arsinclair / instagram-backup-arranger

A small tool that consumes an Instagram backup and produces media files from Stories named chronologically.

A small tool that consumes an Instagram backup and produces media files from Stories named chronologically.

Requires the following tools pre-installed:

• imagemagick for transforming .webp images to .jpg;
• libmagic to determine the file types for files without extensions.

Several quirks of Stories in Instagram Backups

• Sometimes the file won't have an extension, only the name. So far, only observed with .mp4 videos;
• Some images are exported in .webp format and some in .jpg; it's not consistent;
• Some Media URIs in the stories.json won't point to a local file, but instead to a remote instagram CDN. When tried accessing such remote file, the Instagram CDN would return a 403 error indicating that a hash has expired. Such stories are also not displayed correctly in the Story Archive of the official Instagram app. The most likely cause for this seems to be a case when the story is a repost of a reel that has since been taken down;
• the textual content is encoded in latin1 on the server, but the JSON returned is utf-8 and they don't do a proper conversion, so any textual content we have to decode back to latin1, then properly encode it to utf-8.

Run the tool

1. Request an Instagram Data export in JSON format;
2. Extract the archive received in a new folder and change the paths in .env file to point to that folder;
3. Run pnpm start.

The result of the conversion will be saved in an output/ folder.