ktrianta/rust-callgraph-benchmark

A benchmark for Rust call-graph generators

rust-callgraph-benchmark's Issues

Conditionally compiled functions

Conditionally compiled functions might be overlooked by call graph generators.

Add few such examples to make the benchmark more complete.

More on conditional compilation

Macros

It has been observed that call graph generators may fail to analyse macros.

Currently, stdlib macros such as println! exist in the benchmark code. It would be nice to introduce a few of our own macros to the benchmark.

More on macros.

Generic Traits

Rust traits can be generic as describe in the docs.

The following example introduces the generic trait DoubleDrop.

// A trait generic over `T`.
trait DoubleDrop<T> {
    // Define a method on the caller type which takes an
    // additional single parameter `T` and does nothing with it.
    fn double_drop(self, _: T);
}

// Implement `DoubleDrop<T>` for any generic parameter `T` and caller `U`.
impl<T, U> DoubleDrop<T> for U {
    // This method takes ownership of both passed arguments,
    // deallocating both.
    fn double_drop(self, _: T) {}
}

As can be seen, DoubleDrop is implemented for any type U. Generic traits and traits implemented for generic callers are not currently covered by the benchmark.