X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/feb5020b37d7d3ba4005a8bac6f4efece4ce4b8c..d317bf370d3cbd2231f3d6b1970e0a86fddc4e5b:/youtube_dl/extractor/tumblr.py

diff --git a/youtube_dl/extractor/tumblr.py b/youtube_dl/extractor/tumblr.py
index ad5840c..3d3b635 100644
--- a/youtube_dl/extractor/tumblr.py
+++ b/youtube_dl/extractor/tumblr.py
@@ -1,21 +1,56 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-)
 
 
 class TumblrIE(InfoExtractor):
-    _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)'
-    _TEST = {
-        u'url': u'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
-        u'file': u'54196191430.mp4',
-        u'md5': u'479bb068e5b16462f5176a6828829767',
-        u'info_dict': {
-            u"title": u"tatiana maslany news"
+    _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/(?:post|video)/(?P<id>[0-9]+)(?:$|[/?#])'
+    _TESTS = [{
+        'url': 'http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes',
+        'md5': '479bb068e5b16462f5176a6828829767',
+        'info_dict': {
+            'id': '54196191430',
+            'ext': 'mp4',
+            'title': 'tatiana maslany news, Orphan Black || DVD extra - behind the scenes ↳...',
+            'description': 'md5:37db8211e40b50c7c44e95da14f630b7',
+            'thumbnail': 're:http://.*\.jpg',
+        }
+    }, {
+        'url': 'http://5sostrum.tumblr.com/post/90208453769/yall-forgetting-the-greatest-keek-of-them-all',
+        'md5': 'bf348ef8c0ef84fbf1cbd6fa6e000359',
+        'info_dict': {
+            'id': '90208453769',
+            'ext': 'mp4',
+            'title': '5SOS STRUM ;]',
+            'description': 'md5:dba62ac8639482759c8eb10ce474586a',
+            'thumbnail': 're:http://.*\.jpg',
         }
-    }
+    }, {
+        'url': 'http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching',
+        'md5': 'de07e5211d60d4f3a2c3df757ea9f6ab',
+        'info_dict': {
+            'id': 'Wmur',
+            'ext': 'mp4',
+            'title': 'naked smoking & stretching',
+            'upload_date': '20150506',
+            'timestamp': 1430931613,
+        },
+        'add_ie': ['Vidme'],
+    }, {
+        'url': 'http://camdamage.tumblr.com/post/98846056295/',
+        'md5': 'a9e0c8371ea1ca306d6554e3fecf50b6',
+        'info_dict': {
+            'id': '105463834',
+            'ext': 'mp4',
+            'title': 'Cam Damage-HD 720p',
+            'uploader': 'John Moyer',
+            'uploader_id': 'user32021558',
+        },
+        'add_ie': ['Vimeo'],
+    }]
 
     def _real_extract(self, url):
         m_url = re.match(self._VALID_URL, url)
@@ -23,27 +58,30 @@ class TumblrIE(InfoExtractor):
         blog = m_url.group('blog_name')
 
         url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id)
-        webpage = self._download_webpage(url, video_id)
+        webpage, urlh = self._download_webpage_handle(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:
-           raise ExtractorError(u'Unable to extract video')
-        video_url = video.group('video_url')
-        ext = video.group('ext')
+        iframe_url = self._search_regex(
+            r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'',
+            webpage, 'iframe url', default=None)
+        if iframe_url is None:
+            return self.url_result(urlh.geturl(), 'Generic')
 
-        video_thumbnail = self._search_regex(r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22',
-            webpage, u'thumbnail', fatal=False)  # We pick the first poster
-        if video_thumbnail: video_thumbnail = video_thumbnail.replace('\\', '')
+        iframe = self._download_webpage(iframe_url, video_id,
+                                        'Downloading iframe page')
+        video_url = self._search_regex(r'<source src="([^"]+)"',
+                                       iframe, 'video url')
 
         # The only place where you can get a title, it's not complete,
         # but searching in other places doesn't work for all videos
-        video_title = self._html_search_regex(r'<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
-            webpage, u'title', flags=re.DOTALL)
-
-        return [{'id': video_id,
-                 'url': video_url,
-                 'title': video_title,
-                 'thumbnail': video_thumbnail,
-                 'ext': ext
-                 }]
+        video_title = self._html_search_regex(
+            r'(?s)<title>(?P<title>.*?)(?: \| Tumblr)?</title>',
+            webpage, 'title')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': video_title,
+            'description': self._og_search_description(webpage, default=None),
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+        }