Merge pull request #11901 from ThomasChr/randonplaylistorder

[youtube-dl] / youtube_dl / extractor / prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index 7335dc2af971d4bc546de15b8a24e01359037048..5091d8456faf3a4841ba770ea408aa75be806f83 100644 (file)
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -1,4 +1,4 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -85,6 +85,9 @@ class ProSiebenSat1BaseIE(InfoExtractor):
                     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'):
@@ -122,7 +125,17 @@ class ProSiebenSat1BaseIE(InfoExtractor):
 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|7tv)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?
+                        (?:
+                            (?:
+                                prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|7tv|advopedia
+                            )\.(?:de|at|ch)|
+                            ran\.de|fem\.com|advopedia\.de
+                        )
+                        /(?P<id>.+)
+                    '''
 
     _TESTS = [
         {
@@ -134,16 +147,12 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             '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,
             },
-            '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',
@@ -245,7 +254,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             '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',
@@ -259,7 +268,7 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
             '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,
@@ -276,12 +285,13 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
                 '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',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Best of Circus HalliGalli',
                 'description': 'md5:8849752efd90b9772c9db6fdf87fb9e9',
                 'upload_date': '20151229',
@@ -290,6 +300,24 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
                 'skip_download': True,
             },
         },
+        {
+            # 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,
+        },
+        {
+            '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'
@@ -344,7 +372,9 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
         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))
@@ -361,19 +391,28 @@ class ProSiebenSat1IE(ProSiebenSat1BaseIE):
     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)