Merge pull request #7320 from remitamine/adobetv

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

diff --git a/youtube_dl/extractor/gametrailers.py b/youtube_dl/extractor/gametrailers.py
index cd438bd2f4cf425b07301733bc5a62e69b60c4f3..c3f031d9cd4341184cc3b70eea77c1f360a1a3c6 100644 (file)
--- a/youtube_dl/extractor/gametrailers.py
+++ b/youtube_dl/extractor/gametrailers.py
@@ -1,63 +1,62 @@
-import re
-import xml.etree.ElementTree
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse,
-
-    ExtractorError,
+    int_or_none,
+    parse_age_limit,
+    url_basename,
 )
 
+
 class GametrailersIE(InfoExtractor):
-    _VALID_URL = r'http://www.gametrailers.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
+    _VALID_URL = r'http://www\.gametrailers\.com/videos/view/[^/]+/(?P<id>.+)'
+
     _TEST = {
-        u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
-        u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.flv',
-        u'md5': u'c3edbc995ab4081976e16779bd96a878',
-        u'info_dict': {
-            u"title": u"E3 2013: Debut Trailer"
+        'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review',
+        'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a',
+        'info_dict': {
+            'id': '2983958',
+            'ext': 'mp4',
+            'display_id': '116437-Just-Cause-3-Review',
+            'title': 'Just Cause 3 - Review',
+            'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?',
         },
-        u'skip': u'Requires rtmpdump'
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
-                                   r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
-                                  webpage, u'mgid')
-
-        data = compat_urllib_parse.urlencode({'uri': mgid, 'acceptMethods': 'fms'})
-        info_page = self._download_webpage('http://www.gametrailers.com/feeds/mrss?' + data,
-                                           video_id, u'Downloading video info')
-        doc = xml.etree.ElementTree.fromstring(info_page.encode('utf-8'))
-        default_thumb = doc.find('./channel/image/url').text
-
-        media_namespace = {'media': 'http://search.yahoo.com/mrss/'}
-        parts = [{
-            'title': video_doc.find('title').text,
-            'ext': 'flv',
-            'id': video_doc.find('guid').text.rpartition(':')[2],
-            # Videos are actually flv not mp4
-            'url': self._get_video_url(video_doc.find('media:group/media:content', media_namespace).attrib['url'], video_id),
-            # The thumbnail may not be defined, it would be ''
-            'thumbnail': video_doc.find('media:group/media:thumbnail', media_namespace).attrib['url'] or default_thumb,
-            'description': video_doc.find('description').text,
-        } for video_doc in doc.findall('./channel/item')]
-        return parts
-
-    def _get_video_url(self, mediagen_url, video_id):
-        if 'acceptMethods' not in mediagen_url:
-            mediagen_url += '&acceptMethods=fms'
-        links_webpage = self._download_webpage(mediagen_url,
-                                               video_id, u'Downloading video urls info')
-        doc = xml.etree.ElementTree.fromstring(links_webpage)
-        urls = list(doc.iter('src'))
-        if len(urls) == 0:
-            raise ExtractorError(u'Unable to extract video url')
-        # They are sorted from worst to best quality
-        return urls[-1].text
-
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        title = self._html_search_regex(
+            r'<title>(.+?)\|', webpage, 'title').strip()
+        embed_url = self._proto_relative_url(
+            self._search_regex(
+                r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage,
+                'embed url'),
+            scheme='http:')
+        video_id = url_basename(embed_url)
+        embed_page = self._download_webpage(embed_url, video_id)
+        embed_vars_json = self._search_regex(
+            r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page,
+            'embed vars')
+        info = self._parse_json(embed_vars_json, video_id)
+
+        formats = []
+        for media in info['media']:
+            if media['mediaPurpose'] == 'play':
+                formats.append({
+                    'url': media['uri'],
+                    'height': media['height'],
+                    'width:': media['width'],
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': info.get('thumbUri'),
+            'description': self._og_search_description(webpage),
+            'duration': int_or_none(info.get('videoLengthInSeconds')),
+            'age_limit': parse_age_limit(info.get('audienceRating')),
+        }