]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/iconosquare.py
debian/control: Prefer mplayer to mplayer2 in Recommends.
[youtubedl] / youtube_dl / extractor / iconosquare.py
index 1d5a10a3b6349d95387aee00d4be5e6deac618b8..70e4c0d4173816e990749759cf2d36fe902904ee 100644 (file)
@@ -1,40 +1,75 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import int_or_none
 
 
 class IconosquareIE(InfoExtractor):
-    _VALID_URL = r'https?://(www\.)?(?:iconosquare\.com|statigr\.am)/p/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:iconosquare\.com|statigr\.am)/p/(?P<id>[^/]+)'
     _TEST = {
         'url': 'http://statigr.am/p/522207370455279102_24101272',
         'md5': '6eb93b882a3ded7c378ee1d6884b1814',
         'info_dict': {
             'id': '522207370455279102_24101272',
             'ext': 'mp4',
-            'uploader_id': 'aguynamedpatrick',
-            'title': 'Instagram photo by @aguynamedpatrick (Patrick Janelle)',
+            'title': 'Instagram media by @aguynamedpatrick (Patrick Janelle)',
             'description': 'md5:644406a9ec27457ed7aa7a9ebcd4ce3d',
+            'timestamp': 1376471991,
+            'upload_date': '20130814',
+            'uploader': 'aguynamedpatrick',
+            'uploader_id': '24101272',
+            'comment_count': int,
+            'like_count': int,
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        html_title = self._html_search_regex(
-            r'<title>(.+?)</title>',
+
+        media = self._parse_json(
+            self._search_regex(
+                r'window\.media\s*=\s*({.+?});\n', webpage, 'media'),
+            video_id)
+
+        formats = [{
+            'url': f['url'],
+            'format_id': format_id,
+            'width': int_or_none(f.get('width')),
+            'height': int_or_none(f.get('height'))
+        } for format_id, f in media['videos'].items()]
+        self._sort_formats(formats)
+
+        title = self._html_search_regex(
+            r'<title>(.+?)(?: *\(Videos?\))? \| (?:Iconosquare|Statigram)</title>',
             webpage, 'title')
-        title = re.sub(r'(?: *\(Videos?\))? \| (?:Iconosquare|Statigram)$', '', html_title)
-        uploader_id = self._html_search_regex(
-            r'@([^ ]+)', title, 'uploader name', fatal=False)
+
+        timestamp = int_or_none(media.get('created_time') or media.get('caption', {}).get('created_time'))
+        description = media.get('caption', {}).get('text')
+
+        uploader = media.get('user', {}).get('username')
+        uploader_id = media.get('user', {}).get('id')
+
+        comment_count = int_or_none(media.get('comments', {}).get('count'))
+        like_count = int_or_none(media.get('likes', {}).get('count'))
+
+        thumbnails = [{
+            'url': t['url'],
+            'id': thumbnail_id,
+            'width': int_or_none(t.get('width')),
+            'height': int_or_none(t.get('height'))
+        } for thumbnail_id, t in media.get('images', {}).items()]
 
         return {
             'id': video_id,
-            'url': self._og_search_video_url(webpage),
             'title': title,
-            'description': self._og_search_description(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'uploader_id': uploader_id
+            'description': description,
+            'thumbnails': thumbnails,
+            'timestamp': timestamp,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'comment_count': comment_count,
+            'like_count': like_count,
+            'formats': formats,
         }