KEMBAR78
Testing - The Rust Reference

Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Testing attributes

The following attributes are used for specifying functions for performing tests. Compiling a crate in “test” mode enables building the test functions along with a test harness for executing the tests. Enabling the test mode also enables the test conditional compilation option.

The test attribute

The test attribute marks a function to be executed as a test.

Example

#![allow(unused)]
fn main() {
pub fn add(left: u64, right: u64) -> u64 { left + right }
#[test]
fn it_works() {
    let result = add(2, 2);
    assert_eq!(result, 4);
}
}

The test attribute uses the MetaWord syntax and thus does not take any inputs.

The test attribute may only be applied to free functions that are monomorphic, that take no arguments, and where the return type implements the Termination trait.

Note

Some of types that implement the Termination trait include:

  • ()
  • Result<T, E> where T: Termination, E: Debug

Only the first instance of test on a function is honored.

Note

Subsequent test attributes are currently ignored and rustc warns about these.

The test attribute is exported from the standard library prelude as std::prelude::v1::test.

These functions are only compiled when in test mode.

Note

The test mode is enabled by passing the --test argument to rustc or using cargo test.

The test harness calls the returned value’s report method, and classifies the test as passed or failed depending on whether the resulting ExitCode represents successful termination. In particular:

  • Tests that return () pass as long as they terminate and do not panic.
  • Tests that return a Result<(), E> pass as long as they return Ok(()).
  • Tests that return ExitCode::SUCCESS pass, and tests that return ExitCode::FAILURE fail.
  • Tests that do not terminate neither pass nor fail.

Example

#![allow(unused)]
fn main() {
use std::io;
fn setup_the_thing() -> io::Result<i32> { Ok(1) }
fn do_the_thing(s: &i32) -> io::Result<()> { Ok(()) }
#[test]
fn test_the_thing() -> io::Result<()> {
    let state = setup_the_thing()?; // expected to succeed
    do_the_thing(&state)?;          // expected to succeed
    Ok(())
}
}

The ignore attribute

The ignore attribute can be used with the test attribute to tell the test harness to not execute that function as a test.

Example

#![allow(unused)]
fn main() {
#[test]
#[ignore]
fn check_thing() {
    // …
}
}

Note

The rustc test harness supports the --include-ignored flag to force ignored tests to be run.

The ignore attribute uses either the MetaWord or MetaNameValueStr syntax.

The MetaNameValueStr form of the ignore attribute provides a way to specify a reason why the test is ignored.

Example

#![allow(unused)]
fn main() {
#[test]
#[ignore = "not yet implemented"]
fn mytest() {
    // …
}
}

The ignore attribute may be applied to functions annotated with the test attribute.

Note

rustc currently warns when ignore is used in some other situations. This may become an error in the future.

Only the first instance of ignore on a function is honored.

Note

rustc currently ignores duplicate ignore attributes. This may become an error in the future.

Ignored tests are still compiled when in test mode, but they are not executed.

The should_panic attribute

The should_panic attribute causes a test to pass only if the test function to which the attribute is applied panics.

Example

#![allow(unused)]
fn main() {
#[test]
#[should_panic(expected = "values don't match")]
fn mytest() {
    assert_eq!(1, 2, "values don't match");
}
}

The should_panic attribute has one of the following forms:

  • MetaWord

    Example

    #![allow(unused)]
fn main() {
#[test]
#[should_panic]
fn mytest() { panic!("error: some message, and more"); }
}

  • MetaNameValueStr — The given string must appear within the panic message for the test to pass.

    Example

    #![allow(unused)]
fn main() {
#[test]
#[should_panic = "some message"]
fn mytest() { panic!("error: some message, and more"); }
}

  • MetaListNameValueStr — As with the MetaNameValueStr syntax, the given string must appear within the panic message.

    Example

    #![allow(unused)]
fn main() {
#[test]
#[should_panic(expected = "some message")]
fn mytest() { panic!("error: some message, and more"); }
}

    Note

    rustc currently accepts the MetaListNameValueStr form with invalid syntax between the parentheses and emits a future-compatibility warning. This may become a hard error in the future.

The should_panic attribute may only be applied to functions annotated with the test attribute.

Note

rustc currently accepts this attribute in other positions with a warning. This may become a hard error in the future.

Only the first instance of should_panic on a function is honored.

Note

rustc currently ignores subsequent should_panic attributes and emits a future-compatibility warning. This may become a hard error in the future.

When the MetaNameValueStr form or the MetaListNameValueStr form with the expected key is used, the given string must appear somewhere within the panic message for the test to pass.

The return type of the test function must be ().