I wrote this script because I wanted to download my saved posts and the third-party APIs I found on Github didn't work for me.
Just a warning, this script really is not focused on interacting with the API in a way that appears authentic. So if you care about your account and decide to use this, understand the risk.
Files are downloaded into a folder called "saved-posts" created in whatever directory the
script is invocated from.
The script also generates a file named ".id_cache" in this directory to store the current
next_id of request batch. This is primarily used to quickly resume downloading from a fixed
point if anything goes awry with the with a batch.
- Refactor code
- Add download filters
The only cookies you need to be able to receive data back from the API are:
session_id. Other required cookies should be set automatically when interacting
with the web server.
When fetching posts the API returns a list of media confined to the following structure.
Since this project is just about getting saved posts, the most important fields are image_versions2 candidates,
and fields that can provide useful metadata, such as pk, id, epoch, username, and code.