Master MongoDB with PyMongo: Essential CRUD Operations in Python

MongoDB is a C++-written NoSQL database that stores JSON-like documents; this guide shows how to use MongoDB with Python3 via PyMongo.

1. Preparation

Ensure MongoDB is installed and running, and install the PyMongo library.

2. Connect to MongoDB

Use MongoClient from PyMongo, providing host and port (default 27017) or a connection string.

import pymongo
client = pymongo.MongoClient(host='localhost', port=27017)

Or using a connection string:

client = MongoClient('mongodb://localhost:27017/')

3. Select Database

Specify the database, e.g., test:

db = client.test
# or
db = client['test']

4. Select Collection

Declare a collection, e.g., students:

collection = db.students
# or
collection = db['students']

5. Insert Data

Insert a single document:

student = {'id':'20170101','name':'Jordan','age':20,'gender':'male'}
result = collection.insert(student)
print(result)

Insert multiple documents with insert_many (or insert_one for a single record):

student1 = {'id':'20170101','name':'Jordan','age':20,'gender':'male'}
student2 = {'id':'20170202','name':'Mike','age':21,'gender':'male'}
result = collection.insert_many([student1, student2])
print(result.inserted_ids)

6. Query

Use find_one for a single result or find for a cursor of multiple results.

result = collection.find_one({'name':'Mike'})
print(result)

Query by ObjectId:

from bson.objectid import ObjectId
result = collection.find_one({'_id': ObjectId('593278c115c2602667ec6bae')})
print(result)

Range and regex queries use operators such as $gt, $lt, $regex, etc.

7. Count

Count documents with count() on a cursor.

count = collection.find().count()
print(count)

8. Sort

Sort results with sort() and specify ascending or descending.

results = collection.find().sort('name', pymongo.ASCENDING)
print([r['name'] for r in results])

9. Skip & Limit

Use skip() and limit() for pagination.

results = collection.find().sort('name', pymongo.ASCENDING).skip(2).limit(2)
print([r['name'] for r in results])

10. Update

Update a document with update_one or update_many, typically using $set or other operators.

condition = {'name':'Kevin'}
result = collection.update_one(condition, {'$set': {'age':25}})
print(result.modified_count)

11. Delete

Delete documents with delete_one or delete_many (or the older remove).

result = collection.delete_one({'name':'Kevin'})
print(result.deleted_count)

12. Additional Operations

PyMongo also provides methods such as find_one_and_update, create_index, and others for advanced use.