X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fprosiebensat1.py;h=536a42dc88a4e17bbd039289508521d1ea13e282;hb=c8e337450b63c3e3d93d1d6badf9475d0757690f;hp=8c9451f4053036e0575ed5551b137fe6d168b166;hpb=8d1c8cae9ce9027058d90b246b9fa30ea88602ef;p=youtube-dl.git

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index 8c9451f40..536a42dc8 100644
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -5,17 +5,19 @@ import re
 
 from hashlib import sha1
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_urllib_parse,
+)
+from ..utils import (
     unified_strdate,
-    ExtractorError,
+    int_or_none,
 )
 
 
 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)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)'
 
     _TESTS = [
         {
@@ -23,7 +25,7 @@ class ProSiebenSat1IE(InfoExtractor):
             'info_dict': {
                 'id': '2104602',
                 'ext': 'mp4',
-                'title': 'Staffel 2, Episode 18 - Jahresrückblick',
+                'title': 'Episode 18 - Staffel 2',
                 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                 'upload_date': '20131231',
                 'duration': 5845.04,
@@ -158,7 +160,7 @@ class ProSiebenSat1IE(InfoExtractor):
             'info_dict': {
                 'id': '439664',
                 'title': 'Episode 8 - Ganze Folge - Playlist',
-                'description': 'Das finale und härteste Duell aller Zeiten ist vorbei! Der Weltmeister für dieses Jahr steht! Alle packenden Duelle der achten Episode von "Joko gegen Klaas - das Duell um die Welt" seht ihr hier noch einmal in voller Länge!',
+                'description': 'md5:63b8963e71f481782aeea877658dec84',
             },
             'playlist_count': 2,
         },
@@ -175,6 +177,7 @@ class ProSiebenSat1IE(InfoExtractor):
         r'<header class="clearfix">\s*<h3>(.+?)</h3>',
         r'<!-- start video -->\s*<h1>(.+?)</h1>',
         r'<h1 class="att-name">\s*(.+?)</h1>',
+        r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
     ]
     _DESCRIPTION_REGEXES = [
         r'<p itemprop="description">\s*(.+?)</p>',
@@ -189,52 +192,23 @@ 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/>',
     ]
-    _ITEM_TYPE_REGEXES = [
+    _PAGE_TYPE_REGEXES = [
+        r'<meta name="page_type" content="([^"]+)">',
         r"'itemType'\s*:\s*'([^']*)'",
     ]
-    _ITEM_ID_REGEXES = [
+    _PLAYLIST_ID_REGEXES = [
+        r'content[iI]d=(\d+)',
         r"'itemId'\s*:\s*'([^']*)'",
     ]
-    _PLAYLIST_CLIPS_REGEXES = [
-        r'data-qvt=.+?<a href="([^"]+)"',
+    _PLAYLIST_CLIP_REGEXES = [
+        r'(?s)data-qvt=.+?<a href="([^"]+)"',
     ]
 
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        item_type = self._html_search_regex(self._ITEM_TYPE_REGEXES, webpage, 'item type', default='CLIP')
-        if item_type == 'CLIP':
-            return self._clip_extract(url, webpage)
-        elif item_type == 'PLAYLIST':
-            playlist_id = self._html_search_regex(self._ITEM_ID_REGEXES, webpage, 'playlist id')
-
-            for regex in self._PLAYLIST_CLIPS_REGEXES:
-                playlist_clips = re.findall(regex, webpage, re.DOTALL)
-                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)
-                    root_url = re.match('(.+?//.+?)/', url).group(1)
-
-                    return {
-                        '_type': 'playlist',
-                        'id': playlist_id,
-                        'title': title,
-                        'description': description,
-                        'entries': [self._clip_extract(root_url + clip_path) for clip_path in playlist_clips]
-                    }
-        else:
-            raise ExtractorError('Unknown item type "%s"' % item_type)
-
-    def _clip_extract(self, url, webpage=None):
-        if webpage is None:
-            video_id = self._match_id(url)
-            webpage = self._download_webpage(url, video_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-1.12.6'
         client_location = url
 
         videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
@@ -294,18 +268,25 @@ class ProSiebenSat1IE(InfoExtractor):
             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
 
         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>.+)$', source['url'])
+                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', 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': mobj.group('url'),
-                    'app': mobj.group('app'),
-                    'play_path': mobj.group('playpath'),
+                    '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']),
@@ -329,3 +310,31 @@ class ProSiebenSat1IE(InfoExtractor):
             'duration': duration,
             'formats': formats,
         }
+
+    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)