#104 with Rust
How to write and organise tests for rust.
Notes
Rust has first class testing support built into the language and toolchain. It even has a nascent style guide for testing.
Test code is identified with a set of test attributes
#[test]marks a function as a test#[ignore]disables a test function#[should_panic]indicates a test is expected to generate a panic#[cfg(test)]is used to mark a testing module for conditional compilation when running in test mode
The test mode is enabled by passing the --test argument to rustc or using cargo test.
Assertions are supported with a number of macros including:
- assert!
- assert_eq!
- assert_ne!
- there are complementary
debug_*!forms that are only enabled in non optimized builds by default
Where to put test code:
- Unit tests: convention is to create a test module named tests in each file
- Integration tests: included in a
testsdirectory at the top level of the project directory, next tosrc - Doc Tests: executing documentation examples to make sure they still run
Other things to note about testing:
- Rust’s privacy rules allow you to test private functions
Testing a Library
$ cargo new tested_lib --lib
Created library `tested_lib` package
$ cd tested_lib/
Added a few examples/demonstrations to this project including:
- unit test module
- assertions
- expect panic in a test
- integration tests
- doc tests
Running the tests:
$ cargo test
Compiling tested_lib v0.1.0 (.../rust/testing/tested_lib)
Finished test [unoptimized + debuginfo] target(s) in 1.80s
Running target/debug/deps/tested_lib-bd99122334aecbd7
running 3 tests
test tests::add_private_works ... ok
test tests::add_two_works ... ok
test tests::failing_test ... ok
test result: ok. 3 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Running target/debug/deps/integration_test-86734f87d5dcd2ce
running 1 test
test it_adds_two ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Doc-tests tested_lib
running 1 test
test src/lib.rs - add_two (line 5) ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Testing an Application
Integration tests for binary creates face a problem: they can’t use functions from src/main.rs.
Solution:
- put functionality that needs integration testing in supporting library create so it can be tested
- unit test
src/main.rs, or just make it so trivial that it doesn’t need additional testing
$ cargo new tested_app
Created binary (application) `tested_app` package
$ cd tested_app/
$ cargo test
Compiling tested_app v0.1.0 (.../rust/testing/tested_app)
Finished test [unoptimized + debuginfo] target(s) in 1.76s
Running target/debug/deps/tested_app-6874ba1f641ef8b1
running 3 tests
test tests::add_private_works ... ok
test tests::add_two_works ... ok
test tests::failing_test ... ok
test result: ok. 3 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Running target/debug/deps/tested_app-893ff5312f83678e
running 0 tests
test result: ok. 0 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Running target/debug/deps/integration_test-8cd2670c773b8a51
running 1 test
test it_adds_two ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Doc-tests tested_app
running 1 test
test src/lib.rs - add_two (line 5) ... ok
test result: ok. 1 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
Running the app:
$ cargo run
Finished dev [unoptimized + debuginfo] target(s) in 0.01s
Running `target/debug/tested_app`
Adding 2 to 2 = 4
Credits and References
- Writing Automated Tests - The Rust Programming Language
- Rust basic attributes
- Rust test attributes