Update changelog.

[youtubedl] / youtube_dl / extractor / mtv.py

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index e5ca41b4091698ad2a180f9d2ca00b4b96218c1e..dd06395891ac02ce81b40efd471776cb080d328d 100644 (file)
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -3,16 +3,22 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..compat import (
+    compat_urllib_parse_urlencode,
+    compat_str,
+    compat_xpath,
+)

 from ..utils import (
 from ..utils import (

-    compat_urllib_parse,
-    compat_urllib_request,

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

+    float_or_none,

     HEADRequest,
     HEADRequest,

+    sanitized_Request,

     unescapeHTML,
     url_basename,
     RegexNotFoundError,
     unescapeHTML,
     url_basename,
     RegexNotFoundError,

+    xpath_text,

 )
 
 
 )
 
 


@@ -22,6 +28,8 @@ def _media_xml_tag(tag):
 
 class MTVServicesInfoExtractor(InfoExtractor):
     _MOBILE_TEMPLATE = None
 
 class MTVServicesInfoExtractor(InfoExtractor):
     _MOBILE_TEMPLATE = None

+    _LANG = None
+

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


@@ -32,9 +40,12 @@ class MTVServicesInfoExtractor(InfoExtractor):
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
             return rtmp_video_url
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
             return rtmp_video_url

-        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
+        base = 'http://viacommtvstrmfs.fplive.net/'

         return base + m.group('finalid')
 
         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)


@@ -45,62 +56,84 @@ class MTVServicesInfoExtractor(InfoExtractor):
 
     def _extract_mobile_video_formats(self, mtvn_id):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id
 
     def _extract_mobile_video_formats(self, mtvn_id):
         webpage_url = self._MOBILE_TEMPLATE % mtvn_id

-        req = compat_urllib_request.Request(webpage_url)
+        req = sanitized_Request(webpage_url)

         # Otherwise we get a webpage that would execute some javascript
         # 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,
         webpage = self._download_webpage(req, mtvn_id,

-            'Downloading mobile page')
+                                         '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)
         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'}]
+        return [{'url': url, 'ext': 'mp4'}]

 
     def _extract_video_formats(self, mdoc, mtvn_id):
 
     def _extract_video_formats(self, mdoc, mtvn_id):

-        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4|copyright_error\.flv(?:\?geo\b.+?)?)$', 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 '
             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')
+                               'country, trying with the mobile version')

                 return self._extract_mobile_video_formats(mtvn_id)
             raise ExtractorError('This video is not available from your country.',
                 return self._extract_mobile_video_formats(mtvn_id)
             raise ExtractorError('This video is not available from your country.',

-                expected=True)
+                                 expected=True)

 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
             try:
                 _, _, ext = rendition.attrib['type'].partition('/')
                 rtmp_video_url = rendition.find('./src').text
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
             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
+                new_url = self._transform_rtmp_url(rtmp_video_url)
+                formats.append({
+                    'ext': 'flv' if new_url.startswith('rtmp') else ext,
+                    'url': new_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
 
             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)
         self.report_extraction(video_id)
     def _get_video_info(self, itemdoc):
         uri = itemdoc.find('guid').text
         video_id = self._id_from_uri(uri)
         self.report_extraction(video_id)

-        mediagen_url = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content'))).attrib['url']
+        content_el = itemdoc.find('%s/%s' % (_media_xml_tag('group'), _media_xml_tag('content')))
+        mediagen_url = content_el.attrib['url']

         # Remove the templates, like &device={device}
         mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
         # Remove the templates, like &device={device}
         mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:

-            mediagen_url += '&acceptMethods=fms'
+            mediagen_url += '&' if '?' in mediagen_url else '?'
+            mediagen_url += 'acceptMethods=fms'

 
         mediagen_doc = self._download_xml(mediagen_url, video_id,
 
         mediagen_doc = self._download_xml(mediagen_url, video_id,

-            'Downloading video urls')
+                                          'Downloading video urls')

 
 

-        description_node = itemdoc.find('description')
-        if description_node is not None:
-            description = description_node.text.strip()
-        else:
-            description = None
+        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 = xpath_text(itemdoc, 'description')

 
         title_el = None
         if title_el is None:
 
         title_el = None
         if title_el is None:


@@ -108,9 +141,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 itemdoc, './/{http://search.yahoo.com/mrss/}category',
                 'scheme', 'urn:mtvn:video_title')
         if title_el is None:
                 itemdoc, './/{http://search.yahoo.com/mrss/}category',
                 'scheme', 'urn:mtvn:video_title')
         if title_el is None:

-            title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+            title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))

         if title_el is None:
         if title_el is None:

-            title_el = itemdoc.find('.//title')
+            title_el = itemdoc.find(compat_xpath('.//title'))

             if title_el.text is None:
                 title_el = None
 
             if title_el.text is None:
                 title_el = None
 


@@ -122,30 +155,40 @@ class MTVServicesInfoExtractor(InfoExtractor):
         # 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',
         # 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')
+                                       'scheme', 'urn:mtvn:id')

         if mtvn_id_node is not None:
             mtvn_id = mtvn_id_node.text
 
         return {
             'title': title,
             'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
         if mtvn_id_node is not None:
             mtvn_id = mtvn_id_node.text
 
