fast-cat

An ergonomic macro for highly efficient string concatenation with a single memory allocation.

Why use fast-cat?

Standard methods like format! or using the + operator can be inefficient for joining multiple strings, often leading to several memory re-allocations. This crate provides a concat_str! macro that solves this problem by first calculating the total size of all string parts and then performing only a single memory allocation.

It gives you the performance of a manually optimized String::with_capacity approach but with a much cleaner and more convenient syntax.

Features

• ⚡ Blazingly Fast: Significantly outperforms format! and + for string concatenation (see benchmarks below).
• ⚙️ Single Allocation: Avoids intermediate allocations, reducing memory pressure and improving performance.
• 🕊️ Zero Dependencies: This crate has no dependencies, which means fast compile times and no extra bloat for your project.
• ✍️ Ergonomic: Clean, intuitive syntax that is easy to read and write.

Installation

Add this to your Cargo.toml:

[dependencies]
fast-cat = "0.1.0" # Replace with the latest version

Usage

The concat_str! macro accepts any number of comma-separated arguments that can be referenced as a &str, including string literals, String variables, and &str slices.

use fast_cat::concat_str;

fn main() {
    let world = "world".to_string();
    let excitement = "!";

    // Mix and match literals, owned Strings, and slices
    let greeting = concat_str!("Hello, ", &world, excitement);

    assert_eq!(greeting, "Hello, world!");
}

Benchmarks

Benchmarks show that concat_str! is not only significantly faster than standard library alternatives but also manages to outperform a hand-written, manually optimized implementation.

The test scenario involved concatenating 6 string parts of mixed types (&str and String).

Results were obtained on a specific machine and may vary. The relative performance is the key takeaway.

As the data shows, concat_str! provides the best performance, making it an ideal choice for performance-sensitive applications.

License

This project is licensed under either of

• Apache License, Version 2.0, (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.