aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorSergey M․ <dstftw@gmail.com>2018-03-21 23:43:03 +0700
committerSergey M․ <dstftw@gmail.com>2018-03-21 23:43:03 +0700
commitcba5d1b6b36d79fcafe0600d9805e6b82ed5388f (patch)
treec3bbe58eed97de511e848eca4fc5c836d5d05c9d
parent3395958d2befc710181bbde872074ce81eee9158 (diff)
[instagram:user] Add pagination (closes #15934)
-rw-r--r--youtube_dl/extractor/instagram.py104
1 files changed, 63 insertions, 41 deletions
diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py
index ac9d92a8d..f9cd11b8e 100644
--- a/youtube_dl/extractor/instagram.py
+++ b/youtube_dl/extractor/instagram.py
@@ -1,5 +1,6 @@
from __future__ import unicode_literals
+import itertools
import json
import re
@@ -242,48 +243,69 @@ class InstagramUserIE(InfoExtractor):
return int_or_none(try_get(
node, lambda x: x['edge_media_' + suffix]['count']))
- edges = self._download_json(
- 'https://www.instagram.com/graphql/query/', uploader_id, query={
- 'query_hash': '472f257a40c653c64c666ce877d59d2b',
- 'variables': json.dumps({
- 'id': uploader_id,
- 'first': 999999999,
+ cursor = ''
+ for page_num in itertools.count(1):
+ media = self._download_json(
+ 'https://www.instagram.com/graphql/query/', uploader_id,
+ 'Downloading JSON page %d' % page_num, query={
+ 'query_hash': '472f257a40c653c64c666ce877d59d2b',
+ 'variables': json.dumps({
+ 'id': uploader_id,
+ 'first': 100,
+ 'after': cursor,
+ })
+ })['data']['user']['edge_owner_to_timeline_media']
+
+ edges = media.get('edges')
+ if not edges or not isinstance(edges, list):
+ break
+
+ for edge in edges:
+ node = edge.get('node')
+ if not node or not isinstance(node, dict):
+ continue
+ if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
+ continue
+ video_id = node.get('shortcode')
+ if not video_id:
+ continue
+
+ info = self.url_result(
+ 'https://instagram.com/p/%s/' % video_id,
+ ie=InstagramIE.ie_key(), video_id=video_id)
+
+ description = try_get(
+ node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
+ compat_str)
+ thumbnail = node.get('thumbnail_src') or node.get('display_src')
+ timestamp = int_or_none(node.get('taken_at_timestamp'))
+
+ comment_count = get_count('to_comment')
+ like_count = get_count('preview_like')
+ view_count = int_or_none(node.get('video_view_count'))
+
+ info.update({
+ 'description': description,
+ 'thumbnail': thumbnail,
+ 'timestamp': timestamp,
+ 'comment_count': comment_count,
+ 'like_count': like_count,
+ 'view_count': view_count,
})
- })['data']['user']['edge_owner_to_timeline_media']['edges']
-
- for edge in edges:
- node = edge['node']
-
- if node.get('__typename') != 'GraphVideo' and node.get('is_video') is not True:
- continue
- video_id = node.get('shortcode')
- if not video_id:
- continue
-
- info = self.url_result(
- 'https://instagram.com/p/%s/' % video_id,
- ie=InstagramIE.ie_key(), video_id=video_id)
-
- description = try_get(
- node, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'],
- compat_str)
- thumbnail = node.get('thumbnail_src') or node.get('display_src')
- timestamp = int_or_none(node.get('taken_at_timestamp'))
-
- comment_count = get_count('to_comment')
- like_count = get_count('preview_like')
- view_count = int_or_none(node.get('video_view_count'))
-
- info.update({
- 'description': description,
- 'thumbnail': thumbnail,
- 'timestamp': timestamp,
- 'comment_count': comment_count,
- 'like_count': like_count,
- 'view_count': view_count,
- })
-
- yield info
+
+ yield info
+
+ page_info = media.get('page_info')
+ if not page_info or not isinstance(page_info, dict):
+ break
+
+ has_next_page = page_info.get('has_next_page')
+ if not has_next_page:
+ break
+
+ cursor = page_info.get('end_cursor')
+ if not cursor or not isinstance(cursor, compat_str):
+ break
def _real_extract(self, url):
username = self._match_id(url)