MongoDB with MongoEngine

Using a document database like MongoDB is a common alternative to relational SQL databases. This pattern shows how to use MongoEngine, a document mapper library, to integrate with MongoDB.

A running MongoDB server and Flask-MongoEngine are required.

  1. pip install flask-mongoengine

Configuration

Basic setup can be done by defining MONGODB_SETTINGS on app.config and creating a MongoEngine instance.

  1. from flask import Flask
  2. from flask_mongoengine import MongoEngine
  3. app = Flask(__name__)
  4. app.config['MONGODB_SETTINGS'] = {
  5. "db": "myapp",
  6. }
  7. db = MongoEngine(app)

Mapping Documents

To declare a model that represents a Mongo document, create a class that inherits from Document and declare each of the fields.

  1. import mongoengine as me
  2. class Movie(me.Document):
  3. title = me.StringField(required=True)
  4. year = me.IntField()
  5. rated = me.StringField()
  6. director = me.StringField()
  7. actors = me.ListField()

If the document has nested fields, use EmbeddedDocument to defined the fields of the embedded document and EmbeddedDocumentField to declare it on the parent document.

  1. class Imdb(me.EmbeddedDocument):
  2. imdb_id = me.StringField()
  3. rating = me.DecimalField()
  4. votes = me.IntField()
  5. class Movie(me.Document):
  6. ...
  7. imdb = me.EmbeddedDocumentField(Imdb)

Creating Data

Instantiate your document class with keyword arguments for the fields. You can also assign values to the field attributes after instantiation. Then call doc.save().

  1. bttf = Movie(title="Back To The Future", year=1985)
  2. bttf.actors = [
  3. "Michael J. Fox",
  4. "Christopher Lloyd"
  5. ]
  6. bttf.imdb = Imdb(imdb_id="tt0088763", rating=8.5)
  7. bttf.save()

Queries

Use the class objects attribute to make queries. A keyword argument looks for an equal value on the field.

  1. bttf = Movies.objects(title="Back To The Future").get_or_404()

Query operators may be used by concatenating them with the field name using a double-underscore. objects, and queries returned by calling it, are iterable.

  1. some_theron_movie = Movie.objects(actors__in=["Charlize Theron"]).first()
  2. for recents in Movie.objects(year__gte=2017):
  3. print(recents.title)

Documentation

There are many more ways to define and query documents with MongoEngine. For more information, check out the official documentation.

Flask-MongoEngine adds helpful utilities on top of MongoEngine. Check out their documentation as well.