Arios

Arios is a small HTTP client crate written in Rust.

The project started as a study exercise and is being refactored into a publishable library crate. The current focus is on manual HTTP request/response handling, response metadata parsing, and a small public API.

Status

Arios currently supports:

• GET and POST
• HTTP and HTTPS
• configurable Accept and Content-Type headers
• response metadata such as status code, content-type, charset, and content-length
• access to raw response bytes
• text decoding with built-in support for UTF-8, ISO-8859-1, and US-ASCII
• UTF-8 fallback when no charset is provided or when the charset is not supported

Current limitations:

• the API is still evolving
• URL parsing is intentionally simple
• there is no custom crate error type yet
• advanced HTTP features are still limited

Installation

Add the crate to your Cargo.toml:

[dependencies]
arios = "0.1.2"

Example: GET

use arios::{Arios, ContentType};

fn main() -> std::io::Result<()> {
    let client = Arios::create("https://www.google.com")?;
    let response = client.get(ContentType::Html)?;

    println!("status: {}", response.code);
    println!("content-type: {:?}", response.content_type);
    println!("body size: {}", response.bytes().len());

    Ok(())
}

Example: POST

use arios::{Arios, ContentType};

fn main() -> std::io::Result<()> {
    let client = Arios::create("https://example.com/api")?;
    let body = r#"{"name":"arios"}"#;

    let response = client.post(body, ContentType::Json, ContentType::Json)?;

    println!("status: {}", response.code);
    println!("{}", response.text()?);

    Ok(())
}

Response API

AriosResponse exposes:

• bytes() for raw response bytes
• text() for decoded textual content
• public metadata fields such as code, status, content_type, charset, and content_length

text() currently supports utf-8, iso-8859-1, and us-ascii. If the response does not declare a charset, or declares one that Arios does not support yet, decoding falls back to UTF-8.

Goals

The project is being built as a learning-focused HTTP client with crate-oriented organization. The immediate goals are:

• keep the public API small and understandable
• improve test coverage
• continue reducing protocol parsing edge cases
• prepare the crate for publication