From: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date: Sun, 8 Nov 2015 10:49:51 +0000 (+0100)
Subject: [movieclips] Fix extraction (fixes #7404)
X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=commitdiff_plain;h=d5c181a14e08198e400932d591b47683a630c8c1;p=youtube-dl

[movieclips] Fix extraction (fixes #7404)

They use theplatform now.
Changed the test, because the old one seems to be georestricted.
---

diff --git a/youtube_dl/extractor/movieclips.py b/youtube_dl/extractor/movieclips.py
index 04e17d055..e06828b55 100644
--- a/youtube_dl/extractor/movieclips.py
+++ b/youtube_dl/extractor/movieclips.py
@@ -1,80 +1,41 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..compat import (
-    compat_str,
-)
-from ..utils import (
-    ExtractorError,
-    clean_html,
+    compat_urllib_request,
 )
 
 
 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>[^/?#]+)'
     _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?autoPlay=true&playlistId=5',
         'info_dict': {
-            'id': 'Wy7ZU',
-            'display_id': 'my-week-with-marilyn-movie-do-you-love-me',
+            'id': 'pKIGmG83AqD9',
+            'display_id': 'warcraft-trailer-1-561180739597',
             'ext': 'mp4',
-            'title': 'My Week with Marilyn - Do You Love Me?',
-            'description': 'md5:e86795bd332fe3cff461e7c8dc542acb',
+            '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$',
         },
-        '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']
+        display_id = self._match_id(url)
 
-        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(',')
+        req = compat_urllib_request.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)
 
         return {
-            'id': video_id,
-            'display_id': display_id,
+            '_type': 'url_transparent',
+            'url': theplatform_link,
             'title': title,
+            'display_id': display_id,
             'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
-            'formats': formats,
         }