Merge tag 'upstream/2016.12.01'

[youtubedl] / youtube_dl / extractor / prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index c6eee3b72a6e428012644ee7c10caad9be78fa86..7cc07a2ad5b88c51aa9f5d339839fd743727e17e 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
 from __future__ import unicode_literals
 
 import re


@@ -15,10 +15,124 @@ from ..utils import (
 )
 
 
 )
 
 

-class ProSiebenSat1IE(InfoExtractor):
+class ProSiebenSat1BaseIE(InfoExtractor):
+    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)
+
+        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']
+
+        def fix_bitrate(bitrate):
+            bitrate = int_or_none(bitrate)
+            if not bitrate:
+                return None
+            return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
+
+        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={
+                    '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))
+                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': duration,
+            'formats': formats,
+        }
+
+
+class ProSiebenSat1IE(ProSiebenSat1BaseIE):

     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'
     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 = [
         {
 
     _TESTS = [
         {


@@ -186,8 +300,29 @@ class ProSiebenSat1IE(InfoExtractor):
                 'skip_download': True,
             },
         },
                 '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'
+    _SALT = '01!8d8F_)r9]4s[qeuXfP%'
+    _CLIENT_NAME = 'kolibri-2.0.19-splec4'

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


@@ -234,140 +369,48 @@ class ProSiebenSat1IE(InfoExtractor):
     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')

-
-        access_token = 'prosieben'
-        client_name = 'kolibri-2.0.19-splec4'
-        client_location = url
-
-        video = self._download_json(
-            'http://vas.sim-technik.de/vas/live/v2/videos',
-            clip_id, 'Downloading videos JSON', query={
-                'access_token': access_token,
-                'client_location': client_location,
-                'client_name': client_name,
-                'ids': clip_id,
-            })[0]
-
-        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']]
-
-        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 = self._download_json(
-            'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
-            clip_id, 'Downloading sources JSON', query={
-                'access_token': access_token,
-                'client_id': client_id,
-                'client_location': client_location,
-                'client_name': client_name,
-            })
-        server_id = sources['server_id']
-

         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')

-
-        def fix_bitrate(bitrate):
-            bitrate = int_or_none(bitrate)
-            if not bitrate:
-                return None
-            return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
-
-        formats = []
-        for source_id in source_ids:
-            client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, client_location, source_id, g, 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': access_token,
-                    'client_id': client_id,
-                    'client_location': client_location,
-                    'client_name': 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))
-                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)
-
+        info = self._extract_video_info(url, clip_id)

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

-        return {
+        info.update({

             'id': clip_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'upload_date': upload_date,
             'id': clip_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'upload_date': upload_date,

-            'duration': duration,
-            'formats': formats,
-        }
+        })
+        return info

 
     def _extract_playlist(self, url, webpage):
         playlist_id = self._html_search_regex(
             self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
 
     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(
+                '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)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)