Aggregation Framework (Introductie)

Introduction to the Aggregation Framework

The Aggregation Framework in MongoDB is a powerful tool for transforming and analyzing data within the database.

It allows you to perform operations like filtering, grouping, sorting, and calculating summary statistics — similar to SQL’s GROUP BY and aggregate functions.

---

Aggregation Pipelines

An aggregation pipeline is a sequence of stages, where each stage processes the output of the previous one.

Example: Calculate the average price per brand.

db.products.aggregate([
  { $group: { _id: "$brand", avg_price: { $avg: "$price" } } }
])

This aggregation pipeline consists of a single stage — the $group stage.

• The $group stage groups all documents by the value of the brand field (_id: “$brand”).
• For each group, it calculates the average price using the $avg operator.
• The result is one document per brand, containing the brand name (_id) and its corresponding average price.

Example output:

[
  { "_id": "Apple", "avg_price": 799 },
  { "_id": "Samsung", "avg_price": 699 },
  { "_id": "Dell", "avg_price": 299 }
]

Common Stages

| Stage | Description | Example | | ———- | ——————————— | ———————————————————- | | $match | Filters documents (like find()) | { $match: { brand: "Apple" } } | | $group | Groups documents by a field | { $group: { _id: "$brand", total: { $sum: "$price" } } } | | $project | Selects or reshapes fields | { $project: { name: 1, brand: 1, price: 1 } } | | $sort | Sorts documents | { $sort: { price: -1 } } | | $limit | Limits the number of results | { $limit: 5 } |

You can chain these stages together to perform complex analyses.

Example: Top 3 most expensive products per brand

db.products.aggregate([
  { $sort: { brand: 1, price: -1 } },
  { $group: {
      _id: "$brand",
      top_products: { $push: { name: "$name", price: "$price" } }
  }},
  { $project: {
      _id: 1,
      top_3: { $slice: ["$top_products", 3] }
  }}
])

This aggregation pipeline consists of three stages:

• $sort — Sorts all products first by brand in ascending order (A–Z), and then within each brand by price in descending order (highest first).
• $group — Groups the sorted documents by brand. For each brand, it builds an array called top_products that contains the name and price of every product in that brand, in the sorted order.
• $project — Reshapes the output from the previous stage. It creates a new field top_3, which contains only the first three products from the top_products array using the $slice operator. As a result, you get one document per brand, each containing its three most expensive products.

Practice Questions

🧠 1. Which aggregation stage is similar to WHERE in SQL?

• A) $group
• B) $match
• C) $sort
• D) $project

Click to reveal answer

✅ Correct: **B**

🧠 2. What does $group do in an aggregation pipeline?

Click to reveal answer

It groups documents by a field and allows you to calculate aggregated values like `$sum`, `$avg`, or `$count`.

🧠 3. Write an aggregation query that counts the number of products per brand.

Click to reveal example solution

db.products.aggregate([ { $group: { _id: "$brand", count: { $count: {} } } } ])