Meet Malloy

Malloy is an experimental language for describing data relationships and transformations. It is both a semantic modeling language and a querying language that runs queries against a relational database. Malloy is currently available on BigQuery and Postgres. About Malloy:

• Queries compile to SQL, optimized for your database
• Computations are modular, composable, reusable, and extendable in ways that are consistent with modern programming paradigms
• Excels at querying and producing nested data sets
• The fan and chasm traps are solved, making it possible to aggregate anything in one query and reducing need for fact tables and overly complex SQL
• Defaults are smart, and the language is concise (where SQL is verbose and often redundant)

Malloy is a language for anyone who works with SQL--whether you’re an analyst, data scientist, data engineer, or someone building a data application. If you know SQL, Malloy will feel familiar, while more powerful and efficient. Malloy allows you to model as you go, so there is no heavy up-front work before you can start answering complex questions, and you're never held back or restricted by the model.

We've built a Visual Studio Code extension to facilitate interacting with your data using Malloy. The extension provides a rich environment to create Malloy data models, query and transform data, and to create simple visualizations and dashboards.

Syntax Example

We recommend starting with the Quickstart to get acquainted with the syntax. Here is a simple example of a Malloy query:

query: table('malloy-data.faa.flights') -> {
  where: origin: 'SFO'
  group_by: carrier
  aggregate: [
    flight_count is count()
    average_flight_time is flight_time.avg()
  ]
}

In SQL this would be expressed:

SELECT
   carrier,
   COUNT(*) as flight_count,
   AVG(flight_time) as average_flight_time
FROM `malloy-data.faa.flights`
WHERE origin = 'SFO'
GROUP BY carrier
ORDER BY flight_count desc         -- malloy automatically orders by the first aggregate

Learn more about the syntax and language features of Malloy in the Quickstart.

Get Started

GitHub mutes videos by default, so make sure to unmute to hear the walkthrough.

Installing the Extension

Currently, the Malloy extension works on Mac and Linux machines.

1. Download Visual Studio Code: If you don't already have it, download Visual Studio Code

2. Add the Malloy extension from the Visual Studio Code Marketplace: Open VS Code and click the Extensions button on the far left (it looks like 4 blocks with one flying away). This will open the Extension Marketplace. Search for "Malloy" and, once found, click "Install"

3. Connect to your database: Directions here.

4. Write some Malloy!: Start with the Quickstart. It may be helpful to check out one of the walkthroughs under Documentation below, or try some of the BigQuery sample models on public datasets available on the repo before getting started.

Join the Community

• Join the Malloy Slack Community! Use this community to ask questions, meet other Malloy users, and share ideas with one another.
• Use GitHub issues in this Repo to provide feedback, suggest improvements, report bugs, and start new discussions.

Documentation

Malloy Documentation

• Basics - A quick introduction to the language
• eCommerce Example Analysis - a walkthrough of the basics on an ecommerce dataset
• Flights Example Analysis - examples built on the NTSB flights public dataset
• Modeling Walkthrough - introduction to modeling via the Iowa liquor sales public data set

Why do we need another data language?

SQL is complete but ugly: everything is expressible, but nothing is reusable; simple ideas are complex to express; the language is verbose and lacks smart defaults. Malloy is immediately understandable by SQL users, and far easier to use and learn.

Key features and advantages:

• Query and model in the same language - everything is reusable and extensible.
• Malloy reads the schema so you don’t need to model everything. Malloy allows creation of re-usable metrics and logic, but there’s no need for boilerplate code that doesn’t add anything new.
• Pipelining: output one query into the next easily for powerful advanced analysis.
• Aggregating Subqueries let you build nested data sets to delve deeper into data quickly, and return complicated networks of data from single queries (like GraphQL).
• Queries do more: Power an entire dashboard with a single query. Nested queries are batched together, scanning the data only once.
• Indexes for unified suggest/search: Malloy automatically builds search indexes, making it easier to understand a dataset and filter values.
• Built to optimize the database: make the most of BigQuery, utilizing BI engine, caching, reading/writing nested datasets extremely fast, and more.
• Malloy models are purely about data; visualization and “styles” configurations live separately, keeping the model clean and easy to read.
• Aggregates are safe and accurate: Malloy generates distinct keys when they’re needed to ensure it never fans out your data.
• Nested tables are made approachable: you don’t have to model or flatten them; specify a query path and Malloy handles the rest.
• Compiler-based error checking: Malloy understands sql expressions so the compiler catches errors as you write, before the query is run.

Contributing

If you would like to work on Malloy, you can find some helpful instructions about developing Malloy and developing documentation.

Malloy is not an officially supported Google product.