X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/38bb9b1b0a044cabaf5691553815e334cd2e9213..39393b81acfaf4045fb7f20454a0226f0dc9142e:/youtube_dl/extractor/prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index fec008c..e470882 100644
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -1,29 +1,192 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from hashlib import sha1
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse,
-)
+from ..compat import compat_str
 from ..utils import (
+    ExtractorError,
     determine_ext,
+    float_or_none,
     int_or_none,
+    merge_dicts,
     unified_strdate,
 )
 
 
-class ProSiebenSat1IE(InfoExtractor):
+class ProSiebenSat1BaseIE(InfoExtractor):
+    _GEO_BYPASS = False
+    _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
+
+        video = self._download_json(
+            'http://vas.sim-technik.de/vas/live/v2/videos',
+            clip_id, 'Downloading videos JSON', query={
+                'access_token': self._TOKEN,
+                'client_location': client_location,
+                'client_name': self._CLIENT_NAME,
+                'ids': clip_id,
+            })[0]
+
+        if video.get('is_protected') is True:
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        formats = []
+        if self._ACCESS_ID:
+            raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
+            protocols = 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, expected_status=(403,)) or {}
+            error = protocols.get('error') or {}
+            if error.get('title') == 'Geo check failed':
+                self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
+            server_token = protocols.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']]
+
+            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']
+
+            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
+                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:
+                        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 {
+            'duration': float_or_none(video.get('duration')),
+            'formats': formats,
+        }
+
+
+class ProSiebenSat1IE(ProSiebenSat1BaseIE):
     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'
-    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?
+                        (?:
+                            (?:beta\.)?
+                            (?:
+                                prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|advopedia
+                            )\.(?:de|at|ch)|
+                            ran\.de|fem\.com|advopedia\.de|galileo\.tv/video
+                        )
+                        /(?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
@@ -31,14 +194,14 @@ class ProSiebenSat1IE(InfoExtractor):
             'info_dict': {
                 'id': '2104602',
                 'ext': 'mp4',
-                'title': 'Episode 18 - Staffel 2',
+                'title': 'CIRCUS HALLIGALLI - Episode 18 - Staffel 2',
                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
+                'series': 'CIRCUS HALLIGALLI',
+                'season_number': 2,
+                'episode': 'Episode 18 - Staffel 2',
+                'episode_number': 18,
             },
         },
         {
@@ -71,6 +234,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
@@ -86,6 +250,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
@@ -101,6 +266,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # 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',
@@ -116,6 +282,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # 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',
@@ -131,33 +298,33 @@ class ProSiebenSat1IE(InfoExtractor):
                 # 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',
                 'ext': 'mp4',
-                'title': 'Andreas Kümmert: Rocket Man',
+                'title': 'The Voice of Germany - Andreas Kümmert: Rocket Man',
                 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
+                'timestamp': 1382041620,
                 'upload_date': '20131017',
                 'duration': 469.88,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
         {
-            'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
+            'url': 'http://www.fem.com/videos/beauty-lifestyle/kurztrips-zum-valentinstag',
             'info_dict': {
                 'id': '2156342',
                 'ext': 'mp4',
                 'title': 'Kurztrips zum Valentinstag',
-                'description': 'Romantischer Kurztrip zum Valentinstag? Wir verraten, was sich hier wirklich lohnt.',
+                'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
                 'duration': 307.24,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -169,14 +336,64 @@ class ProSiebenSat1IE(InfoExtractor):
                 'description': 'md5:63b8963e71f481782aeea877658dec84',
             },
             'playlist_count': 2,
+            'skip': 'This video is unavailable',
+        },
+        {
+            # 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',
+            'only_matching': True,
+        },
+        {
+            # geo restricted to Germany
+            '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.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
+            'only_matching': True,
         },
     ]
 
+    _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+)',
+        r'clip[iI][dD]\s*=\s*["\'](\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>',
@@ -184,15 +401,19 @@ class ProSiebenSat1IE(InfoExtractor):
         r'<!-- start video -->\s*<h1>(.+?)</h1>',
         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>',
+        r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
     ]
     _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 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="(.+?)">',
         r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
         r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
         r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
@@ -211,131 +432,58 @@ class ProSiebenSat1IE(InfoExtractor):
     ]
 
     def _extract_clip(self, url, webpage):
-        clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
-
-        access_token = 'prosieben'
-        client_name = 'kolibri-2.0.19-splec4'
-        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,
-        })
-
-        videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
-
-        duration = float(videos[0]['duration'])
-        source_ids = [source['id'] for source in videos[0]['sources']]
-        source_ids_str = ','.join(map(str, source_ids))
-
-        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()
-
-        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')
-        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, webpage, 'title')
-        description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
+        clip_id = self._html_search_regex(
+            self._CLIPID_REGEXES, webpage, 'clip id')
+        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(
+            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_meta('og:published_time', webpage,
+                                   'upload date', default=None)
+            or self._html_search_regex(self._UPLOAD_DATE_REGEXES,
+                                       webpage, 'upload date', default=None))
 
-        upload_date = unified_strdate(self._html_search_regex(
-            self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
-
-        formats = []
-
-        urls_sources = urls['sources']
-        if isinstance(urls_sources, dict):
-            urls_sources = urls_sources.values()
-
-        def fix_bitrate(bitrate):
-            bitrate = int_or_none(bitrate)
-            if not bitrate:
-                return None
-            return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
+        json_ld = self._search_json_ld(webpage, clip_id, default={})
 
-        for source in urls_sources:
-            protocol = source['protocol']
-            source_url = source['url']
-            if protocol == 'rtmp' or protocol == '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',
-                    'vbr': fix_bitrate(source['bitrate']),
-                    'ext': 'mp4',
-                    'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
-                })
-            elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
-                formats.extend(self._extract_f4m_formats(source_url, clip_id))
-            else:
-                formats.append({
-                    'url': source_url,
-                    'vbr': fix_bitrate(source['bitrate']),
-                })
-
-        self._sort_formats(formats)
-
-        return {
+        return merge_dicts(info, {
             'id': clip_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'upload_date': upload_date,
-            'duration': duration,
-            'formats': formats,
-        }
+        }, json_ld)
 
     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)
+        playlist = self._parse_json(
+            self._search_regex(
+                r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
+                webpage, 'playlist'),
+            playlist_id)
+        entries = []
+        for item in playlist:
+            clip_id = item.get('id') or item.get('upc')
+            if not clip_id:
+                continue
+            info = self._extract_video_info(url, clip_id)
+            info.update({
+                'id': clip_id,
+                'title': item.get('title') or item.get('teaser', {}).get('headline'),
+                'description': item.get('teaser', {}).get('description'),
+                'thumbnail': item.get('poster'),
+                'duration': float_or_none(item.get('duration')),
+                'series': item.get('tvShowTitle'),
+                'uploader': item.get('broadcastPublisher'),
+            })
+            entries.append(info)
+        return self.playlist_result(entries, playlist_id)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -347,3 +495,6 @@ class ProSiebenSat1IE(InfoExtractor):
             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)