MongoDB and Python

Chapter 2. Reading and Writing to MongoDB with Python

MongoDB is a document-oriented database. This is different from a relational database in two significant ways. Firstly, not all entries must adhere to the same schema. Secondly you can embed entries inside of one another. Despite these major differences, there are analogs to SQL concepts in MongoDB. A logical group of entries in a SQL database is termed a table. In MongoDB, the analogous term is a collection. A single entry in a SQL databse is termed a row. In MongoDB, the analog is a document.

Table 2-1. Comparison of SQL/RDBMS and MongoDB Concepts and Terms

Concept	SQL	MongoDB
One User	One Row	One Document
All Users	Users Table	Users Collection
One Username Per User (1-to-1)	Username Column	Username Property
Many Emails Per User (1-to-many)	SQL JOIN with Emails Table	Embed relevant email doc in User Document
Many Items Owned by Many Users (many-to-many)	SQL JOIN with Items Table	Programmatically Join with Items Collection

Hence, in MongoDB, you are mostly operating on documents and collections of documents. If you are familiar with JSON, a MongoDB document is essentially a JSON document with a few extra features. From a Python perspective, it is a Python dictionary.

Consider the following example of a user document with a username, first name, surname, date of birth, email address and score:

from datetime import datetime user_doc = { "username" : "janedoe", "firstname" : "Jane", "surname" : "Doe", "dateofbirth" : datetime(1974, 4, 12), "email" : ...