A test runner that builds tests with rustc or cargo (or any other compiler with some configuration effort) and compares the output of the compiler with a file that you check into git. This allows you to test how your libraries show up to your users when the library is used wrongly and emits errors.

Usage

See examples directory for how to use this in your own crate. To be able to use it with cargo test, you need to put

[[test]]
name = "your_test_file"
harness = false

into your Cargo.toml, otherwise cargo test will only look for #[test]s and not run your fn main() that actually executes ui_test

Implicit (and possibly surprising) behavior

• Tests are run in order of their filenames (files first, then recursing into folders). So if you have any slow tests, prepend them with a small integral number to make them get run first, taking advantage of parallelism as much as possible (instead of waiting for the slow tests at the end).
• cargo test --test your_test_name -- --help lists the commands you can specify for filtering, blessing and making your tests less verbose.
  • Since cargo test on its own runs all tests, using cargo test -- --check will not work on its own, but cargo test -- --quiet and cargo test -- some_test_name will work just fine, as the CLI matches.
• if there is a .stdin file with the same filename as your test, it will be piped as standard input to your program.

Supported comment annotations

If your test tests for failure, you need to add a //~ annotation where the error is happening to ensure that the test will always keep failing at the annotated line. This will match any substring of the error. These comments can take two forms:

• //~ LEVEL: XXX matches by error level and message text
  • LEVEL can be one of the following (descending order): ERROR, HELP, WARN, NOTE or ICE
  • If a level is specified explicitly, all diagnostics of that level or higher need an annotation. To avoid this see //@require-annotations-for-level
  • This checks the output before normalization, so you can check things that get normalized away, but need to be careful not to accidentally have a pattern that differs between platforms.
  • if XXX is of the form /XXX/ it is treated as a regex instead of a substring and will succeed if the regex matches.
• //~ CODE matches by diagnostic code.
  • CODE can take multiple forms such as: E####, lint_name, tool::lint_name.
  • This will only match a diagnostic at the ERROR level.

In order to change how a single test is tested, you can add various //@ comments to the test. Any other comments will be ignored, and all //@ comments must be formatted precisely as their command specifies, or the test will fail without even being run.

• //@ignore-C avoids running the test when condition C is met.

  • C can be target: XXX YYY, which checks whether the target triple contains XXX or YYY.
  • C can be host: XXX YYY, which checks whether the host triple contains XXX or YYY.
  • C can also be bitwidth: followed by one or more space separated integer size like 64, 32 or 16.
  • C can also be on-host, which will only run the test during cross compilation testing.

• //@only-C only runs the test when condition C is met. The conditions are the same as with ignore.

• //@needs-asm-support only runs the test when the target supports asm!.

• //@stderr-per-bitwidth produces one stderr file per bitwidth, as they may differ significantly sometimes

• //@error-in-other-file: XXX can be used to check for errors that can't have //~ patterns due to being reported in other files.

• //@revisions: XXX YYY runs the test once for each space separated name in the list

  • emits one stderr file per revision
  • //~ comments can be restricted to specific revisions by adding the revision name after the ~ in square brackets: //~[XXX]
  • //@ comments can be restricted to specific revisions by adding the revision name after the @ in square brackets: //@[XXX]
    • Note that you cannot add revisions to the revisions command.

  For example:

  //@revisions: edition2021 edition2024
  //@[edition2021] edition:2021
  //@[edition2024] edition:2024
  //@[edition2024] check-pass
  
  // Then in code:
  let x = 12; //~[edition2021] dead_code
  let x = 12;
  //~[edition2021]^ dead_code

• //@compile-flags: XXX appends XXX to the command line arguments passed to the rustc driver

  • you can specify this multiple times, and all the flags will accumulate

• //@rustc-env: XXX=YYY sets the env var XXX to YYY for the rustc driver execution.

  • for Miri these env vars are used during compilation via rustc and during the emulation of the program
  • you can specify this multiple times, accumulating all the env vars

• //@normalize-stdout-test: "REGEX" -> "REPLACEMENT" / //@normalize-stderr-test: "REGEX" -> "REPLACEMENT" replaces all matches of REGEX in the stdout/stderr with REPLACEMENT. The replacement may specify $1 and similar backreferences to paste captures. Note that this doesn't apply to error strings specified by //~ comments.

  • you can specify multiple such commands, there is no need to create a single regex that handles multiple replacements that you want to perform.

• //@require-annotations-for-level: LEVEL can be used to change the level of diagnostics that require a corresponding annotation.

  • this is only useful if there are any annotations like HELP, WARN or NOTE, as these would automatically require annotations for all other diagnostics of the same or higher level.

• //@check-pass requires that a test has no error annotations, emits no errors, and exits successfully with exit/status code 0.

• //@edition: EDITION overwrites the default edition (2021) to the given edition.

• //@no-rustfix do not run rustfix on tests that have machine applicable suggestions.

• //@aux-build: filename looks for a file in the auxiliary directory (within the directory of the test), compiles it as a library and links the current crate against it. This allows you import the crate with extern crate or just via use statements. This will automatically detect aux files that are proc macros and build them as proc macros.

• //@run compiles the test and runs the resulting binary. The resulting binary must exit successfully. Stdout and stderr are taken from the resulting binary. Any warnings during compilation are ignored.

  • You can also specify a different exit code/status that is expected via e.g. //@run: 1 or //@run: 101 (the latter is the standard Rust exit code for panics).
  • run tests collect the run output into .run.stderr and .run.stdout respectively.
  • if a .run.stdin file exists, it will be piped as standard input to your test's execution.

Significant differences to compiletest-rs

• target selection is done via ignore-target: xxx and only-target: xxx instead of compiletest's ignore-xxx/only-xxx. The xxx must also be a substring of the target triple; special collections such as macos/unix in compiletest is not supported.
• only supports ui tests
• tests are run in named order, so you can prefix slow tests with 0 in order to make them get run first
• aux-builds require specifying nested aux builds explicitly and will not allow you to reference sibling aux-builds' artifacts.