'upload_date': '20131017',


@@ -269,7 +309,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             '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': 'flv',
+                'ext': 'mp4',

                 'title': 'Kurztrips zum Valentinstag',
                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
                 'duration': 307.24,
                 'title': 'Kurztrips zum Valentinstag',
                 'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
                 'duration': 307.24,


@@ -286,12 +326,13 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
                 'description': 'md5:63b8963e71f481782aeea877658dec84',
             },
             'playlist_count': 2,
                 'description': 'md5:63b8963e71f481782aeea877658dec84',
             },
             'playlist_count': 2,

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

         },
         {
             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
             'info_dict': {
                 'id': '4187506',
         },
         {
             'url': 'http://www.7tv.de/circus-halligalli/615-best-of-circus-halligalli-ganze-folge',
             'info_dict': {
                 'id': '4187506',

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

                 'title': 'Best of Circus HalliGalli',
                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
                 'upload_date': '20151229',
                 'title': 'Best of Circus HalliGalli',
                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
                 'upload_date': '20151229',


@@ -300,6 +341,21 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
                 'skip_download': True,
             },
         },
                 'skip_download': True,
             },
         },

+        {
+            # title in <h2 class="subtitle">
+            'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
+            'info_dict': {
+                'id': '4895826',
+                'ext': 'mp4',
+                'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
+                'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
+                'upload_date': '20170302',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'skip': 'geo restricted to Germany',
+        },

         {
             # geo restricted to Germany
             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
         {
             # geo restricted to Germany
             'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',


@@ -310,6 +366,11 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
             'only_matching': True,
         },
             'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
             'only_matching': True,
         },

+        {
+            # geo restricted to Germany
+            'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
+            'only_matching': True,
+        },

         {
             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
             'only_matching': True,
         {
             'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
             'only_matching': True,


@@ -323,12 +384,19 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
     _TOKEN = 'prosieben'
     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
     _CLIENT_NAME = 'kolibri-2.0.19-splec4'
     _TOKEN = 'prosieben'
     _SALT = '01!8d8F_)r9]4s[qeuXfP%'
     _CLIENT_NAME = 'kolibri-2.0.19-splec4'

+
+    _ACCESS_ID = 'x_prosiebenmaxx-de'
+    _ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
+    _IV = 'Aeluchoc6aevechuipiexeeboowedaok'
+

     _CLIPID_REGEXES = [
         r'"clip_id"\s*:\s+"(\d+)"',
         r'clipid: "(\d+)"',
         r'clip[iI]d=(\d+)',
     _CLIPID_REGEXES = [
         r'"clip_id"\s*:\s+"(\d+)"',
         r'clipid: "(\d+)"',
         r'clip[iI]d=(\d+)',

-        r'clip[iI]d\s*=\s*["\'](\d+)',
+        r'clip[iI][dD]\s*=\s*["\'](\d+)',

         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
         r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",

+        r'proMamsId&quot;\s*:\s*&quot;(\d+)',
+        r'proMamsId"\s*:\s*"(\d+)',

     ]
     _TITLE_REGEXES = [
         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
     ]
     _TITLE_REGEXES = [
         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',


@@ -338,6 +406,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
         r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
         r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
         r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',

+        r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',

     ]
     _DESCRIPTION_REGEXES = [
         r'<p itemprop="description">\s*(.+?)</p>',
     ]
     _DESCRIPTION_REGEXES = [
         r'<p itemprop="description">\s*(.+?)</p>',


@@ -369,10 +438,14 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
     def _extract_clip(self, url, webpage):
         clip_id = self._html_search_regex(
             self._CLIPID_REGEXES, webpage, 'clip id')
     def _extract_clip(self, url, webpage):
         clip_id = self._html_search_regex(
             self._CLIPID_REGEXES, webpage, 'clip id')

-        title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
+        title = self._html_search_regex(
+            self._TITLE_REGEXES, webpage, 'title',
+            default=None) or self._og_search_title(webpage)

         info = self._extract_video_info(url, clip_id)
         description = self._html_search_regex(
         info = self._extract_video_info(url, clip_id)
         description = self._html_search_regex(

-            self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
+            self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
+        if description is None:
+            description = self._og_search_description(webpage)

         thumbnail = self._og_search_thumbnail(webpage)
         upload_date = unified_strdate(self._html_search_regex(
             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
         thumbnail = self._og_search_thumbnail(webpage)
         upload_date = unified_strdate(self._html_search_regex(
             self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))


@@ -391,7 +464,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
         playlist = self._parse_json(
             self._search_regex(
             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
         playlist = self._parse_json(
             self._search_regex(

-                'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
+                r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',

                 webpage, 'playlist'),
             playlist_id)
         entries = []
                 webpage, 'playlist'),
             playlist_id)
         entries = []


@@ -422,3 +495,6 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             return self._extract_clip(url, webpage)
         elif page_type == 'playlist':
             return self._extract_playlist(url, webpage)
             return self._extract_clip(url, webpage)
         elif page_type == 'playlist':
             return self._extract_playlist(url, webpage)

+        else:
+            raise ExtractorError(
+                'Unsupported page type %s' % page_type, expected=True)