         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,
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,

+            'duration': float_or_none(content_el.attrib.get('duration')),

         }
 
         }
 

+    def _get_feed_query(self, uri):
+        data = {'uri': uri}
+        if self._LANG:
+            data['lang'] = self._LANG
+        return compat_urllib_parse_urlencode(data)
+

     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)

-        data = compat_urllib_parse.urlencode({'uri': uri})
+        feed_url = self._get_feed_url(uri)
+        info_url = feed_url + '?' + self._get_feed_query(uri)
+        return self._get_videos_info_from_url(info_url, video_id)

 
 

+    def _get_videos_info_from_url(self, url, video_id):

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

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

             'Downloading info', transform_source=fix_xml_ampersands)
             'Downloading info', transform_source=fix_xml_ampersands)

-        return [self._get_video_info(item) for item in idoc.findall('.//item')]
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')])

 
 

-    def _real_extract(self, url):
-        title = url_basename(url)
-        webpage = self._download_webpage(url, title)
+    def _extract_mgid(self, webpage):

         try:
             # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
             # or http://media.mtvnservices.com/{mgid}
         try:
             # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
             # or http://media.mtvnservices.com/{mgid}


@@ -154,9 +197,64 @@ 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:.*?)"'],
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],

-                webpage, u'mgid')
+                webpage, 'mgid', default=None)
+
+        if not mgid:
+            sm4_embed = self._html_search_meta(
+                'sm4:video:embed', webpage, 'sm4 embed', default='')
+            mgid = self._search_regex(
+                r'embed/(mgid:.+?)["\'&?/]', sm4_embed, 'mgid')
+        return mgid
+
+    def _real_extract(self, url):
+        title = url_basename(url)
+        webpage = self._download_webpage(url, title)
+        mgid = self._extract_mgid(webpage)
+        videos_info = self._get_videos_info(mgid)
+        return videos_info
+
+
+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.',
+        },
+    }
+
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage)
+        if mobj:
+            return mobj.group('url')
+
+    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}/'
+                      'context4/context5/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)
 
 
         return self._get_videos_info(mgid)
 
 


@@ -170,25 +268,14 @@ class MTVIE(MTVServicesInfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',
     _TESTS = [
         {
             'url': 'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml',

-            'file': '853555.mp4',

             'md5': '850f3f143316b1e71fa56a4edfd6e0f8',
             'info_dict': {
             '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.',
             },
         },
                 '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):
     ]
 
     def _get_thumbnail_url(self, uri, itemdoc):


@@ -200,15 +287,15 @@ class MTVIE(MTVServicesInfoExtractor):
         uri = mobj.groupdict().get('mgid')
         if uri is None:
             webpage = self._download_webpage(url, video_id)
         uri = mobj.groupdict().get('mgid')
         if uri is None:
             webpage = self._download_webpage(url, video_id)

-    
+

             # Some videos come from Vevo.com
             # 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:
             if m_vevo:

-                vevo_id = m_vevo.group(1);
+                vevo_id = m_vevo.group(1)

                 self.to_screen('Vevo video detected: %s' % vevo_id)
                 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')
                 self.to_screen('Vevo video detected: %s' % vevo_id)
                 return self.url_result('vevo:%s' % vevo_id, ie='Vevo')

-    
+

             uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
         return self._get_videos_info(uri)
 
             uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
         return self._get_videos_info(uri)
 


@@ -225,3 +312,65 @@ class MTVIggyIE(MTVServicesInfoExtractor):
         }
     }
     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
         }
     }
     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'

+
+
+class MTVDEIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtv.de'
+    _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
+    _TESTS = [{
+        'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
+        'info_dict': {
+            'id': 'music_video-a50bc5f0b3aa4b3190aa',
+            'ext': 'mp4',
+            'title': 'MusicVideo_cro-traum',
+            'description': 'Cro - Traum',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # mediagen URL without query (e.g. http://videos.mtvnn.com/mediagen/e865da714c166d18d6f80893195fcb97)
+        'url': 'http://www.mtv.de/shows/933-teen-mom-2/staffeln/5353/folgen/63565-enthullungen',
+        'info_dict': {
+            'id': 'local_playlist-f5ae778b9832cc837189',
+            'ext': 'mp4',
+            'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        # single video in pagePlaylist with different id
+        'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
+        'info_dict': {
+            'id': 'local_playlist-4e760566473c4c8c5344',
+            'ext': 'mp4',
+            'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
+            'description': 'MTV Movies Supercut',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        playlist = self._parse_json(
+            self._search_regex(
+                r'window\.pagePlaylist\s*=\s*(\[.+?\]);\n', webpage, 'page playlist'),
+            video_id)
+
+        # news pages contain single video in playlist with different id
+        if len(playlist) == 1:
+            return self._get_videos_info_from_url(playlist[0]['mrss'], video_id)
+
+        for item in playlist:
+            item_id = item.get('id')
+            if item_id and compat_str(item_id) == video_id:
+                return self._get_videos_info_from_url(item['mrss'], video_id)