+ _VALID_URL = r'(?P<url>https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+))'
+ _TESTS = [{
+ 'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
+ 'md5': '0d2da106a9d2631273e192b372806516',
+ 'info_dict': {
+ 'id': 'aye83DjauH',
+ 'ext': 'mp4',
+ 'title': 'Video by naomipq',
+ 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
+ 'thumbnail': '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',
+ 'title': 'Video by britneyspears',
+ 'thumbnail': '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,
+ },
+ }, {
+ 'url': 'https://instagram.com/p/-Cmh1cukG2/',
+ 'only_matching': True,
+ }, {
+ 'url': 'http://instagram.com/p/9o6LshA7zy/embed/',
+ 'only_matching': True,
+ }]
+
+ @staticmethod
+ def _extract_embed_url(webpage):
+ mobj = re.search(
+ r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?instagram\.com/p/[^/]+/embed.*?)\1',
+ webpage)
+ if mobj:
+ return mobj.group('url')
+
+ 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):
+ mobj = re.match(self._VALID_URL, url)
+ video_id = mobj.group('id')
+ url = mobj.group('url')
+
+ webpage = self._download_webpage(url, video_id)
+
+ (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,
+ 'formats': formats,
+ 'ext': 'mp4',
+ 'title': 'Video by %s' % uploader_id,
+ 'description': description,
+ 'thumbnail': thumbnail,
+ 'timestamp': timestamp,
+ 'uploader_id': uploader_id,
+ 'uploader': uploader,
+ 'like_count': like_count,
+ 'comment_count': comment_count,
+ 'comments': comments,
+ }
+
+
+class InstagramUserIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
+ IE_DESC = 'Instagram user profile'
+ IE_NAME = 'instagram:user'