X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fmtv.py;h=15df62649ff3d1366ff7fefd4de396f2c263d648;hb=071c10137b6b17b79ecfc8676736d5cc243022f6;hp=5ebc78033a4abbb98310096c279fe11459b4a791;hpb=2f15832f569834a37ac3ee6140a3b997407c04cd;p=youtube-dl

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index 5ebc78033..15df62649 100644
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+    compat_str,
 )
 from ..utils import (
     ExtractorError,
@@ -24,6 +25,7 @@ def _media_xml_tag(tag):
 
 class MTVServicesInfoExtractor(InfoExtractor):
     _MOBILE_TEMPLATE = None
+    _LANG = None
 
     @staticmethod
     def _id_from_uri(uri):
@@ -53,7 +55,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
         req = compat_urllib_request.Request(webpage_url)
         # Otherwise we get a webpage that would execute some javascript
-        req.add_header('Youtubedl-user-agent', 'curl/7')
+        req.add_header('User-Agent', 'curl/7')
         webpage = self._download_webpage(req, mtvn_id,
                                          'Downloading mobile page')
         metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
@@ -78,17 +80,32 @@ class MTVServicesInfoExtractor(InfoExtractor):
             try:
                 _, _, ext = rendition.attrib['type'].partition('/')
                 rtmp_video_url = rendition.find('./src').text
-                formats.append({'ext': ext,
-                                'url': self._transform_rtmp_url(rtmp_video_url),
-                                'format_id': rendition.get('bitrate'),
-                                'width': int(rendition.get('width')),
-                                'height': int(rendition.get('height')),
-                                })
+                if rtmp_video_url.endswith('siteunavail.png'):
+                    continue
+                formats.append({
+                    'ext': ext,
+                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'format_id': rendition.get('bitrate'),
+                    'width': int(rendition.get('width')),
+                    'height': int(rendition.get('height')),
+                })
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)
         return formats
 
+    def _extract_subtitles(self, mdoc, mtvn_id):
+        subtitles = {}
+        for transcript in mdoc.findall('.//transcript'):
+            if transcript.get('kind') != 'captions':
+                continue
+            lang = transcript.get('srclang')
+            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
         video_id = self._id_from_uri(uri)
@@ -102,6 +119,14 @@ class MTVServicesInfoExtractor(InfoExtractor):
         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()
@@ -135,6 +160,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         return {
             'title': title,
             'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
+            'subtitles': self._extract_subtitles(mediagen_doc, mtvn_id),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
@@ -144,8 +170,12 @@ class MTVServicesInfoExtractor(InfoExtractor):
         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')])
@@ -167,7 +197,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, 'mgid')
-        return self._get_videos_info(mgid)
+
+        videos_info = self._get_videos_info(mgid)
+        return videos_info
 
 
 class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
@@ -212,25 +244,14 @@ class MTVIE(MTVServicesInfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
-            'file': '853555.mp4',
             'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
             'info_dict': {
+                'id': '853555',
+                'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
             },
         },
-        {
-            'add_ie': ['Vevo'],
-            'url': 'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml',
-            'file': 'USCJY1331283.mp4',
-            'md5': '73b4e7fcadd88929292fe52c3ced8caf',
-            'info_dict': {
-                'title': 'Everything Has Changed',
-                'upload_date': '20130606',
-                'uploader': 'Taylor Swift',
-            },
-            'skip': 'VEVO is only available in some countries',
-        },
     ]
 
     def _get_thumbnail_url(self, uri, itemdoc):
@@ -244,8 +265,8 @@ class MTVIE(MTVServicesInfoExtractor):
             webpage = self._download_webpage(url, video_id)
 
             # Some videos come from Vevo.com
-            m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";',
-                               webpage, re.DOTALL)
+            m_vevo = re.search(
+                r'(?s)isVevoVideo = true;.*?vevoVideoId = "(.*?)";', webpage)
             if m_vevo:
                 vevo_id = m_vevo.group(1)
                 self.to_screen('Vevo video detected: %s' % vevo_id)
@@ -267,3 +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')])