X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fprosiebensat1.py;h=385681d06e3dda356193d9f89c7ccbdd4cbde453;hb=9932a65370564e08be11b0611e52b63ff451742d;hp=c6e539e5aa911ceb7d9cb79d8d9d317abbaf8dbe;hpb=c84890f7082237b02a5fc61657b2ea513836a85e;p=youtube-dl

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index c6e539e5a..385681d06 100644
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -5,10 +5,11 @@ 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,
 )
 
 
@@ -86,7 +87,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Im Interview: Kai Wiesinger',
                 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
-                'upload_date': '20140225',
+                'upload_date': '20140203',
                 'duration': 522.56,
             },
             'params': {
@@ -101,7 +102,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
                 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
-                'upload_date': '20140225',
+                'upload_date': '20141014',
                 'duration': 2410.44,
             },
             'params': {
@@ -158,7 +159,7 @@ 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,
         },
@@ -168,6 +169,7 @@ class ProSiebenSat1IE(InfoExtractor):
         r'"clip_id"\s*:\s+"(\d+)"',
         r'clipid: "(\d+)"',
         r'clip[iI]d=(\d+)',
+        r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
     ]
     _TITLE_REGEXES = [
         r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
@@ -188,48 +190,19 @@ 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'
@@ -328,3 +301,31 @@ 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)