[vine] Extract more metadata and support low format

[youtube-dl] / youtube_dl / extractor / vine.py

diff --git a/youtube_dl/extractor/vine.py b/youtube_dl/extractor/vine.py
index 651ba317dcd8fffefb2ac938c9f09d5de8356865..5bbc8ba88bbdb6a0d069324a3fd191844ccf7773 100644 (file)
--- a/youtube_dl/extractor/vine.py
+++ b/youtube_dl/extractor/vine.py
@@ -1,40 +1,60 @@
+from __future__ import unicode_literals
+
 import re
+import json
 
 from .common import InfoExtractor
+from ..utils import unified_strdate
 
 
 class VineIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?vine\.co/v/(?P<id>\w+)'
+    _VALID_URL = r'https?://(?:www\.)?vine\.co/v/(?P<id>\w+)'
     _TEST = {
-        u'url': u'https://vine.co/v/b9KOOWX7HUx',
-        u'file': u'b9KOOWX7HUx.mp4',
-        u'md5': u'2f36fed6235b16da96ce9b4dc890940d',
-        u'info_dict': {
-            u"uploader": u"Jack Dorsey", 
-            u"title": u"Chicken."
-        }
+        'url': 'https://vine.co/v/b9KOOWX7HUx',
+        'md5': '2f36fed6235b16da96ce9b4dc890940d',
+        'info_dict': {
+            'id': 'b9KOOWX7HUx',
+            'ext': 'mp4',
+            'title': 'Chicken.',
+            'description': 'Chicken.',
+            'upload_date': '20130519',
+            'uploader': 'Jack Dorsey',
+            'uploader_id': '76',
+        },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-
         video_id = mobj.group('id')
-        webpage_url = 'https://vine.co/v/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
-
-        self.report_extraction(video_id)
-
-        video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
-            webpage, u'video URL')
-
-        uploader = self._html_search_regex(r'<p class="username">(.*?)</p>',
-            webpage, u'uploader', fatal=False, flags=re.DOTALL)
 
-        return [{
-            'id':        video_id,
-            'url':       video_url,
-            'ext':       'mp4',
-            'title':     self._og_search_title(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'uploader':  uploader,
-        }]
+        webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id)
+
+        data = json.loads(self._html_search_regex(
+            r'window\.POST_DATA = { %s: ({.+?}) }' % video_id, webpage, 'vine data'))
+
+        formats = [
+            {
+                'url': data['videoLowURL'],
+                'ext': 'mp4',
+                'format_id': 'low',
+            },
+            {
+                'url': data['videoUrl'],
+                'ext': 'mp4',
+                'format_id': 'standard',
+            }
+        ]
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'description': data['description'],
+            'thumbnail': data['thumbnailUrl'],
+            'upload_date': unified_strdate(data['created']),
+            'uploader': data['username'],
+            'uploader_id': data['userIdStr'],
+            'like_count': data['likes']['count'],
+            'comment_count': data['comments']['count'],
+            'repost_count': data['reposts']['count'],
+            'formats': formats,
+        }
\ No newline at end of file