[kickstarter] Fix extraction, extract more info and modernize

youtube_dl/extractor/kickstarter.py

@@ -1,37 +1,39 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class KickStarterIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>\d*)/.*'
+    _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*'
     _TEST = {
-        u"url": u"https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location",
+        'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location',
-        u"file": u"1404461844.mp4",
+        'md5': 'c81addca81327ffa66c642b5d8b08cab',
-        u"md5": u"c81addca81327ffa66c642b5d8b08cab",
+        'info_dict': {
-        u"info_dict": {
+            'id': '1404461844',
-            u"title": u"Intersection: The Story of Josh Grant by Kyle Cowling",
+            'ext': 'mp4',
+            'title': 'Intersection: The Story of Josh Grant by Kyle Cowling',
+            'description': 'A unique motocross documentary that examines the '
+                'life and mind of one of sports most elite athletes: Josh Grant.',
         },
     }
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
         video_id = m.group('id')
-        webpage_src = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(r'data-video="(.*?)">',
+        video_url = self._search_regex(r'data-video-url="(.*?)"',
-            webpage_src, u'video URL')
+            webpage, 'video URL')
-        if 'mp4' in video_url:
+        video_title = self._html_search_regex(r'<title>(.*?)</title>',
-            ext = 'mp4'
+            webpage, 'title').rpartition('— Kickstarter')[0].strip()
-        else:
-            ext = 'flv'
-        video_title = self._html_search_regex(r"<title>(.*?)</title>",
-            webpage_src, u'title').rpartition(u'\u2014 Kickstarter')[0].strip()
 
-        results = [{
+        return {
-                    'id': video_id,
+            'id': video_id,
-                    'url': video_url,
+            'url': video_url,
-                    'title': video_title,
+            'title': video_title,
-                    'ext': ext,
+            'description': self._og_search_description(webpage),
-                    }]
+            'thumbnail': self._og_search_thumbnail(webpage),
-        return results
+        }