Prefix Matching (#2947)

WARNING

All experimental features are subject to breaking changes and/or removal at any time. We strongly recommend that you do not use these features in a production environment. They are intended for testing and feedback only.

INFO

To enable this experiment, set the environment variable: TASK_X_PREFIX_MATCHING=1 or configure it in .taskrc.yml:

experiments:
  PREFIX_MATCHING: 1

Check out our guide to enabling experiments for more information.

This experiment adds support for Shortest Unique Prefix Matching (segment-wise abbreviation matching) when running tasks.

In large Taskfiles with multi-level namespaces (for example, api:openapi:export), typing the full task name or manually defining short aliases for each task can be tedious.

How It Works

1. Unique Match: When an input prefix uniquely matches a task (or one of its aliases), Task will execute it immediately.

   • task a:o:e matches api:openapi:export
   • task api:o matches api:openapi:export (if no other api:o* tasks exist)
   • task b matches build (if no other b* tasks exist)

2. Ambiguous Match: If the prefix matches multiple tasks, Task will halt execution and return a conflict error listing all candidate tasks:

   task: Found multiple tasks (docker:build:production, docker:build:staging) that match "doc"

3. Precedence: Direct task name matches, alias matches, and wildcard matches always take precedence over prefix matching.