When working with MongoDB in Python using PyMongo, you typically retrieve data from a collection using the find() method. But instead of getting all the matching documents at once, PyMongo returns something called a cursor.
A PyMongo cursor is a special iterable object that acts like a pointer to the result set of a MongoDB query. It allows you to fetch and process query results one document at a time rather than loading everything into memory at once.
Why do we need Cursor
Using a cursor provides several important advantages, especially when dealing with large datasets:
- Memory-efficient: Documents are retrieved in small batches instead of all at once, preventing memory overload.
- Lazy loading: Results are fetched as you iterate, which avoids blocking your application.
- Incremental processing: You can handle one document at a time in a loop, which is great for streaming, filtering or transforming data.
- Improved performance: Reduces pressure on both the client and the MongoDB server.
Cursor Customization Options
You can fine-tune the data returned by the cursor using:
- projection : to select specific fields from documents
- sort() : to order results based on one or more fields
- limit() : to restrict the number of documents returned
- skip() : to skip a specified number of documents
Sample database is as follows: 
Example: Using Cursor in PyMongo
from pymongo import MongoClient
# Connecting to mongodb
client = MongoClient('mongodb://localhost:27017/')
with client:
db = client.GFG
lectures = db.lecture.find()
print(lectures.next())
print(lectures.next())
print(lectures.next())
print("\nRemaining Lectures\n")
print(list(lectures))
Output:
Explanation:
- find() method returns the cursor object.
- next() method we get the next document in the collection.
- list() method, we can transform the cursor to a Python list.