To bridge my conceptual leap from SQL to NoSQL, I wanted to convert MySQL tables/CSV files to Google Cloud Platform's much looser Firestore database structure (where, effectively, a table is a collection and a row is a document).
This script is designed to run in Google Cloud Platform's App Engine. For larger CSV files, this may be a problem and you may need to reconfigure it to run on a different server (or locally); if so, you'll find this useful to authenticate, etc.
- the location of the CSV (remote or local)
- the name of the collection you'd like it in (think table name)
- optionally, whether you want to set each document ID (think row ID) to auto increment or simply want Firestore to set it randomly (the default)
If you're not needing to do this dynamically, take a look at Google's options for importing and exporting to/from Firestore.