Mangle: Google's Bold Take on Database Programming

Google introduced Mangle, a new open-source programming language that extends the classic logic-based language Datalog for modern deductive databases.

Implemented as a Go package, Mangle is designed to make it easier to query and reason about data spread across multiple sources.

At its core, Mangle builds on Datalog, a declarative logic programming language with roots in database theory. While traditional Datalog is powerful for expressing complex queries, it often lacks features needed for real-world applications. Mangle bridges this gap by introducing practical extensions while keeping the accessibility and simplicity of its predecessor.

Key Features and Extensions

Mangle enhances Datalog with features that are essential for modern development, security, and data analysis:

Recursive Rules

A hallmark of Datalog, fully supported in Mangle. Recursive rules allow developers to express transitive relationships, like tracing a project's dependency tree or mapping access rights across a hierarchy.

Uniform Data Access

Mangle can treat multiple data sources as one logical database. It can pull facts from files, APIs, or databases, letting developers join information seamlessly without worrying about where it comes from.

Aggregation and Function Calls

Mangle supports aggregation functions (e.g., count, sum) and external function calls. This makes it easier to run computations and integrate logical queries with existing code or business logic.

Optional Type Checking

Unlike plain Datalog, Mangle gives developers the option to add type safety when needed. This helps balance flexibility with reliability, especially for larger projects.

Practical Applications

Mangle is not just a research toy—it's useful for real-world problems:

Vulnerability Detection

Security teams can write rules like: . Mangle can then recursively scan dependency graphs to flag issues. This makes it great for software supply chain security.