]> gitweb @ CieloNegro.org - youtube-dl.git/commitdiff
TumblrIE
authorJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Mon, 22 Apr 2013 19:07:49 +0000 (21:07 +0200)
committerJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Mon, 22 Apr 2013 19:27:27 +0000 (21:27 +0200)
I haven't found many videos to test, so it may not work for all.

test/tests.json
youtube_dl/InfoExtractors.py

index 7808a07de24e7326ee7a690fb8d122e99fd6cfa1..2b56738a07c71fb9ab9282ed0d2e9fd8e1db210e 100644 (file)
         "title": "11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden"
     },
     "skip": "Requires rtmpdump"
+  },
+  {
+    "name": "Tumblr",
+    "url": "http://birthdayproject2012.tumblr.com/post/17258355236/a-sample-video-from-leeann-if-you-need-an-idea",
+    "file": "17258355236.mp4",
+    "md5": "7c6a514d691b034ccf8567999e9e88a3",
+    "info_dict": {
+        "title": "A sample video from LeeAnn. (If you need an idea..."
+    }
   }
 
 ]
index ff1f07e9b80a35705f4474a59b39df00d578ae72..6a6545c9b578fba948962542d602bc5352b86840 100755 (executable)
@@ -4327,6 +4327,40 @@ class ARDIE(InfoExtractor):
             info["url"] = stream["video_url"]
         return [info]
 
+class TumblrIE(InfoExtractor):
+    _VALID_URL = r'http://(?P<blog_name>.*?).tumblr.com/((post)|(video))/(?P<id>\d*)/(.*?)'
+
+    def _real_extract(self, url):
+        m_url = re.match(self._VALID_URL, url)
+        video_id = m_url.group('id')
+        blog = m_url.group('blog_name')
+
+        url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
+        webpage = self._download_webpage(url, video_id)
+
+        re_video = r'src=\\x22(?P<video_url>http://%s.tumblr.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id)
+        video = re.search(re_video, webpage)
+        if video is None:
+            self.to_screen("No video founded")
+            return []
+        video_url = video.group('video_url')
+        ext = video.group('ext')
+
+        re_thumb = r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22'  # We pick the first poster
+        thumb = re.search(re_thumb, webpage).group('thumb').replace('\\', '')
+
+        # The only place where you can get a title, it's not complete,
+        # but searching in other places doesn't work for all videos
+        re_title = r'<title>(.*?) - (?P<title>.*?)</title>'
+        title = unescapeHTML(re.search(re_title, webpage).group('title'))
+
+        return [{'id': video_id,
+                 'url': video_url,
+                 'title': title,
+                 'thumbnail': thumb,
+                 'ext': ext
+                 }]
+
 
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
@@ -4381,6 +4415,7 @@ def gen_extractors():
         SpiegelIE(),
         LiveLeakIE(),
         ARDIE(),
+        TumblrIE(),
         GenericIE()
     ]