AQL Expression

An AQL expression is a structured combination of data types, operators, and functions, allowing you to query your database or define reusable metrics.

Types

There are two main types of AQL expressions:

• Table expressions: Expressions that represent a query that returns a table output, similar to SQL
• Metric expressions: Expressions that represent a reusable metric, which is basically some aggregation logic with added context

Structure

A typical AQL expression often is a combination of AQL functions combined together using the AQL pipe operator, like the followings:

A query that returns the sum of values of all orders from a order_items table:

order_items | select(value: quantity * price) | sum(value)

A metric that returns the running total of number of orders in 2023:

count(orders.id)
| where(orders.created_at is @2023)
| running_total(run: orders.created_at | month())

Dataset

An AQL expression only works within the context of a dataset as it requires knowledge of the models and relationships between them defined in the dataset.