[instagram:user] Fix extraction (closes #14699)

This commit is contained in:
Sergey M․ 2017-11-12 18:35:17 +07:00
parent af85ce29c6
commit 5fc12b9549
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 59 additions and 62 deletions

View File

@ -1,5 +1,6 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import itertools
import re import re
from .common import InfoExtractor from .common import InfoExtractor
@ -7,7 +8,6 @@ from ..compat import compat_str
from ..utils import ( from ..utils import (
get_element_by_attribute, get_element_by_attribute,
int_or_none, int_or_none,
limit_length,
lowercase_escape, lowercase_escape,
try_get, try_get,
) )
@ -212,7 +212,7 @@ class InstagramIE(InfoExtractor):
class InstagramUserIE(InfoExtractor): class InstagramUserIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])' _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<id>[^/]{2,})/?(?:$|[?#])'
IE_DESC = 'Instagram user profile' IE_DESC = 'Instagram user profile'
IE_NAME = 'instagram:user' IE_NAME = 'instagram:user'
_TEST = { _TEST = {
@ -221,82 +221,79 @@ class InstagramUserIE(InfoExtractor):
'id': 'porsche', 'id': 'porsche',
'title': 'porsche', 'title': 'porsche',
}, },
'playlist_mincount': 2, 'playlist_count': 5,
'playlist': [{
'info_dict': {
'id': '614605558512799803_462752227',
'ext': 'mp4',
'title': '#Porsche Intelligent Performance.',
'thumbnail': r're:^https?://.*\.jpg',
'uploader': 'Porsche',
'uploader_id': 'porsche',
'timestamp': 1387486713,
'upload_date': '20131219',
},
}],
'params': { 'params': {
'extract_flat': True, 'extract_flat': True,
'skip_download': True, 'skip_download': True,
'playlistend': 5,
} }
} }
def _real_extract(self, url): def _entries(self, uploader_id):
mobj = re.match(self._VALID_URL, url) query = {
uploader_id = mobj.group('username') '__a': 1,
}
entries = [] def get_count(kind):
page_count = 0 return int_or_none(try_get(
media_url = 'http://instagram.com/%s/media' % uploader_id node, lambda x: x['%ss' % kind]['count']))
while True:
for page_num in itertools.count(1):
page = self._download_json( page = self._download_json(
media_url, uploader_id, 'https://instagram.com/%s/' % uploader_id, uploader_id,
note='Downloading page %d ' % (page_count + 1), note='Downloading page %d' % page_num,
) fatal=False, query=query)
page_count += 1 if not page:
break
for it in page['items']: nodes = try_get(page, lambda x: x['user']['media']['nodes'], list)
if it.get('type') != 'video': if not nodes:
break
max_id = None
for node in nodes:
node_id = node.get('id')
if node_id:
max_id = node_id
if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
continue
video_id = node.get('code')
if not video_id:
continue continue
like_count = int_or_none(it.get('likes', {}).get('count'))
user = it.get('user', {})
formats = [{ info = self.url_result(
'format_id': k, 'https://instagram.com/p/%s/' % video_id,
'height': v.get('height'), ie=InstagramIE.ie_key(), video_id=video_id)
'width': v.get('width'),
'url': v['url'],
} for k, v in it['videos'].items()]
self._sort_formats(formats)
thumbnails_el = it.get('images', {}) description = try_get(
thumbnail = thumbnails_el.get('thumbnail', {}).get('url') node, [lambda x: x['caption'], lambda x: x['text']['id']],
compat_str)
thumbnail = node.get('thumbnail_src') or node.get('display_src')
timestamp = int_or_none(node.get('date'))
# In some cases caption is null, which corresponds to None comment_count = get_count('comment')
# in python. As a result, it.get('caption', {}) gives None like_count = get_count('like')
title = (it.get('caption') or {}).get('text', it['id']) view_count = int_or_none(node.get('video_views'))
entries.append({ info.update({
'id': it['id'], 'description': description,
'title': limit_length(title, 80),
'formats': formats,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'webpage_url': it.get('link'), 'timestamp': timestamp,
'uploader': user.get('full_name'), 'comment_count': comment_count,
'uploader_id': user.get('username'),
'like_count': like_count, 'like_count': like_count,
'timestamp': int_or_none(it.get('created_time')), 'view_count': view_count,
}) })
if not page['items']: yield info
break
max_id = page['items'][-1]['id'].split('_')[0]
media_url = (
'http://instagram.com/%s/media?max_id=%s' % (
uploader_id, max_id))
return { if not max_id:
'_type': 'playlist', break
'entries': entries,
'id': uploader_id, query['max_id'] = max_id
'title': uploader_id,
} def _real_extract(self, url):
uploader_id = self._match_id(url)
return self.playlist_result(
self._entries(uploader_id), uploader_id, uploader_id)