Run an Aggregation Pipeline

This guide explains how to build and run an aggregation pipeline. The aggregation pipeline lets you perform complex transformations and calculations on data items. A pipeline consists of 1 or more stages that process your data sequentially, with each stage's output becoming the input for the next stage.

The pipeline allows you to:

• Group items and calculate aggregate values.
• Transform result items by including, excluding, or reshaping fields.
• Filter, sort, and limit results.

Anatomy of a pipeline

An aggregation pipeline has the following key components:

• aggregatePipeline(): Runs an aggregation pipeline on the specified collection and returns the results.
• pipelineBuilder(): Builds an aggregation pipeline with stages and pagination configuration.
• Stages: Operations that process and transform data sequentially. Each stage type performs a different operation.
• Expressions: Operators and values for performing calculations, transformations, and field manipulations.

Build and run an aggregation pipeline

Building and running an aggregation pipeline involves 2 main phases: building the pipeline with your desired stages, then running it on a collection:

1. Import the items module and deconstruct the required submodules:

   Copy

2. Add 1 or more stages to build the pipeline:

   Copy

   Some stages resolve expressions to perform calculations and transformations. For example, you can create a new field by combining existing fields:

   Copy

3. Pass the stages array to the withStages() method. You can also add paging configuration with the withPaging() method. Finally, call build() to build the pipeline:

   Copy

   The pipeline is now ready.

4. Pass the configured pipeline to the aggregatePipeline() method to run the pipeline on the collection:

   Copy

See also

• Pipeline builder
• Aggregation pipeline stages
• Aggregation pipeline expressions