| Crates.io | bitreq |
| lib.rs | bitreq |
| version | 0.3.1 |
| created_at | 2025-10-21 21:53:54.712439+00 |
| updated_at | 2026-01-19 22:34:09.907352+00 |
| description | Simple, minimal-dependency HTTP client |
| homepage | |
| repository | https://github.com/rust-bitcoin/corepc |
| max_upload_size | |
| id | 1894506 |
| size | 192,220 |
This crate is a fork for the very nice minreq. I chose to fork and rename it because I wanted to totally gut it and provide a crate with different goals. Many thanks to the original author.
Simple, minimal-dependency HTTP client. Optional features for http
proxies (proxy), async support (async, async-https), and https
with various TLS implementations (https-rustls, https-rustls-probe,
and https which is an alias for https-rustls).
Without any optional features, my casual testing indicates about 100
KB additional executable size for stripped release builds using this
crate. Compiled with rustc 1.45.2, println!("Hello, World!"); is 239
KB on my machine, where the hello example is 347
KB. Both are pure Rust, so aside from libc, everything is statically
linked.
Note: some of the dependencies of this crate (especially the various
https libraries) are a lot more complicated than this library, and
their impact on executable size reflects that.
Build your own with cargo doc --all-features, or browse the online
documentation at docs.rs/bitreq.
If you don't care about the MSRV, you can ignore this section entirely, including the commands instructed.
We use an MSRV per major release, i.e., with a new major release we reserve the right to change the MSRV.
The current major version of this library should always compile with
default features (i.e., std) on Rust 1.75. Other features may
require a higher MSRV.
This crate is distributed under the terms of the ISC license.
This is a list of features I'll implement once it gets long enough, or a severe enough issue is found that there's good reason to make a major version bump.
http:// or https://.Just thinking out loud, might not end up doing some or all of these.
#![no_std], maybe even exposing a
less convenient API for #![no_std] usage. Keep the current API as
in any case (at the very least, as a default feature).