]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/gamespot.py
debian/changelog: Annotate bugs to close.
[youtubedl] / youtube_dl / extractor / gamespot.py
index 47373e21540030d4c9a19dbfc1c5943f468fea4f..2d33fa7f50d203a4522229d319467595881b138a 100644 (file)
@@ -14,8 +14,8 @@ from ..utils import (
 
 
 class GameSpotIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
-    _TEST = {
+    _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': {
@@ -23,8 +23,16 @@ class GameSpotIE(InfoExtractor):
             '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):
         page_id = self._match_id(url)
@@ -32,25 +40,37 @@ class GameSpotIE(InfoExtractor):
         data_video_json = self._search_regex(
             r'data-video=["\'](.*?)["\']', webpage, 'data video')
         data_video = json.loads(unescapeHTML(data_video_json))
+        streams = data_video['videoStreams']
 
-        # 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,
-            })
+        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': data_video['guid'],