[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / movieclips.py

diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py
index 04e17d0551c7a46feff1822c4dc4be38d00cc520..5453da1acfe19a103d97a026b725b81e86d7859d 100644 (file)
--- a/youtube_dl/extractor/movieclips.py
+++ b/youtube_dl/extractor/movieclips.py
@@ -1,80 +1,49 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-)
 from ..utils import (
-    ExtractorError,
-    clean_html,
+    smuggle_url,
+    float_or_none,
+    parse_iso8601,
+    update_url_query,
 )
 
 
 class MovieClipsIE(InfoExtractor):
-    _VALID_URL = r'https?://movieclips\.com/(?P<id>[\da-zA-Z]+)(?:-(?P<display_id>[\da-z-]+))?'
+    _VALID_URL = r'https?://(?:www\.)?movieclips\.com/videos/.+-(?P<id>\d+)(?:\?|$)'
     _TEST = {
-        'url': 'http://movieclips.com/Wy7ZU-my-week-with-marilyn-movie-do-you-love-me/',
+        'url': 'http://www.movieclips.com/videos/warcraft-trailer-1-561180739597',
+        'md5': '42b5a0352d4933a7bd54f2104f481244',
         'info_dict': {
-            'id': 'Wy7ZU',
-            'display_id': 'my-week-with-marilyn-movie-do-you-love-me',
+            'id': 'pKIGmG83AqD9',
             'ext': 'mp4',
-            'title': 'My Week with Marilyn - Do You Love Me?',
-            'description': 'md5:e86795bd332fe3cff461e7c8dc542acb',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            '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': r're:^https?://.*\.jpg$',
+            'timestamp': 1446843055,
+            'upload_date': '20151106',
+            'uploader': 'Movieclips',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        }
+        'add_ie': ['ThePlatform'],
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
-        show_id = display_id or video_id
-
-        config = self._download_xml(
-            'http://config.movieclips.com/player/config/%s' % video_id,
-            show_id, 'Downloading player config')
-
-        if config.find('./country-region').text == 'false':
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, config.find('./region_alert').text), expected=True)
-
-        properties = config.find('./video/properties')
-        smil_file = properties.attrib['smil_file']
-
-        smil = self._download_xml(smil_file, show_id, 'Downloading SMIL')
-        base_url = smil.find('./head/meta').attrib['base']
-
-        formats = []
-        for video in smil.findall('./body/switch/video'):
-            vbr = int(video.attrib['system-bitrate']) / 1000
-            src = video.attrib['src']
-            formats.append({
-                'url': base_url,
-                'play_path': src,
-                'ext': src.split(':')[0],
-                'vbr': vbr,
-                'format_id': '%dk' % vbr,
-            })
-
-        self._sort_formats(formats)
-
-        title = '%s - %s' % (properties.attrib['clip_movie_title'], properties.attrib['clip_title'])
-        description = clean_html(compat_str(properties.attrib['clip_description']))
-        thumbnail = properties.attrib['image']
-        categories = properties.attrib['clip_categories'].split(',')
+        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 {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
-            'formats': formats,
+            '_type': 'url_transparent',
+            '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'),
         }