[movieclips] fix extraction

1 files changed, 26 insertions, 17 deletions

diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py
index 1564cb71f..d0cb8278e 100644
--- a/youtube_dl/extractor/movieclips.py
+++ b/youtube_dl/extractor/movieclips.py
@@ -2,39 +2,48 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import sanitized_Request
+from ..utils import (
+    smuggle_url,
+    float_or_none,
+    parse_iso8601,
+    update_url_query,
+)
 
 
 class MovieClipsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www.)?movieclips\.com/videos/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www.)?movieclips\.com/videos/.+-(?P<id>\d+)(?:\?|$)'
     _TEST = {
-        'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597?autoPlay=true&playlistId=5',
+        'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597',
+        'md5': '42b5a0352d4933a7bd54f2104f481244',
         'info_dict': {
             'id': 'pKIGmG83AqD9',
-            'display_id': 'warcraft-trailer-1-561180739597',
             'ext': 'mp4',
             'title': 'Warcraft Trailer 1',
             'description': 'Watch Trailer 1 from Warcraft (2016). Legendary’s WARCRAFT is a 3D epic adventure of world-colliding conflict based.',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'timestamp': 1446843055,
+            'upload_date': '20151106',
+            'uploader': 'Movieclips',
         },
         'add_ie': ['ThePlatform'],
     }
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-
-        req = sanitized_Request(url)
-        # it doesn't work if it thinks the browser it's too old
-        req.add_header('User-Agent', 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/43.0 (Chrome)')
-        webpage = self._download_webpage(req, display_id)
-        theplatform_link = self._html_search_regex(r'src="(http://player.theplatform.com/p/.*?)"', webpage, 'theplatform link')
-        title = self._html_search_regex(r'<title[^>]*>([^>]+)-\s*\d+\s*|\s*Movieclips.com</title>', webpage, 'title')
-        description = self._html_search_meta('description', webpage)
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        video = next(v for v in self._parse_json(self._search_regex(
+            r'var\s+__REACT_ENGINE__\s*=\s*({.+});',
+            webpage, 'react engine'), video_id)['playlist']['videos'] if v['id'] == video_id)
 
         return {
             '_type': 'url_transparent',
-            'url': theplatform_link,
-            'title': title,
-            'display_id': display_id,
-            'description': description,
+            'ie_key': 'ThePlatform',
+            'url': smuggle_url(update_url_query(
+                video['contentUrl'], {'mbr': 'true'}), {'force_smil_url': True}),
+            'title': self._og_search_title(webpage),
+            'description': self._html_search_meta('description', webpage),
+            'duration': float_or_none(video.get('duration')),
+            'timestamp': parse_iso8601(video.get('dateCreated')),
+            'thumbnail': video.get('defaultImage'),
+            'uploader': video.get('provider'),
         }