Google releases Mangle: a programming language for deductive database programming

Google introduced it MangleThis is a new open source programming language that extends the classic logic-based language data of modern deductive database programming. Mangle is used as a GO library to simplify complex tasks about data scattered across multiple different sources.

This release has an increasing challenge for developers and security engineers: data fragmentation. In a modern software ecosystem, information about dependencies, vulnerabilities, configurations, and infrastructure is in a variety of formats and locations. Mangle provides a unified declarative framework to cohesively analyze this information.

The core of Mangle is the extension of Datalog, a declarative logic programming language whose roots lie in database theory. While traditional data is very powerful for expressing complex queries, it often lacks the functionality required by practical real-life applications. Mangle bridges this gap by introducing multiple key extensions while aiming to keep its predecessor accessible and simplicity.

Key features and extensions

Mangle enhances traditional data with features that are critical to modern development and security workflows:

• Recursive rules: Mangle’s flag is a Datalog’s flag, fully supporting recursive rules. This allows developers to express pass-through relationships gracefully, which are common in tasks that track a complete dependency tree of projects or draw access through a hierarchy.
• Unified data access: Mangle’s main advantage is its ability to treat multiple data sources as a single logical database. It can ingest facts from files, APIs, or other datastores, allowing developers to write queries that seamlessly connect information without having to worry about the underlying source.
• Aggregation and function calls: Beyond pure logic, Mangle combines practical extensions such as aggregation capabilities (e.g. count,,,,, sum) and the ability to call external functions. This allows for more complex computations and can be integrated with existing code bases, allowing developers to enrich their logical analysis using custom business logic.

Key practical applications

Mangle’s design makes it particularly suitable for several key areas:

1. Vulnerability Testing: Security teams can use Mangle to model security policies and code bases into a set of facts and rules. For example, one could write a rule that the project depends on a library with a known CVE. Mangle can then recursively check the entire dependency graph to mark the affected items.
2. Software dependency analysis: Mangle is naturally choosing the complexity of modern software supply chains. It can be used to analyze software bill of materials (SBOM), execute version policies in an organization, or identify items affected by deprecated libraries.
3. Knowledge graph modeling: The language provides powerful tools for building and querying knowledge graphs. By representing entities and their relationships as logical facts, organizations can use Mangle to discover hidden connections and perform complex inferences on large interconnected datasets.

Implementation and developer accessibility

By implementing Mangle as a GO library, Google has ensured it can be easily embedded into a wide range of existing applications and analytics tools. This approach avoids the overhead of independent database systems and places powerful deductive query capabilities directly in the hands of developers. The project’s documentation emphasizes practicality and is intended to enable logic-based programming to be accessible to developers who may be more familiar with command or object-oriented paradigms.

The release of Mangle provides a powerful new tool for any developer, SRE or security professional whose task is to understand complex, distributed information. By combining the declarative elegance of Datalog with the practical capabilities required by modern software, Google provides a solution that simplifies everything from security analytics to infrastructure management.

Check Github page. Check out ours anytime Tutorials, codes and notebooks for github pages. Also, please stay tuned for us twitter And don’t forget to join us 100K+ ml reddit And subscribe Our newsletter.