Skip to content

ROADMAP.md

Latest commit

 

History

History
81 lines (51 loc) · 4.75 KB

ROADMAP.md

File metadata and controls

81 lines (51 loc) · 4.75 KB
title Aria Roadmap
description Overall vision doc highlighting the goals for Aria.

Aria Roadmap

This document outlines the future direction of the Aria programming language. As an open-source project, we welcome community feedback and contributions to help shape the future of Aria. Ideas outside this roadmap are very welcome, and can be filed as GitHub issues to begin discussion.

This document is best understood alongside the Release Policy, which defines the criteria for versioning and releasing new versions of Aria.

Very-Short-Term Goal (Aria 1.0, by April 2026)

Aria is currently at version 0.9. While we intend not to break things for its own sake, this version indicates that the language is still under development. Up until version 1.0, we reserve the right to make breaking changes to the language and to the library. Ideally, we will only do this if the balance between the value of the change and the cost of breaking existing code is a net-positive.

After version 1.0 is released, we will only consider breaking changes on a must basis, i.e. we will only break existing code if there is no other realistic path to achieve a significant language goal.

Rewriting the Aria compiler or VM in Aria is a non-goal.

Bugs and improvements for Aria 1.0 are tracked on GitHub.

At this stage of Aria's development, most items under consideration for 1.0 are bugs. For an improvement not already planned to be added to the roadmap it would need to be:

  • significant enough that its exclusion would severely limit Aria's value to its users (e.g. if integer addition was missing, we would add it);
  • breaking enough that including it after 1.0 would widely affect existing Aria code (e.g. a change to Maybe's core API)

Improvements that don't meet this bar are most well suited to the 1.5 or 2.0 roadmaps.

Medium-Term Goals (Aria 1.5, by September 2026)

  • Improve the Core Language: Quality of life additions to the language itself, including:

    • per-case methods in enums;
    • char / byte builtin types;
    • improved pattern matching and destructuring;
    • try_unwrap helpers for enums;
    • Either enumeration;
    • multithreading.

  • Enhance the Standard Library: Add more utility modules and expand the functionality of existing ones, including:

    • server support for aria.network;
    • additional aria.fs.path helpers for filesystem manipulation;
    • better subprocess control;
    • crypto (as in, cryptography) support;
    • add more operations to aria.numerics.Complex.

Improvements to Aria 1.5 are tracked here.

Long-Term Goals (Aria 2.0, by June 2027)

  • Implement support for Reflection: Allow introspecting Aria values in a more structured and uniform manner.

  • Implement a Foreign Function Interface (FFI): Allow Aria to call directly into C APIs without writing a glue layer.

  • Filtered Exception Handling: Allow pattern matching in catch clauses.

  • Improvements to 3rd party library ecosystem: Make it easy to create an Aria library (possibly with a native binding) and ship it to users.

  • Create a Language Server Protocol (LSP) Implementation: Provide a better development experience in code editors with features like autocompletion, go-to-definition, and inline documentation.

  • Serializable Modules: Allow compiling Aria modules once and storing the output for faster execution. This is not intended to allow shipping binaries that cannot be reverse-engineered, its goal is purely speed of execution for unmodified programs.

Improvements to Aria 2.0 are tracked here.

Very-Long-Term Goals (Aria 2.x, second half of 2027 onwards)

  • Improve the import experience: Remedy shortcomings in the way import works, including:

    • exclude extensions from being imported based on importer decision;
    • exclude some symbols from being import based on module author decision (e.g. private impl details);
    • allow local imports within a function/type scope.

  • Expand Platform Support: Add support for Aria to run on Windows.

  • Implement a Debugger: Develop tooling to allow interactive debugging of Aria progams.

  • Allow GUI/TUI apps in Aria: Add support for UI toolkits (text and graphics based).

  • Fuzzing: Fuzz inputs to the Aria compiler and VM and fix resulting issues.

How to Contribute

We encourage you to get involved! Please see our Contribution Guide for more details on how you can help.