perf(shuffle): use the Fisher-Yates algorithm

[eumkz]

Tip

The owner of this PR can publish a preview release by commenting /publish in this PR. Afterwards, anyone can try it out by running pnpm add radashi@pr<PR_NUMBER>.

Summary

This PR replaces the current implementation of the shuffle function with a more efficient version using the Fisher-Yates algorithm.

Key changes:

1. Switched from a functional to an imperative approach for better performance.
2. Implemented Fisher-Yates shuffle algorithm for in-place shuffling.
3. Added option to pass custom random function, defaulting to Radashi's random function.
4. Ensured original array is not mutated by creating a copy.

Performance implications:

• Improved time complexity from O(n log n) to O(n).
• Reduced memory usage by avoiding creation of intermediate arrays.
• Potentially faster for larger arrays due to fewer object creations and method calls.

Testing:

• Existing tests should pass without modification.
• Added new tests for custom random function and array immutability.

Note: This change may impact the specific shuffle results but maintains the same randomness guarantees.

Related issue, if any:

For any code change,

• Related documentation has been updated, if needed
• Related tests have been added or updated, if needed
• Related benchmarks have been added or updated, if needed

Does this PR introduce a breaking change?

No

[Minhir]

That is nice!

[aleclarson]

Looks good, thanks @eumkz. 👍

[github-actions]

A new beta version 12.2.0-beta.85fb266 has been published to NPM. 🚀

To install:

pnpm add radashi@12.2.0-beta.85fb266

The radashi@beta tag also includes this PR.