The Rust team is happy to announce the latest version of Rust, 1.7. Rust is a systems programming language focused on safety, speed, and concurrency.

As always, you can install Rust 1.7 from the appropriate page on our website, and check out the detailed release notes for 1.7 on GitHub. About 1300 patches were landed in this release.

What's in 1.7 stable

This release is primarily about library features. While we have several language features cooking for future releases, the timeframe in which 1.7 was developed included the holidays, which means less time for commenting on GitHub and more time for spending with loved ones.

Library stabilizations

About 40 library functions and methods are now stable in 1.7. One of the largest APIs stabilized was support for custom hash algorithms in the standard library’s HashMap<K, V> type. Previously all hash maps would use SipHash as the hashing algorithm, which provides protection against DOS attacks by default. SipHash, however, is not very fast at hashing small keys. As shown, however, the FNV hash algorithm is much faster for these size of inputs. This means that by switching hash algorithms for types like HashMap<usize, V> there can be a significant speedup so long as the loss of DOS protection is acceptable.

To see this in action, you can check out the fnv crate on crates.io and create a HashMap via:

extern crate fnv; use std::collections::HashMap; use std::hash::BuildHasherDefault; use fnv::FnvHasher; type MyHasher = BuildHasherDefault<FnvHasher>; fn main() { let mut map: HashMap<_, _, MyHasher> = HashMap::default(); map.insert(1, "Hello"); map.insert(2, ", world!"); println!("{:?}", map); }

Note that most of the time you don’t even need to specify the hasher as type inference will take care of it, so HashMap::default() should be all you need to get up to 2x faster hashes. It’s also worth pointing out that Hash trait is agnostic to the hashing algorithm used, so no changes are needed to the types being inserted into hash maps to reap the benefits!

Other notable improvements include:

<[T]>::clone_from_slice() , an efficient way to copy the data from one slice and put it into another slice.

, an efficient way to copy the data from one slice and put it into another slice. Various convenience methods on Ipv4Addr and Ipv6Addr , such as is_loopback() , which returns true or false if the address is a loopback address according to RFC 6890.

and , such as , which returns or if the address is a loopback address according to RFC 6890. Various improvements to CString , used for FFI.

, used for FFI. checked, saturated, and overflowing operations for various numeric types. These aren’t counted in that ‘40’ number above, because there are a lot of them, but they all do the same thing.

See the detailed release notes for more.

Cargo features

There were a few small updates to Cargo:

An improvement to build scripts that allows them to precisely inform Cargo about dependencies to ensure that they’re only rerun when those files change. This should help development quite a bit in repositories with build scripts.

A modification to the cargo rustc subcommand, which allows specifying profiles to pull in dev-dependencies during testing and such.

Contributors to 1.7

We had 144 individuals contribute to 1.7. Thank you so much!