Vacro

The Philosophy

Writing procedural macros in Rust shouldn't be a nightmare filled with boilerplate code and black-box debugging.

Vacro has evolved from a simple parsing library into a complete toolchain designed to improve the Developer Experience (DevX) across the entire macro lifecycle:

1. Parsing: Write parsing logic in a declarative way.
2. Debugging: Visualize the parsing path to see exactly what happens inside the macro.
3. Reporting: Easily generate elegant and precise compiler error messages.

The Ecosystem

Vacro is designed as a modular framework. You can use the fully-featured vacro entry point or pick specific underlying components as needed.

Quick Start

Add vacro to your Cargo.toml and enable the DevX features you need:

[dependencies]
vacro = { version = "0.2", features = ["full"] }

1. Declarative Parsing (vacro-parser)

Define your macro's input grammar like writing regex:

use vacro::prelude::*;

// Define syntax: "fn" <name> "(" <args> ")"
vacro::define!(MyMacroInput:
    fn
    #(name: syn::Ident)
    ( #(args*[,]: syn::Type) )
);

See more: vacro-parser

2. Visual Debugging (vacro-trace)

Take snapshots of your TokenStream to see how it evolves. View the diffs in vacro-cli.

use vacro::prelude::*;

// Capture a snapshot with a tag.
// If called multiple times with the same tag, vacro-cli will show the diff.
vacro::snapshot!("expand", tokens);

See more: vacro-trace

3. Diagnostic Reporting (vacro-report)

Provides superior error reporting capabilities, saying goodbye to generic unexpected identifier errors.

use vacro::prelude::*;

#[vacro::report::scope]
fn my_macro_impl(input: TokenStream) -> TokenStream {
    // If this fails (e.g., invalid syntax constructed),
    // Vacro will catch it and emit a precise error pointing to the code.
    let f: ItemFn = parse_quote!( fn foo () { >>invalid<< } );
    quote!(#f)
}

See more: vacro-report

4. Visualization Tool (vacro-cli)

Install and run the TUI tool to view trace data and snapshot diffs.

cargo install vacro-cli
# 1. Run tests to generate trace data
cargo test
# 2. Launch the visualizer
cargo vacro

Run the following test, then open the CLI to inspect the captured logs and snapshot evolution:

#[test]
#[instrument]
fn test_function() {
    // 1. Log
    info!("Function started");
    warn!("This is a warning");
    error!("This is an error");

    // 2. Snapshot
    // Initial state
    let code_snippet = quote! { x: i32 };
    snapshot!("Field", code_snippet);

    // State change: Wrap in a struct
    // vacro-cli will automatically diff multiple snapshots with the "Struct" tag
    let code_snippet = quote! { struct A { #code_snippet }};
    snapshot!("Struct", code_snippet);

    // State change: Add derive
    let code_snippet = quote! {
        #[derive(Debug)]
        #code_snippet
    };
    snapshot!("Struct", code_snippet);

    let x = 1 + 1;
    debug!("Calculation result: {}", x);
}

See more: vacro-cli

Roadmap

We are currently in active development, transitioning towards a DevX Framework.

• Documentation: Multi-language support (vacro-doc-i18n).
• Parsing: Core DSL implementation (vacro-parser).
• Diagnostics: Error reporting integration (vacro-report).
• Debugging: Implementation of vacro-trace and vacro-cli.

License

Licensed under either of Apache License, Version 2.0 or MIT license at your option.