Merge tag 'upstream/2016.08.17'

[youtubedl] / youtube_dl / extractor / mtv.py

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index bc7f49ebbac86cda7aa1bb711076b783e24bfea8..2f455680ebba41074513f7dd33b76e5c269cb142 100644 (file)
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -2,20 +2,25 @@ from __future__ import unicode_literals
 
 import re
 
 
 import re
 

-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor

 from ..compat import (
 from ..compat import (

-    compat_urllib_parse,
-    compat_urllib_request,
+    compat_urllib_parse_urlencode,

     compat_str,
     compat_str,

+    compat_xpath,

 )
 from ..utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
 )
 from ..utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,

+    float_or_none,

     HEADRequest,
     HEADRequest,

+    sanitized_Request,
+    strip_or_none,
+    timeconvert,

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

+    xpath_text,

 )
 
 
 )
 
 


@@ -23,21 +28,22 @@ def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
 
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
 

-class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
+class MTVServicesInfoExtractor(InfoExtractor):

     _MOBILE_TEMPLATE = None
     _MOBILE_TEMPLATE = None

+    _LANG = None

 
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
 
     # This was originally implemented for ComedyCentral, but it also works here
 
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
 
     # This was originally implemented for ComedyCentral, but it also works here

-    @staticmethod
-    def _transform_rtmp_url(rtmp_video_url):
+    @classmethod
+    def _transform_rtmp_url(cls, rtmp_video_url):

         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:
         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url)
         if not m:

-            return rtmp_video_url
+            return {'rtmp': rtmp_video_url}

         base = 'http://viacommtvstrmfs.fplive.net/'
         base = 'http://viacommtvstrmfs.fplive.net/'

-        return base + m.group('finalid')
+        return {'http': base + m.group('finalid')}

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


@@ -52,7 +58,7 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
 
     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
         req.add_header('User-Agent', 'curl/7')
         webpage = self._download_webpage(req, mtvn_id,
         # Otherwise we get a webpage that would execute some javascript
         req.add_header('User-Agent', 'curl/7')
         webpage = self._download_webpage(req, mtvn_id,


@@ -66,7 +72,7 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
         return [{'url': url, 'ext': 'mp4'}]
 
     def _extract_video_formats(self, mdoc, mtvn_id):
         return [{'url': url, 'ext': 'mp4'}]
 
     def _extract_video_formats(self, mdoc, mtvn_id):

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


@@ -81,13 +87,14 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
                 rtmp_video_url = rendition.find('./src').text
                 if rtmp_video_url.endswith('siteunavail.png'):
                     continue
                 rtmp_video_url = rendition.find('./src').text
                 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'),
+                new_urls = self._transform_rtmp_url(rtmp_video_url)
+                formats.extend([{
+                    'ext': 'flv' if new_url.startswith('rtmp') else ext,
+                    'url': new_url,
+                    'format_id': '-'.join(filter(None, [kind, rendition.get('bitrate')])),

                     'width': int(rendition.get('width')),
                     'height': int(rendition.get('height')),
                     'width': int(rendition.get('width')),
                     'height': int(rendition.get('height')),

-                })
+                } for kind, new_url in new_urls.items()])

             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)
             except (KeyError, TypeError):
                 raise ExtractorError('Invalid rendition field.')
         self._sort_formats(formats)


@@ -95,44 +102,42 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
 
     def _extract_subtitles(self, mdoc, mtvn_id):
         subtitles = {}
 
     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 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
         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,
                                           'Downloading video urls')
 
 
         mediagen_doc = self._download_xml(mediagen_url, video_id,
                                           '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 = strip_or_none(xpath_text(itemdoc, 'description'))
+
+        timestamp = timeconvert(xpath_text(itemdoc, 'pubDate'))

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


@@ -140,9 +145,9 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
                 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
 


@@ -165,21 +170,35 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
             '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')),
+            'timestamp': timestamp,

         }
 
         }
 

+    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)
         feed_url = self._get_feed_url(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})
+        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(

-            feed_url + '?' + data, video_id,
+            url, video_id,

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

+
+        title = xpath_text(idoc, './channel/title')
+        description = xpath_text(idoc, './channel/description')
+

         return self.playlist_result(
         return self.playlist_result(

-            [self._get_video_info(item) for item in idoc.findall('.//item')])
+            [self._get_video_info(item) for item in idoc.findall('.//item')],
+            playlist_title=title, playlist_description=description)

 
 

-    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}


@@ -193,11 +212,20 @@ class MTVServicesInfoExtractor(SubtitlesInfoExtractor):
         if mgid is None or ':' not in mgid:
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
         if mgid is None or ':' not in mgid:
             mgid = self._search_regex(
                 [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],

-                webpage, '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)
         videos_info = self._get_videos_info(mgid)

-        if self._downloader.params.get('listsubtitles', False):
-            return

         return videos_info
 
 
         return videos_info
 
 


@@ -214,9 +242,18 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor):
             '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.',
             '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.',

+            'timestamp': 1400126400,
+            'upload_date': '20140515',

         },
     }
 
         },
     }
 

+    @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, '')
     def _get_feed_url(self, uri):
         video_id = self._id_from_uri(uri)
         site_id = uri.replace(video_id, '')


@@ -249,6 +286,8 @@ class MTVIE(MTVServicesInfoExtractor):
                 'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',
                 'ext': 'mp4',
                 'title': 'Taylor Swift - "Ours (VH1 Storytellers)"',
                 'description': 'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.',

+                'timestamp': 1352610000,
+                'upload_date': '20121111',

             },
         },
     ]
             },
         },
     ]


@@ -275,15 +314,68 @@ class MTVIE(MTVServicesInfoExtractor):
         return self._get_videos_info(uri)
 
 
         return self._get_videos_info(uri)
 
 

-class MTVIggyIE(MTVServicesInfoExtractor):
-    IE_NAME = 'mtviggy.com'
-    _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
-    _TEST = {
-        'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
+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': {
         'info_dict': {

-            'id': '984696',
+            'id': 'music_video-a50bc5f0b3aa4b3190aa',
+            'ext': 'flv',
+            'title': 'MusicVideo_cro-traum',
+            'description': 'Cro - Traum',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+        'skip': 'Blocked at Travis CI',
+    }, {
+        # 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': 'flv',
+            'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+        'skip': 'Blocked at Travis CI',
+    }, {
+        'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
+        'info_dict': {
+            'id': 'local_playlist-4e760566473c4c8c5344',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
-        }
-    }
-    _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'
+            'title': 'Article_mtv-movies-spotlight-pixels-teil-3_short-clips_part1',
+            'description': 'MTV Movies Supercut',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+        'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
+    }]
+
+    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)
+
+        def _mrss_url(item):
+            return item['mrss'] + item.get('mrssvars', '')
+
+        # news pages contain single video in playlist with different id
+        if len(playlist) == 1:
+            return self._get_videos_info_from_url(_mrss_url(playlist[0]), 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(_mrss_url(item), video_id)