X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fmtv.py;h=15df62649ff3d1366ff7fefd4de396f2c263d648;hb=071c10137b6b17b79ecfc8676736d5cc243022f6;hp=bc7f49ebbac86cda7aa1bb711076b783e24bfea8;hpb=6789defea9b1fc7ff631e9da8a281504167ced10;p=youtube-dl

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index bc7f49ebb..15df62649 100644
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -2,7 +2,7 @@ from __future__ import unicode_literals
 
 import re
 
-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
@@ -23,8 +23,9 @@ def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
 
-class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
+class MTVServicesInfoExtractor(InfoExtractor):
     _MOBILE_TEMPLATE = None
+    _LANG = None
 
     @staticmethod
     def _id_from_uri(uri):
@@ -95,25 +96,15 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
 
     def _extract_subtitles(self, mdoc, mtvn_id):
         subtitles = {}
-        FORMATS = {
-            'scc': 'cea-608',
-            'eia-608': 'cea-608',
-            'xml': 'ttml',
-        }
-        subtitles_format = FORMATS.get(
-            self._downloader.params.get('subtitlesformat'), 'ttml')
         for transcript in mdoc.findall('.//transcript'):
             if transcript.get('kind') != 'captions':
                 continue
             lang = transcript.get('srclang')
-            for typographic in transcript.findall('./typographic'):
-                captions_format = typographic.get('format')
-                if captions_format == subtitles_format:
-                    subtitles[lang] = compat_str(typographic.get('src'))
-                    break
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(mtvn_id, subtitles)
-        return self.extract_subtitles(mtvn_id, subtitles)
+            subtitles[lang] = [{
+                'url': compat_str(typographic.get('src')),
+                'ext': typographic.get('format')
+            } for typographic in transcript.findall('./typographic')]
+        return subtitles
 
     def _get_video_info(self, itemdoc):
         uri = itemdoc.find('guid').text
@@ -128,6 +119,14 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
         mediagen_doc = self._download_xml(mediagen_url, video_id,
                                           'Downloading video urls')
 
+        item = mediagen_doc.find('./video/item')
+        if item is not None and item.get('type') == 'text':
+            message = '%s returned error: ' % self.IE_NAME
+            if item.get('code') is not None:
+                message += '%s - ' % item.get('code')
+            message += item.text
+            raise ExtractorError(message, expected=True)
+
         description_node = itemdoc.find('description')
         if description_node is not None:
             description = description_node.text.strip()
@@ -171,8 +170,12 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
         video_id = self._id_from_uri(uri)
         feed_url = self._get_feed_url(uri)
         data = compat_urllib_parse.urlencode({'uri': uri})
+        info_url = feed_url + '?'
+        if self._LANG:
+            info_url += 'lang=%s&' % self._LANG
+        info_url += data
         idoc = self._download_xml(
-            feed_url + '?' + data, video_id,
+            info_url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
         return self.playlist_result(
             [self._get_video_info(item) for item in idoc.findall('.//item')])
@@ -196,8 +199,6 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
                 webpage, 'mgid')
 
         videos_info = self._get_videos_info(mgid)
-        if self._downloader.params.get('listsubtitles', False):
-            return
         return videos_info
 
 
@@ -287,3 +288,40 @@ class MTVIggyIE(MTVServicesInfoExtractor):
         }
     }
     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
+
+class MTVDEIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtv.de'
+    _VALID_URL = r'''(?x)^https?://(?:www\.)?mtv\.de(?P<video_path>/artists/.*)'''
+    _TESTS = [
+        {
+            'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
+            'info_dict': {
+                'id': 'a50bc5f0b3aa4b3190aa',
+                'ext': 'mp4',
+                'title': 'cro-traum',
+                'description': 'Cro - Traum',
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        return self._get_videos_info(url, mobj.group('video_path'))
+
+    def _get_videos_info(self, url, video_path):
+        webpage = self._download_webpage(url, video_path)
+        playlist_js = self._search_regex(r'<script>\s*window.pagePlaylist =(.*?\]);\s*window.trackingParams =', webpage, 'playlist', flags=re.DOTALL)
+        playlist = self._parse_json(playlist_js, video_path)
+        info = None
+        for item in playlist:
+            if item['video_path'] == video_path:
+                info = item
+                break
+        if info == None:
+            raise ExtractorError('video not in playlist')
+        mrss_url = info['mrss']
+        idoc = self._download_xml(
+            mrss_url, video_path,
+            'Downloading info', transform_source=fix_xml_ampersands)
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')])