X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/4231113a7be907539052ce59df44e23c54d21d38..6245a4128588146e39f66b371624ee39bdbb7720:/youtube_dl/extractor/mtv.py

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index 1154a35..fedd5f4 100644
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -115,10 +116,17 @@ class MTVServicesInfoExtractor(InfoExtractor):
             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')]
+            for typographic in transcript.findall('./typographic'):
+                sub_src = typographic.get('src')
+                if not sub_src:
+                    continue
+                ext = typographic.get('format')
+                if ext == 'cea-608':
+                    ext = 'scc'
+                subtitles.setdefault(lang, []).append({
+                    'url': compat_str(sub_src),
+                    'ext': ext
+                })
         return subtitles
 
     def _get_video_info(self, itemdoc, use_hls=True):
@@ -342,33 +350,29 @@ class MTVIE(MTVServicesInfoExtractor):
     }]
 
 
-class MTV81IE(InfoExtractor):
-    IE_NAME = 'mtv81'
-    _VALID_URL = r'https?://(?:www\.)?mtv81\.com/videos/(?P<id>[^/?#.]+)'
+class MTVJapanIE(MTVServicesInfoExtractor):
+    IE_NAME = 'mtvjapan'
+    _VALID_URL = r'https?://(?:www\.)?mtvjapan\.com/videos/(?P<id>[0-9a-z]+)'
 
     _TEST = {
-        'url': 'http://www.mtv81.com/videos/artist-to-watch/the-godfather-of-japanese-hip-hop-segment-1/',
-        'md5': '1edbcdf1e7628e414a8c5dcebca3d32b',
+        'url': 'http://www.mtvjapan.com/videos/prayht/fresh-info-cadillac-escalade',
         'info_dict': {
-            'id': '5e14040d-18a4-47c4-a582-43ff602de88e',
+            'id': 'bc01da03-6fe5-4284-8880-f291f4e368f5',
             'ext': 'mp4',
-            'title': 'Unlocking The Truth|July 18, 2016|1|101|Trailer',
-            'description': '"Unlocking the Truth" premieres August 17th at 11/10c.',
-            'timestamp': 1468846800,
-            'upload_date': '20160718',
+            'title': '【Fresh Info】Cadillac ESCALADE Sport Edition',
+        },
+        'params': {
+            'skip_download': True,
         },
     }
+    _GEO_COUNTRIES = ['JP']
+    _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 
-    def _extract_mgid(self, webpage):
-        return self._search_regex(
-            r'getTheVideo\((["\'])(?P<id>mgid:.+?)\1', webpage,
-            'mgid', group='id')
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        mgid = self._extract_mgid(webpage)
-        return self.url_result('http://media.mtvnservices.com/embed/%s' % mgid)
+    def _get_feed_query(self, uri):
+        return {
+            'arcEp': 'mtvjapan.com',
+            'mgid': uri,
+        }
 
 
 class MTVVideoIE(MTVServicesInfoExtractor):
@@ -418,14 +422,14 @@ class MTVVideoIE(MTVServicesInfoExtractor):
 
 class MTVDEIE(MTVServicesInfoExtractor):
     IE_NAME = 'mtv.de'
-    _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:artists|shows|news)/(?:[^/]+/)*(?P<id>\d+)-[^/#?]+/*(?:[#?].*)?$'
+    _VALID_URL = r'https?://(?:www\.)?mtv\.de/(?:musik/videoclips|folgen|news)/(?P<id>[0-9a-z]+)'
     _TESTS = [{
-        'url': 'http://www.mtv.de/artists/10571-cro/videos/61131-traum',
+        'url': 'http://www.mtv.de/musik/videoclips/2gpnv7/Traum',
         'info_dict': {
-            'id': 'music_video-a50bc5f0b3aa4b3190aa',
-            'ext': 'flv',
-            'title': 'MusicVideo_cro-traum',
-            'description': 'Cro - Traum',
+            'id': 'd5d472bc-f5b7-11e5-bffd-a4badb20dab5',
+            'ext': 'mp4',
+            'title': 'Traum',
+            'description': 'Traum',
         },
         'params': {
             # rtmp download
@@ -434,11 +438,12 @@ class MTVDEIE(MTVServicesInfoExtractor):
         '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',
+        'url': 'http://www.mtv.de/folgen/6b1ylu/teen-mom-2-enthuellungen-S5-F1',
         'info_dict': {
-            'id': 'local_playlist-f5ae778b9832cc837189',
-            'ext': 'flv',
-            'title': 'Episode_teen-mom-2_shows_season-5_episode-1_full-episode_part1',
+            'id': '1e5a878b-31c5-11e7-a442-0e40cf2fc285',
+            'ext': 'mp4',
+            'title': 'Teen Mom 2',
+            'description': 'md5:dc65e357ef7e1085ed53e9e9d83146a7',
         },
         'params': {
             # rtmp download
@@ -446,7 +451,7 @@ class MTVDEIE(MTVServicesInfoExtractor):
         },
         'skip': 'Blocked at Travis CI',
     }, {
-        'url': 'http://www.mtv.de/news/77491-mtv-movies-spotlight-pixels-teil-3',
+        'url': 'http://www.mtv.de/news/glolix/77491-mtv-movies-spotlight--pixels--teil-3',
         'info_dict': {
             'id': 'local_playlist-4e760566473c4c8c5344',
             'ext': 'mp4',
@@ -459,25 +464,11 @@ class MTVDEIE(MTVServicesInfoExtractor):
         },
         'skip': 'Das Video kann zur Zeit nicht abgespielt werden.',
     }]
+    _GEO_COUNTRIES = ['DE']
+    _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed'
 
-    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)
+    def _get_feed_query(self, uri):
+        return {
+            'arcEp': 'mtv.de',
+            'mgid': uri,
+        }