Merge pull request #8130 from dyn888/master
[youtube-dl] / youtube_dl / extractor / letv.py
index c096cb1ab97d258e1cc5b9e8c2b649553c2217cb..08bdae8a2182be28d2f7c5b3bc6d3fdcb66e1a49 100644 (file)
@@ -19,6 +19,7 @@ from ..utils import (
     int_or_none,
     str_or_none,
     encode_data_uri,
+    url_basename,
 )
 
 
@@ -245,7 +246,7 @@ class LetvPlaylistIE(LetvTvIE):
 
 class LetvCloudIE(InfoExtractor):
     IE_DESC = '乐视云'
-    _VALID_URL = r'http://yuntv\.letv\.com/bcloud.html\?.*$'
+    _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
 
     _TESTS = [{
         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
@@ -253,46 +254,68 @@ class LetvCloudIE(InfoExtractor):
         'info_dict': {
             'id': 'p7jnfw5hw9_467623dedf',
             'ext': 'mp4',
-            'title': 'p7jnfw5hw9_467623dedf',
+            'title': 'Video p7jnfw5hw9_467623dedf',
         },
     }, {
         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
         'info_dict': {
             'id': 'p7jnfw5hw9_ec93197892',
             'ext': 'mp4',
-            'title': 'p7jnfw5hw9_ec93197892',
+            'title': 'Video p7jnfw5hw9_ec93197892',
         },
     }, {
         'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
         'info_dict': {
             'id': 'p7jnfw5hw9_187060b6fd',
             'ext': 'mp4',
-            'title': 'p7jnfw5hw9_187060b6fd',
+            'title': 'Video p7jnfw5hw9_187060b6fd',
         },
     }]
 
     def _real_extract(self, url):
-        uu = re.search('uu=([\w]+)', url).group(1)
-        vu = re.search('vu=([\w]+)', url).group(1)
+        uu_mobj = re.search('uu=([\w]+)', url)
+        vu_mobj = re.search('vu=([\w]+)', url)
+
+        if not uu_mobj or not vu_mobj:
+            raise ExtractorError('Invalid URL: %s' % url, expected=True)
+
+        uu = uu_mobj.group(1)
+        vu = vu_mobj.group(1)
         media_id = uu + '_' + vu
 
         play_json_req = sanitized_Request(
             'http://api.letvcloud.com/gpc.php?cf=html5&sign=signxxxxx&ver=2.2&format=json&' +
-            "uu=" + uu + "&vu=" + vu)
+            'uu=' + uu + '&vu=' + vu)
         play_json = self._download_json(play_json_req, media_id, 'Downloading playJson data')
 
-        formats = [{
-            'url': base64.b64decode(media['play_url']['main_url'].encode('utf-8')).decode("utf-8"),
-            'ext': 'mp4',
-            'format_id': int_or_none(media.get('play_url', {}).get('vtype')),
-            'format_note': str_or_none(media.get('play_url', {}).get('definition')),
-            'width': int_or_none(media.get('play_url', {}).get('vwidth')),
-            'height': int_or_none(media.get('play_url', {}).get('vheight')),
-        } for media in play_json['data']['video_info']['media'].values()]
+        if not play_json.get('data'):
+            if play_json.get('message'):
+                raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
+            elif play_json.get('code'):
+                raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
+            else:
+                raise ExtractorError('Letv cloud returned an unknwon error')
+
+        def b64decode(s):
+            return base64.b64decode(s.encode('utf-8')).decode('utf-8')
+
+        formats = []
+        for media in play_json['data']['video_info']['media'].values():
+            play_url = media['play_url']
+            url = b64decode(play_url['main_url'])
+            decoded_url = b64decode(url_basename(url))
+            formats.append({
+                'url': url,
+                'ext': determine_ext(decoded_url),
+                'format_id': int_or_none(play_url.get('vtype')),
+                'format_note': str_or_none(play_url.get('definition')),
+                'width': int_or_none(play_url.get('vwidth')),
+                'height': int_or_none(play_url.get('vheight')),
+            })
         self._sort_formats(formats)
 
         return {
             'id': media_id,
-            'title': media_id,
+            'title': 'Video %s' % media_id,
             'formats': formats,
         }