youtube-dl

commit 7129bed51bc2452524c69874023ee4a0b351a046
parent 1cc79574fc5df21bf35dccf61eac0e9e75ed8d20
Author: Philipp Hagemeister <phihag@phihag.de>
Date:   Sat, 13 Dec 2014 12:35:13 +0100

[keek] Modernize and extract uploader

Diffstat:
M
youtube_dl/extractor/keek.py
 | 
28
++++++++++++++++++----------

1 file changed, 18 insertions(+), 10 deletions(-)
diff --git a/youtube_dl/extractor/keek.py b/youtube_dl/extractor/keek.py
@@ -6,29 +6,36 @@ from .common import InfoExtractor
 
 
 class KeekIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
+    _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<id>\w+)'
     IE_NAME = 'keek'
     _TEST = {
         'url': 'https://www.keek.com/ytdl/keeks/NODfbab',
-        'file': 'NODfbab.mp4',
-        'md5': '9b0636f8c0f7614afa4ea5e4c6e57e83',
+        'md5': '09c5c109067536c1cec8bac8c21fea05',
         'info_dict': {
-            'uploader': 'ytdl',
+            'id': 'NODfbab',
+            'ext': 'mp4',
+            'uploader': 'youtube-dl project',
+            'uploader_id': 'ytdl',
             'title': 'test chars: "\'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de .',
         },
     }
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('videoID')
+        video_id = self._match_id(url)
 
         video_url = 'http://cdn.keek.com/keek/video/%s' % video_id
         thumbnail = 'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
         webpage = self._download_webpage(url, video_id)
 
-        uploader = self._html_search_regex(
-            r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
-            webpage, 'uploader', fatal=False)
+        raw_desc = self._html_search_meta('description', webpage)
+        if raw_desc:
+            uploader = self._html_search_regex(
+                r'Watch (.*?)\s+\(', raw_desc, 'uploader', fatal=False)
+            uploader_id = self._html_search_regex(
+                r'Watch .*?\(@(.+?)\)', raw_desc, 'uploader_id', fatal=False)
+        else:
+            uploader = None
+            uploader_id = None
 
         return {
             'id': video_id,
@@ -36,5 +43,6 @@ class KeekIE(InfoExtractor):
             'ext': 'mp4',
             'title': self._og_search_title(webpage),
             'thumbnail': thumbnail,
-            'uploader': uploader
+            'uploader': uploader,
+            'uploader_id': uploader_id,
         }