X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Finstagram.py;h=98f408c18650cf8393869432a861a3486575b533;hb=ec85ded83cbfa652ba94cb080aab52d8b270212a;hp=3cbe77ad80f2fc9a03c738745524d5dac98c9d37;hpb=c23533a100eb0a383b60ed2e5990602e9318fe4b;p=youtube-dl

diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py
index 3cbe77ad8..98f408c18 100644
--- a/youtube_dl/extractor/instagram.py
+++ b/youtube_dl/extractor/instagram.py
@@ -8,6 +8,7 @@ from ..utils import (
     int_or_none,
     limit_length,
     lowercase_escape,
+    try_get,
 )
 
 
@@ -19,18 +20,32 @@ class InstagramIE(InfoExtractor):
         'info_dict': {
             'id': 'aye83DjauH',
             'ext': 'mp4',
-            'uploader_id': 'naomipq',
             'title': 'Video by naomipq',
             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
-        }
+            'thumbnail': r're:^https?://.*\.jpg',
+            'timestamp': 1371748545,
+            'upload_date': '20130620',
+            'uploader_id': 'naomipq',
+            'uploader': 'Naomi Leonor Phan-Quang',
+            'like_count': int,
+            'comment_count': int,
+            'comments': list,
+        },
     }, {
         # missing description
         'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
         'info_dict': {
             'id': 'BA-pQFBG8HZ',
             'ext': 'mp4',
-            'uploader_id': 'britneyspears',
             'title': 'Video by britneyspears',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'timestamp': 1453760977,
+            'upload_date': '20160125',
+            'uploader_id': 'britneyspears',
+            'uploader': 'Britney Spears',
+            'like_count': int,
+            'comment_count': int,
+            'comments': list,
         },
         'params': {
             'skip_download': True,
@@ -67,21 +82,74 @@ class InstagramIE(InfoExtractor):
         url = mobj.group('url')
 
         webpage = self._download_webpage(url, video_id)
-        uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
-                                         webpage, 'uploader id', fatal=False)
-        desc = self._search_regex(
-            r'"caption":"(.+?)"', webpage, 'description', default=None)
-        if desc is not None:
-            desc = lowercase_escape(desc)
+
+        (video_url, description, thumbnail, timestamp, uploader,
+         uploader_id, like_count, comment_count, height, width) = [None] * 10
+
+        shared_data = self._parse_json(
+            self._search_regex(
+                r'window\._sharedData\s*=\s*({.+?});',
+                webpage, 'shared data', default='{}'),
+            video_id, fatal=False)
+        if shared_data:
+            media = try_get(
+                shared_data, lambda x: x['entry_data']['PostPage'][0]['media'], dict)
+            if media:
+                video_url = media.get('video_url')
+                height = int_or_none(media.get('dimensions', {}).get('height'))
+                width = int_or_none(media.get('dimensions', {}).get('width'))
+                description = media.get('caption')
+                thumbnail = media.get('display_src')
+                timestamp = int_or_none(media.get('date'))
+                uploader = media.get('owner', {}).get('full_name')
+                uploader_id = media.get('owner', {}).get('username')
+                like_count = int_or_none(media.get('likes', {}).get('count'))
+                comment_count = int_or_none(media.get('comments', {}).get('count'))
+                comments = [{
+                    'author': comment.get('user', {}).get('username'),
+                    'author_id': comment.get('user', {}).get('id'),
+                    'id': comment.get('id'),
+                    'text': comment.get('text'),
+                    'timestamp': int_or_none(comment.get('created_at')),
+                } for comment in media.get(
+                    'comments', {}).get('nodes', []) if comment.get('text')]
+
+        if not video_url:
+            video_url = self._og_search_video_url(webpage, secure=False)
+
+        formats = [{
+            'url': video_url,
+            'width': width,
+            'height': height,
+        }]
+
+        if not uploader_id:
+            uploader_id = self._search_regex(
+                r'"owner"\s*:\s*{\s*"username"\s*:\s*"(.+?)"',
+                webpage, 'uploader id', fatal=False)
+
+        if not description:
+            description = self._search_regex(
+                r'"caption"\s*:\s*"(.+?)"', webpage, 'description', default=None)
+            if description is not None:
+                description = lowercase_escape(description)
+
+        if not thumbnail:
+            thumbnail = self._og_search_thumbnail(webpage)
 
         return {
             'id': video_id,
-            'url': self._og_search_video_url(webpage, secure=False),
+            'formats': formats,
             'ext': 'mp4',
             'title': 'Video by %s' % uploader_id,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
             'uploader_id': uploader_id,
-            'description': desc,
+            'uploader': uploader,
+            'like_count': like_count,
+            'comment_count': comment_count,
+            'comments': comments,
         }
 
 
@@ -101,7 +169,7 @@ class InstagramUserIE(InfoExtractor):
                 'id': '614605558512799803_462752227',
                 'ext': 'mp4',
                 'title': '#Porsche Intelligent Performance.',
-                'thumbnail': 're:^https?://.*\.jpg',
+                'thumbnail': r're:^https?://.*\.jpg',
                 'uploader': 'Porsche',
                 'uploader_id': 'porsche',
                 'timestamp': 1387486713,