Master MongoDB with Python: Complete Guide to CRUD Operations

1 Connect to Database

1.1 Install PyMongo

Use pip to install the PyMongo library, which enables Python to communicate with MongoDB.

<code>python3 -m pip install pymongo</code>

You can also install a specific version or upgrade the package:

<code>python3 -m pip install pymongo==3.5.1</code>

<code>python3 -m pip install --upgrade pymongo</code>

1.2 Create a MongoClient

Initialize a connection to MongoDB. For a local instance without authentication:

<code>import pymongo
conn = pymongo.MongoClient()</code>

For a remote server, provide a URI that includes username, password, host and port:

<code>import pymongo
conn = pymongo.MongoClient('mongodb://test:[email protected]:27019')</code>

If the server does not require authentication, omit the credentials:

<code>import pymongo
conn = pymongo.MongoClient('mongodb://45.10.110.77:27019')</code>

1.3 Select Database and Collection

Two common ways exist to obtain a database and a collection.

Method 1 – attribute style:

<code>from pymongo import MongoClient
conn = MongoClient()
database = conn.chapter_1
collection = database.example_data_1</code>

Method 2 – dictionary‑style indexing (useful for variable names):

<code>from pymongo import MongoClient
db_name = 'chapter_1'
collection_name = 'example_data_1'
conn = MongoClient()
 database = conn[db_name]
 collection = database[collection_name]
</code>

Both methods are equivalent; the second is handy when iterating over multiple databases or collections.

Note: A collection is created only after the first document is inserted.

2 MongoDB Commands Mapped to PyMongo Methods

Most MongoDB shell commands correspond directly to PyMongo methods, with naming converted from camelCase to snake_case.

insertOne → insert_one

insertMany → insert_many

find → find

updateOne → update_one

updateMany → update_many

deleteOne → delete_one

deleteMany → delete_many

3 Inserting Data

Basic Syntax

<code>collection.insert_one(document)      # single document
collection.insert_many([doc1, doc2]) # multiple documents
</code>

Example

<code>collection.insert_many([
    {'name': '王小二','age':21,'student':True,'address':'广州'},
    {'name': '赵小三','age':20,'student':True,'address':'北京'},
    {'name': '钱小四','age':21,'student':True,'address':'上海'},
    {'name': '孙小五','age':20,'student':True,'address':'山东'},
    {'name': '李小六','age':None,'student':True,'address':'河北'},
    {'name': '欧阳小七','age':24,'student':False,'address':'杭州'},
    {'name': '公孙小八','age':25,'student':False,'address':'广州'}
])
</code>

PyMongo automatically creates the database and collection if they do not already exist.

4 Querying Data

Find One Document

<code>from pymongo import MongoClient
conn = MongoClient()
collection = conn.chapter_1.example_data_1
print(collection.find_one({}))
</code>

Find All Documents

<code>for doc in collection.find():
    print(doc)
</code>

Conditional Queries

<code>rows = collection.find({
    'age': {'$gt': 21, '$lt': 25},
    'name': {'$ne': '公孙小八'}
})
for r in rows:
    print(r)
</code>

Sorting and Distinct

<code># Sort by age descending
rows = collection.find().sort('age', -1)
# Distinct values of a field
unique_ages = collection.distinct('age')
</code>

When sorting large result sets, avoid executing the operation directly in GUI tools to prevent freezing.

5 Updating and Deleting Documents

Update Syntax

<code># Update a single document
collection.update_one({'name': '公孙小八'}, {'$set': {'age': 80, 'address': '美国'}})
# Update many documents
collection.update_many({'student': False}, {'$set': {'status': 'inactive'}})
# Upsert (insert if not found)
collection.update_one({'name': '隐身人'}, {'$set': {'age': 0, 'address': '里世界'}}, upsert=True)
</code>

Delete Syntax

<code># Delete one matching document
collection.delete_one({'age': 0})
# Delete all matching documents
collection.delete_many({'age': 0})
</code>

Always preview the documents with a find query before performing delete operations.

6 Special Cases When Translating MongoDB Commands to PyMongo

Null Values

MongoDB uses

null

; Python uses

None

. Replace

null

with

None

in queries.

Boolean Values

MongoDB uses

true

/

false

; Python uses

True

/

False

. Adjust the case accordingly.

Sorting Parameters

In the shell,

sort({field: 1})

is used. In PyMongo, call

sort('field', 1)

or

sort('field', -1)

.

Querying by _id

Import

ObjectId

from the

bson

package and use it in the query:

<code>from bson import ObjectId
collection.find({'_id': ObjectId('5e8ac5dfdd9cf99b7a446e99')})
</code>

Conclusion

The article introduced MongoDB installation, demonstrated how to use the graphical tool Robo 3T, and showed that most MongoDB operations can be directly ported to Python with PyMongo. Only a few edge cases—null handling, boolean literals, sorting syntax, and _id queries—require minor adjustments.