Important TL;DR:
If you are upgrading: 1.) Test methods must start with the prefix "test" and 2.) If you are using LUnit add-ons, you need to upgrade these also.

This release (v1.9.x) contains the following improvements and new features:
- Change: test methods must start with the prefix "test". This reverts the update in v1.8.x, where this requirement was removed. The reason for this is an improvement which significantly reduces test discovery time.
- Change: Internal Runnable.lvclass interface updated. This may require add-ons using these interfaces to update for compatibility.
- Feature: Introduce diff tool which highlights difference between expected and actual result.
- Improvement: Performance improvements to test discovery, test loading and UI loading
- Improvement: Result comparison message calculated on demand and not during test execution.
- Fix: comparison of equal typedeffed datatypes could in edge cases show as unequal
- Fix: show description when comparing 2D arrays
- Fix: minor typos

antsundq was a contributor to this release

This release (v1.8.17) contains the following improvements and new features:
- Fix:: Config file moved under ~/.astemes on linux to prevent access right issues

antsundq and j-medland were contributors to this release

This release (v1.8.x) contains the following improvements and new features:
- Tests are no longer required to have the "test" prefix in the name. Now the only requirement is that the VI is public. If you have utility vi:s belonging to a test class , the access scope should be set to either private or protected.
- Test vi:s implemented as Dynamic Dispatch (not recommended) still require the "test" prefix.
- Test Inheritance features migrated to own add-on package
- Fix error at comparison of numeric arrays when the tolerance input is wired
- Add result boolean indicator to API
- Improved test index cache invalidation

antsundq was a contributor to this release

This release (v1.8.x) contains the following improvements and new features:
- Tests are no longer required to have the "test" prefix in the name. Now the only requirement is that the VI is public. If you have utility vi:s belonging to a test class , the access scope should be set to either private or protected.
- Test vi:s implemented as Dynamic Dispatch (not recommended) still require the "test" prefix.
- Test Inheritance features migrated to own add-on package
- Fix error at comparison of numeric arrays when the tolerance input is wired
- Add result boolean indicator to API
- Improved test index cache invalidation

antsundq was a contributor to this release

This release (v1.7.x) contains the following improvements and new features:
- Add default banner to test case classes when created
- Cosmetic improvements to report generation dialog
- Add documented support for third party reporting extenssions
- Minor devex improvements
- Add "verify" as valid test method prefix in addition to "test"
- Allow for test names containing . and +
- Highlight BD when opening test from UI
- Add example for organizing tests in nested projects
- Minor bugfixes

antsundq was a contributor to this release

This release (v1.6.x) contains the following improvements and new features:
- Default to open block diagram only when opening a test case from the Execution UI.
- Add LUnit check mark overlay to Test Case classes in the LabVIEW Project Explorer to make it easier to distinguise test cases from normal classes
- Show test result overlays for test methods
- Performance improvements when running tests in parallell. Test execution overhead reduced significantly for large test suites.
- Added Pass/Fail if Matching String advanced assertion

antsundq was a contributor to this release

This release (v1.5.x) contains the following improvements:
- Revolutionary new way for running a test: Just run the test VI using the run-arrow. This will execute the test, open the user interface, and show the results. This is similar to how tests are executed in Caraya.
- Bugfix: Retain earlier results when rediscovering tests in UI

antsundq was a contributor to this release

This release (v1.4.x) contains the following improvements:
- The command line interface has been separated into its own package (LUnit CLI)*
- "Run All" now executes only failed tests on second run. If no tests have failed, all tests are run.
- Assertions added testing for specific error code
- Assertions added for testing if element in array/set/map
- Support for test discovery in auto-populating folders
- Updated Templates
- Improved clarity in failure messages comparing different types of numerics
- Test inheritance feature moved to sub-class of Test Case

* The reason for not including it in the main package is becuase installing the CLI requires write access to C:\Program Files (x86)\National Instruments\Shared\LabVIEW CLI. This is not granted by default and caused bad user experience as the installation would return an error. The CLI is typically not needed in a normal installation.

antsundq , sas_workshops and gt_thetic were contributors to this release

This release contains the following improvements:
- New feature: Ignore tests by adding an underscore "_" as the first character in the test name. These test may be toggled on and off using the File menu option. For further details, see [the documentation](https://astemes.github.io/astemes-lunit/10_Basics/)
- Performance: Improves the performance of test discovery when running tests using the API. Thanks to Sam Taggart for contributions to this improvement.
- Bugfix: Errors during test execution now appears in test description

sas_workshops and antsundq were contributors to this release

Version 1.2 is a complete rewrite of LUnit with feature parity to earlier versions. The rewrite was done using test first test driven development and as a consequence the code base is now smaller, less complex, significantly more performant and more maintainable.

For users of LUnit, very little has changed functionally. The performance and responsiveness have been improved through the rewrite. The test scaffolding additionally provides a foundation for further improving on performance and useability. Please contribute ideas.

The reason for the increment of the minor version is a change to the API for running test programmatically. The old API is still available, with one minor change. All results are now returned in one event with an enum indicating the type of result. If someone needs to use the legacy format, it is straight forward to develop an adapter, please reach out for support if this is a concern.

Version 1.2 is a complete rewrite of LUnit with feature parity to earlier versions. The rewrite was done using test first test driven development and as a consequence the code base is now smaller, less complex, significantly more performant and more maintainable.

For users of LUnit, very little has changed functionally. The performance and responsiveness have been improved through the rewrite. The test scaffolding additionally provides a foundation for further improving on performance and useability. Please contribute ideas.

The reason for the increment of the minor version is a change to the API for running test programmatically. The old API is still available, with one minor change. All results are now returned in one event with an enum indicating the type of result. If someone needs to use the legacy format, it is straight forward to develop an adapter, please reach out for support if this is a concern.

Version 1.2 is a complete rewrite of LUnit with feature parity to earlier versions. The rewrite was done using test first test driven development and as a consequence the code base is now smaller, less complex, significantly more performant and more maintainable.

For users of LUnit, very little has changed functionally. The performance and responsiveness have been improved through the rewrite. The test scaffolding additionally provides a foundation for further improving on performance and useability. Please contribute ideas.

The reason for the increment of the minor version is a change to the API for running test programmatically. The old API is still available, with one minor change. All results are now returned in one event with an enum indicating the type of result. If someone needs to use the legacy format, it is straight forward to develop an adapter, please reach out for support if this is a concern.

Version 1.2 is a complete rewrite of LUnit with feature parity to earlier versions. The rewrite was done using test first test driven development and as a consequence the code base is now smaller, less complex, significantly more performant and more maintainable.

For users of LUnit, very little has changed functionally. The performance and responsiveness have been improved through the rewrite. The test scaffolding additionally provides a foundation for further improving on performance and useability. Please contribute ideas.

The reason for the increment of the minor version is a change to the API for running test programmatically. The old API is still available, with one minor change. All results are now returned in one event with an enum indicating the type of result. If someone needs to use the legacy format, it is straight forward to develop an adapter, please reach out for support if this is a concern.

- Add buttons to LabVIEW Toolbar for running all tests in project and for creating new test case
- Add option to ignore inherited tests
- Add tool for measuring code coverage.
- Add support for organizing tests in nested LabVIEW Project files.
- Internal implementation for test results updated from class based to regular type defs. The only way to access these classes is through the API, so the impact on users is limited.
- Minor useability and performance improvements.

antsundq was a contributor to this release