X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Finstagram.py;h=11bb58d8a66edfc6ae06adb3d7e0d692262d757d;hb=6d4fc66bfc9bb3ed2a4f68366f372a9bedf6e708;hp=ed3e0711815bc36a6b8702c994087ae5827f5dab;hpb=a9d5f12fecf2f3e408cbc05409d1b296d47650ee;p=youtube-dl.git

diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py
index ed3e07118..11bb58d8a 100644
--- a/youtube_dl/extractor/instagram.py
+++ b/youtube_dl/extractor/instagram.py
@@ -4,8 +4,10 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    get_element_by_attribute,
     int_or_none,
     limit_length,
+    lowercase_escape,
 )
 
 
@@ -38,6 +40,18 @@ class InstagramIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_embed_url(webpage):
+        blockquote_el = get_element_by_attribute(
+            'class', 'instagram-media', webpage)
+        if blockquote_el is None:
+            return
+
+        mobj = re.search(
+            r'<a[^>]+href=([\'"])(?P<link>[^\'"]+)\1', blockquote_el)
+        if mobj:
+            return mobj.group('link')
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -46,6 +60,8 @@ class InstagramIE(InfoExtractor):
                                          webpage, 'uploader id', fatal=False)
         desc = self._search_regex(
             r'"caption":"(.+?)"', webpage, 'description', default=None)
+        if desc is not None:
+            desc = lowercase_escape(desc)
 
         return {
             'id': video_id,
@@ -136,7 +152,7 @@ class InstagramUserIE(InfoExtractor):
 
             if not page['items']:
                 break
-            max_id = page['items'][-1]['id']
+            max_id = page['items'][-1]['id'].split('_')[0]
             media_url = (
                 'http://instagram.com/%s/media?max_id=%s' % (
                     uploader_id, max_id))