Register new version in the changelog.

[youtubedl] / youtube_dl / extractor / gamespot.py

diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py
index 7585b70618d1e4f92e8297fbf4d1397359a5224b..b3f1bafcc37ee98f1c5b89a644909f3ee0a32049 100644 (file)
--- a/youtube_dl/extractor/gamespot.py
+++ b/youtube_dl/extractor/gamespot.py
@@ -1,55 +1,82 @@
+from __future__ import unicode_literals
+

 import re
 import re

-import xml.etree.ElementTree
+import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from ..compat import (
+    compat_urllib_parse_unquote,
+    compat_urlparse,
+)

 from ..utils import (
 from ..utils import (

-    unified_strdate,
-    compat_urllib_parse,
+    unescapeHTML,

 )
 
 )
 

-class GameSpotIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
-    _TEST = {
-        u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"6410818.mp4",
-        u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
-        u"info_dict": {
-            u"title": u"Arma III - Community Guide: SITREP I",
-            u"upload_date": u"20130627", 
-        }
-    }

 
 

+class GameSpotIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
+    _TESTS = [{
+        'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
+        'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
+        'info_dict': {
+            'id': 'gs-2300-6410818',
+            'ext': 'mp4',
+            'title': 'Arma 3 - Community Guide: SITREP I',
+            'description': 'Check out this video where some of the basics of Arma 3 is explained.',
+        },
+    }, {
+        'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/',
+        'info_dict': {
+            'id': 'gs-2300-6424837',
+            'ext': 'flv',
+            'title': 'The Witcher 3: Wild Hunt [Xbox ONE]  - Now Playing',
+            'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.',
+        },
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        page_id = mobj.group('page_id')
+        page_id = self._match_id(url)

         webpage = self._download_webpage(url, page_id)
         webpage = self._download_webpage(url, page_id)

-        video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
-                                            r'http://www\.gamespot\.com/videoembed/(\d+)'],
-                                           webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
-        info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
+        data_video_json = self._search_regex(
+            r'data-video=["\'](.*?)["\']', webpage, 'data video')
+        data_video = json.loads(unescapeHTML(data_video_json))
+        streams = data_video['videoStreams']

 
 

-        http_urls = [{'url': node.find('filePath').text,
-                      'rate': int(node.find('rate').text)}
-            for node in clip_el.find('./httpURI')]
-        best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
-        video_url = best_quality['url']
-        title = clip_el.find('./title').text
-        ext = video_url.rpartition('.')[2]
-        thumbnail_url = clip_el.find('./screenGrabURI').text
-        view_count = int(clip_el.find('./views').text)
-        upload_date = unified_strdate(clip_el.find('./postDate').text)
+        formats = []
+        f4m_url = streams.get('f4m_stream')
+        if f4m_url is not None:
+            # Transform the manifest url to a link to the mp4 files
+            # they are used in mobile devices.
+            f4m_path = compat_urlparse.urlparse(f4m_url).path
+            QUALITIES_RE = r'((,\d+)+,?)'
+            qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
+            http_path = f4m_path[1:].split('/', 1)[1]
+            http_template = re.sub(QUALITIES_RE, r'%s', http_path)
+            http_template = http_template.replace('.csmil/manifest.f4m', '')
+            http_template = compat_urlparse.urljoin(
+                'http://video.gamespotcdn.com/', http_template)
+            for q in qualities:
+                formats.append({
+                    'url': http_template % q,
+                    'ext': 'mp4',
+                    'format_id': q,
+                })
+        else:
+            for quality in ['sd', 'hd']:
+                # It's actually a link to a flv file
+                flv_url = streams.get('f4m_{0}'.format(quality))
+                if flv_url is not None:
+                    formats.append({
+                        'url': flv_url,
+                        'ext': 'flv',
+                        'format_id': quality,
+                    })

 
 

-        return [{
-            'id'          : video_id,
-            'url'         : video_url,
-            'ext'         : ext,
-            'title'       : title,
-            'thumbnail'   : thumbnail_url,
-            'upload_date' : upload_date,
-            'view_count'  : view_count,
-        }]
+        return {
+            'id': data_video['guid'],
+            'display_id': page_id,
+            'title': compat_urllib_parse_unquote(data_video['title']),
+            'formats': formats,
+            'description': self._html_search_meta('description', webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }