repository·master·Indexed 12 days ago
https://github.com/rust-lang/rust-analyzerA high-performance language server for Rust that implements the Language Server Protocol (LSP) to provide IDE features such as code completion, refactoring, diagnostics, and navigation. It integrates with tools like rustfmt, rustc, and clippy, and is compatible with editors including VS Code, Vim, Emacs, and Zed.
rust-analyzer is a language server that provides IDE functionality for writing Rust programs. It implements the Language Server Protocol (LSP), allowing it to be used with any compatible editor such as VS Code, Vim, Emacs, or Zed.
Key features include:
rustfmt) and integrated diagnostics (via rustc and clippy).rust-analyzer provides a comprehensive suite of IDE features for Rust development, including:
rust-analyzer provides several core features to enhance your development experience:
The text-size library provides type-safe newtype wrappers for representing text offsets. Instead of using raw u32 or tuples, use these wrappers to distinguish between different types of text positions (e.g., byte offsets vs. character offsets) in your code.
It provides wrappers for:
u32: Typically used for a single offset (like a byte offset).(u32, u32): Typically used for a range or a specific position (like line and column).A SmolStr is an immutable string type designed for efficient storage of small strings, such as tokens in a programming language. It is optimized to avoid heap allocations in several common scenarios:
WS pattern) are also stack-allocated.SmolStr can be created from a &'static str without any allocation.SmolStr is an $O(1)$ operation because it does not require a new allocation.Memory Footprint: On 64-bit platforms, size_of::<SmolStr>() is 24 bytes, which is identical to size_of::<String>().
Note: If a string is longer than 23 bytes and does not follow the WS pattern, it will be heap-allocated. For use cases requiring heavy deduplication, a specialized interner may be more efficient.
rust-analyzer assumes that all code in a project is trusted. This means that opening a project in an IDE using rust-analyzer can lead to the execution of arbitrary code through several mechanisms. Users should be cautious when opening untrusted repositories.By default, rust-analyzer does not perform network access itself. However, it executes cargo metadata to resolve project dependencies, which may trigger the downloading or updating of the crate registry and dependency source code.
Additionally, if enabled (which is the default behavior), build scripts and procedural macros executed during the analysis process have full system permissions and can perform network access.
rust-analyzer server does not control the behavior of other editor plugins. If you are using additional extensions alongside rust-analyzer, you should consult their respective developers regarding their network access and privacy policies.rust-analyzer uses a client-server architecture via RPC (over stdio) to handle procedural macro expansion. This separation serves two purposes:
rustc versions and require matching internal APIs. By shipping this binary as a rustup component, the server can always match the current rustc version regardless of the rust-analyzer version being used.rust-analyzer has partial support for VS Code Live Share:
Crates located in the lib/ directory are published to crates.io and follow semantic versioning (semver). To release a new version of a crate:
Cargo.toml file.master, an automated workflow will publish the new version to crates.io.Note: You must release a new version to crates.io before a new API can be used by the main rust-analyzer project, as it relies on the published versions.