w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[instagram] fix user videos extraction(fixes #15858)

Browse source
This commit is contained in:
Remita Amine 2018-03-15 14:33:36 +01:00
parent 46c6742d4f
commit 27b1c73f14
1 changed files with 47 additions and 59 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

62
youtube_dl/extractor/instagram.py
View file

@ -1,6 +1,6 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import itertools import json
import re import re
from .common import InfoExtractor from .common import InfoExtractor
@ -238,36 +238,25 @@ class InstagramUserIE(InfoExtractor):
} }
def _entries(self, uploader_id): def _entries(self, uploader_id):
query = { def get_count(suffix):
'__a': 1,
}
def get_count(kind):
return int_or_none(try_get( return int_or_none(try_get(
node, lambda x: x['%ss' % kind]['count'])) node, lambda x: x['edge_media_' + suffix]['count']))
for page_num in itertools.count(1): edges = self._download_json(
page = self._download_json( 'https://www.instagram.com/graphql/query/', uploader_id, query={
'https://instagram.com/%s/' % uploader_id, uploader_id, 'query_hash': '472f257a40c653c64c666ce877d59d2b',
note='Downloading page %d' % page_num, 'variables': json.dumps({
fatal=False, query=query) 'id': uploader_id,
if not page: 'first': 999999999,
break })
})['data']['user']['edge_owner_to_timeline_media']['edges']
nodes = try_get(page, lambda x: x['user']['media']['nodes'], list) for edge in edges:
if not nodes: node = edge['node']
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: if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
continue continue
video_id = node.get('code') video_id = node.get('shortcode')
if not video_id: if not video_id:
continue continue
@ -276,14 +265,14 @@ class InstagramUserIE(InfoExtractor):
ie=InstagramIE.ie_key(), video_id=video_id) ie=InstagramIE.ie_key(), video_id=video_id)
description = try_get( description = try_get(
node, [lambda x: x['caption'], lambda x: x['text']['id']], node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
compat_str) compat_str)
thumbnail = node.get('thumbnail_src') or node.get('display_src') thumbnail = node.get('thumbnail_src') or node.get('display_src')
timestamp = int_or_none(node.get('date')) timestamp = int_or_none(node.get('taken_at_timestamp'))
comment_count = get_count('comment') comment_count = get_count('to_comment')
like_count = get_count('like') like_count = get_count('preview_like')
view_count = int_or_none(node.get('video_views')) view_count = int_or_none(node.get('video_view_count'))
info.update({ info.update({
'description': description, 'description': description,
@ -296,12 +285,11 @@ class InstagramUserIE(InfoExtractor):
yield info yield info
if not max_id:
break
query['max_id'] = max_id
def _real_extract(self, url): def _real_extract(self, url):
uploader_id = self._match_id(url) username = self._match_id(url)
uploader_id = self._download_json(
'https://instagram.com/%s/' % username, username, query={
'__a': 1,
})['graphql']['user']['id']
return self.playlist_result( return self.playlist_result(
self._entries(uploader_id), uploader_id, uploader_id) self._entries(uploader_id), username, username)