Merge tag 'upstream/2014.07.11'

[youtubedl] / youtube_dl / extractor / gamespot.py

diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py
index cec3b7ac863247e8ddc2c2add953372bd809eed4..3d67b9d60242760ff3e32c9fbbbcab39542f01da 100644 (file)
--- a/youtube_dl/extractor/gamespot.py
+++ b/youtube_dl/extractor/gamespot.py
@@ -1,45 +1,59 @@
+from __future__ import unicode_literals
+

 import re
 import re

-import xml.etree.ElementTree
+import json

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    unified_strdate,
+    compat_urllib_parse,
+    compat_urlparse,
+    unescapeHTML,
+    get_meta_content,

 )
 
 )
 

+

 class GameSpotIE(InfoExtractor):
 class GameSpotIE(InfoExtractor):

-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/([^/]+)/videos/([^/]+)-([^/d]+)/'
+    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'

     _TEST = {
     _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"5569d64ca98db01f0177c934fe8c1e9b",
-        u"info_dict": {
-            u"title": u"Arma III - Community Guide: SITREP I",
-            u"upload_date": u"20130627", 
+        '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.',

         }
     }
 
         }
     }
 

-

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group(3).split("-")[-1]
-        info_url = "http://www.gamespot.com/pages/video_player/xml.php?id="+str(video_id)
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
+        page_id = mobj.group('page_id')
+        webpage = self._download_webpage(url, page_id)
+        data_video_json = self._search_regex(r'data-video=["\'](.*?)["\']', webpage, 'data video')
+        data_video = json.loads(unescapeHTML(data_video_json))

 
 

-        video_url = clip_el.find('./URI').text
-        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)
+        # Transform the manifest url to a link to the mp4 files
+        # they are used in mobile devices.
+        f4m_url = data_video['videoStreams']['f4m_stream']
+        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)
+        formats = []
+        for q in qualities:
+            formats.append({
+                'url': http_template % q,
+                'ext': 'mp4',
+                'format_id': q,
+            })

 
 

-        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'],
+            'title': compat_urllib_parse.unquote(data_video['title']),
+            'formats': formats,
+            'description': get_meta_content('description', webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }