PyMongo and Beanie, Deep Dive

MongoDB from Python: queries without SQL, and typed documents through an asynchronous mapper.

Complete 16 notebooks 16 written

Before this guide: APIs and JSON.

The notebooks

Open in Colab runs a notebook in your browser, and Read shows it on GitHub without running it. Every notebook runs on its own, and the answers to its exercises are in a separate notebook, linked as Solutions.

1	Why DocumentsOne object instead of five tables, and a MongoDB running inside this notebook	Open in Colab Read Solutions
2	Collections and Documentsinsert_one, insert_many, the _id you did not choose, and the collection a typo invents	Open in Colab Read Solutions
3	BSON TypesThe set you cannot store, the datetime that comes back naive, and the ObjectId json.dumps refuses	Open in Colab Read Solutions
4	find and find_oneOne document against a cursor you can walk only once, and a projection that keeps it small	Open in Colab Read Solutions
5	Query OperatorsWhy {'size': {'w': 1}} matches nothing when {'size.w': 1} matches every document where w is 1	Open in Colab Read Solutions
6	Update Operators$set against the replacement that dropped every other field, and the upsert that inserted a second copy	Open in Colab Read Solutions
7	Bulk Writes and Transactionsbulk_write, ordered against unordered, and the write you forgot to pass the session to	Open in Colab Read Solutions
8	IndexesCompound order, explain, $text, and the unique index that refuses to build	Open in Colab Read Solutions
9	The Aggregation Pipeline$match, $group, $unwind, and the stage order that decides how long it runs	Open in Colab Read Solutions
10	Modeling Without JoinsEmbed until the document will not fit, and the $lookup that scans because the foreign field has no index	Open in Colab Read Solutions
11	Beanie DocumentsA Pydantic model that is also a collection, init_beanie, and the await a Colab cell already allows	Open in Colab Read Solutions
12	Async Queriesfind, to_list, beanie.operators, and the projection model that will not validate	Open in Colab Read Solutions
13	Saving Changesset and inc against save, save_changes, and the revision that moved under you	Open in Colab Read Solutions
14	Link and BackLinkLink, BackLink, fetch_links, and the single $lookup pipeline that saves a round trip per document	Open in Colab Read Solutions
15	Migrationsbeanie new-migration, iterative against free fall, and the replica set beanie migrate assumes	Open in Colab Read Solutions
16	A Product CatalogOne real job: loaded with PyMongo, indexed for the queries it serves, typed with Beanie	Open in Colab Read Solutions