Login required for locations as of 04/2022

[do-me]

After recent changes, the old endpoints, e.g.:

https://instagram.com/graphql/query/?query_hash=ac38b90f0f3981c42092016a37c59bf7&variables={"id":"<some_location_id>","first":50,"after":""}

seem not to work anymore.

However location IDs can still be mined successfully via a different endpoint:

https://www.instagram.com/explore/locations/<some_location_id>/?__a=1&max_id=<last_cursor>

Currently don't have time to investigate further but this quick fix works for me:

1. Change lines 47-49 to:

    if location_or_hashtag == "location":
        instalink = 'https://www.instagram.com/explore/locations/' + str(object_id_or_string) + '/?__a=1&max_id=' + cursor 
        return instalink

2. Due to different JSON respone replace all idata["data"] with idata["graphql"].

Same applies to hashtags. Just updated the code so hashtags can be mined again.

[arifluthfi16]

Hey man, this is an appreciation comment of how amazing this implementation is :D thank you very much i learned a lot!

This repo is definitely underrated!

[do-me]

Locations cannot be retrieved anymore without login. Hashtags as of 04/2022 still do work without login but seldomly in TOR (you must be lucky to hit a good endpoint), e.g.

• https://www.instagram.com/explore/tags/brazil/ for website
• https://www.instagram.com/explore/tags/brazil/?__a=1 for the API endpoint

but it might be only a matter of time until it's going to be closed entirely (without login).

However, posts seem to be consistently publicly available (maybe even for a longer time to positive Instagram SEO). Hence, I propose

1. Mining location or hashtag sites with Instagram Scraper with login (at a reduced pace obviously).
2. If needed re-mine single posts with fast-instagram-scraper but in a tweaked version for posts only. This would require some slight changes which are not implemented yet.

I described it here as well.

[do-me]

Works again for some reason, test e.g. here. Closing for now.