Initiate release.

[youtubedl] / youtube_dl / extractor / mtv.py

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index 5447b6c0cab098b895eda0e9f2b3b266fb65a7b0..228b42d2b940d8eadd0fa3d5e61d0836fd19b7b7 100644 (file)
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -5,9 +5,12 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,

+    compat_urllib_request,

     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,

+    HEADRequest,
+    unescapeHTML,

     url_basename,
     RegexNotFoundError,
 )
     url_basename,
     RegexNotFoundError,
 )


@@ -18,6 +21,8 @@ def _media_xml_tag(tag):
 
 
 class MTVServicesInfoExtractor(InfoExtractor):
 
 
 class MTVServicesInfoExtractor(InfoExtractor):

+    _MOBILE_TEMPLATE = None
+

     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]


@@ -31,6 +36,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 

+    def _get_feed_url(self, uri):
+        return self._FEED_URL
+

     def _get_thumbnail_url(self, uri, itemdoc):
         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
         thumb_node = itemdoc.find(search_path)
     def _get_thumbnail_url(self, uri, itemdoc):
         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
         thumb_node = itemdoc.find(search_path)


@@ -39,9 +47,29 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             return thumb_node.attrib['url']
 
         else:
             return thumb_node.attrib['url']
 

-    def _extract_video_formats(self, mdoc):
-        if re.match(r'.*/error_country_block\.swf$', mdoc.find('.//src').text) is not None:
-            raise ExtractorError('This video is not available from your country.', expected=True)
+    def _extract_mobile_video_formats(self, mtvn_id):
+        webpage_url = self._MOBILE_TEMPLATE % mtvn_id
+        req = compat_urllib_request.Request(webpage_url)
+        # Otherwise we get a webpage that would execute some javascript
+        req.add_header('Youtubedl-user-agent', 'curl/7')
+        webpage = self._download_webpage(req, mtvn_id,
+            'Downloading mobile page')
+        metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
+        req = HEADRequest(metrics_url)
+        response = self._request_webpage(req, mtvn_id, 'Resolving url')
+        url = response.geturl()
+        # Transform the url to get the best quality:
+        url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
+        return [{'url': url,'ext': 'mp4'}]
+
+    def _extract_video_formats(self, mdoc, mtvn_id):
+        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+            if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
+                self.to_screen('The normal version is not available from your '
+                    'country, trying with the mobile version')
+                return self._extract_mobile_video_formats(mtvn_id)
+            raise ExtractorError('This video is not available from your country.',
+                expected=True)

 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):


@@ -56,6 +84,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
                                 })
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
                                 })
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')

+        self._sort_formats(formats)

         return formats
 
     def _get_video_info(self, itemdoc):
         return formats
 
     def _get_video_info(self, itemdoc):


@@ -94,9 +123,16 @@ class MTVServicesInfoExtractor(InfoExtractor):
             raise ExtractorError('Could not find video title')
         title = title.strip()
 
             raise ExtractorError('Could not find video title')
         title = title.strip()
 

+        # This a short id that's used in the webpage urls
+        mtvn_id = None
+        mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
+                'scheme', 'urn:mtvn:id')
+        if mtvn_id_node is not None:
+            mtvn_id = mtvn_id_node.text
+

         return {
             'title': title,
         return {
             'title': title,

-            'formats': self._extract_video_formats(mediagen_doc),
+            'formats': self._extract_video_formats(mediagen_doc, mtvn_id),

             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,


@@ -104,10 +140,10 @@ class MTVServicesInfoExtractor(InfoExtractor):
 
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)
 
     def _get_videos_info(self, uri):
         video_id = self._id_from_uri(uri)

+        feed_url = self._get_feed_url(uri)

         data = compat_urllib_parse.urlencode({'uri': uri})
         data = compat_urllib_parse.urlencode({'uri': uri})

-

         idoc = self._download_xml(
         idoc = self._download_xml(

-            self._FEED_URL + '?' + data, video_id,
+            feed_url + '?' + data, video_id,

             'Downloading info', transform_source=fix_xml_ampersands)
         return [self._get_video_info(item) for item in idoc.findall('.//item')]
 
             'Downloading info', transform_source=fix_xml_ampersands)
         return [self._get_video_info(item) for item in idoc.findall('.//item')]
 


@@ -122,12 +158,46 @@ class MTVServicesInfoExtractor(InfoExtractor):
             if mgid.endswith('.swf'):
                 mgid = mgid[:-4]
         except RegexNotFoundError:
             if mgid.endswith('.swf'):
                 mgid = mgid[:-4]
         except RegexNotFoundError:

+            mgid = None
+
+        if mgid is None or ':' not in mgid:

             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, u'mgid')
         return self._get_videos_info(mgid)
 
 
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
                 webpage, u'mgid')
         return self._get_videos_info(mgid)
 
 

+class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtvservices:embedded'
+    _VALID_URL = r'https?://media\.mtvnservices\.com/embed/(?P<mgid>.+?)(\?|/|$)'
+
+    _TEST = {
+        # From http://www.thewrap.com/peter-dinklage-sums-up-game-of-thrones-in-45-seconds-video/
+        'url': 'http://media.mtvnservices.com/embed/mgid:uma:video:mtv.com:1043906/cp~vid%3D1043906%26uri%3Dmgid%3Auma%3Avideo%3Amtv.com%3A1043906',
+        'md5': 'cb349b21a7897164cede95bd7bf3fbb9',
+        'info_dict': {
+            'id': '1043906',
+            'ext': 'mp4',
+            'title': 'Peter Dinklage Sums Up \'Game Of Thrones\' In 45 Seconds',
+            'description': '"Sexy sexy sexy, stabby stabby stabby, beautiful language," says Peter Dinklage as he tries summarizing "Game of Thrones" in under a minute.',
+        },
+    }
+
+    def _get_feed_url(self, uri):
+        video_id = self._id_from_uri(uri)
+        site_id = uri.replace(video_id, '')
+        config_url = 'http://media.mtvnservices.com/pmt/e1/players/{0}/config.xml'.format(site_id)
+        config_doc = self._download_xml(config_url, video_id)
+        feed_node = config_doc.find('.//feed')
+        feed_url = feed_node.text.strip().split('?')[0]
+        return feed_url
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        mgid = mobj.group('mgid')
+        return self._get_videos_info(mgid)
+
+

 class MTVIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)^https?://
         (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|
 class MTVIE(MTVServicesInfoExtractor):
     _VALID_URL = r'''(?x)^https?://
         (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$|