mantra-rust-macros

This crate provides procedural macros to trace requirements using mantra. It also offers features to get requirement coverage via logs.

Traces

Requirement traces may be created using either the attribute macro req, or the function-like macro reqcov. See mantra-lang-tracing on how to specify requirement IDs using req or reqcov.

The attribute macro may be set anywhere attribute macros are allowed. At the moment, coverage logs are only generated if the macro is set on a function.

Examples:

use mantra_rust_macros::{req, reqcov};

/// Coverage log is generated
#[req(fn_trace)]
fn some_fn() {
    // coverage log is generated
    reqcov!(function_like_trace);
}

#[req(const_trace)]
const SOME_CONST: usize = 1;

#[req(type_trace)]
type SomeType = bool;

#[req(struct_trace)]
struct SomeStruct {
    /// Attribute macros cannot be set for fields.
    some_field: bool,
}

#[req(mod_trace)]
mod some_mod {}

#[req(trait_trace)]
trait SomeTrait {
    #[req(trait_type_trace)]
    type A;

    /// Coverage log is generated
    #[req(trait_fn_default_trace)]
    fn some_trait_fn() {}
}

Coverage log

• Feature log

  Enabling this feature will create coverage TRACE logs using the log crate.

• Feature stdout

  Enabling this feature will print coverage logs to stdout.

• Feature defmt

  Enabling this feature will print coverage logs using the defmt crate. This is intended for embedded devices.

Examples:

#[req(fn_trace)]
fn some_fn() {}

The generated coverage log for the code above has the form:

mantra: req-id=`fn_trace`; file='<resolved by file!()>'; line='<resolved by line!()>';

Trace extraction

The extract feature may be enabled to extract coverage data from logs. With the feature enabled, the functions extract_first_coverage() or extract_covered_reqs() may be used.