Skip to content
Commits on Sep 30, 2011
  1. add ColorTracePrinter to colorise test results

    committed
    - OK in green
    - FAILURE in red
    - Ignore in yellow
    - Warning in yellow
Commits on Mar 26, 2011
  1. @lukebayes
  2. @lukebayes
  3. @lukebayes

    Added performanceCount configuration variable to TextPrinter

    lukebayes committed
    Set default performanceCount to 10 so that test result times no longer push the results off screen in a normal terminal
  4. @lukebayes

    Added support for XMLPrinter

    lukebayes committed
    Extracted base class out of FlashBuilderPrinter and exposed as XMLPrinter
    Added tests for XMLPrinter output
    Added assertMatches assertion
  5. @lukebayes
  6. @lukebayes
  7. @lukebayes
Commits on Jan 19, 2011
  1. @lukebayes
Commits on Dec 28, 2010
  1. @benbjohnson

    Added assertion for the error message with new method,

    benbjohnson committed
    assertThrowsWithMessage().
Commits on Dec 4, 2010
  1. @robertpenner
Commits on Nov 9, 2010
  1. @robertpenner
  2. @robertpenner
  3. @robertpenner
  4. @robertpenner

    TextPrinter now exposes an option to hideLocalPaths from the stack tr…

    robertpenner committed
    …ace, which AsUnitRunner now uses.
    
    Made TextPrinter default constants more consistent.
Commits on Nov 8, 2010
  1. @robertpenner
Commits on Nov 7, 2010
  1. @robertpenner
  2. @robertpenner

    Remove the TestObserver empty marker interface and "observer" termino…

    robertpenner committed
    …logy, which duplicated the existing "listener" paradigm.
  3. @robertpenner
  4. @robertpenner
  5. @robertpenner
  6. @robertpenner

    Removed CallbackBridge because it was no different than a Result. Ren…

    robertpenner committed
    …amed CallbackBridge to ResultObserverTest. May merge with the older-style ResultTest later.
  7. @robertpenner

    Call stacks for failures are now much shorter. Now using a while loop…

    robertpenner committed
    … to iterate as far as possible before hitting async methods.
  8. @robertpenner

    Overhauled test injection in TestRunner to use the SwiftSuspenders in…

    robertpenner committed
    …jector.
    
    Removed support for injecting arbitrary properties from [Inject] metadata--it's not an established convention. [Inject] in frameworks like Robotlegs and Swiz is understood to support a "name" property, but that's to identify the injection rule; it's not a property to set on the instance.
  9. @robertpenner
  10. @robertpenner

    AsUnit 4 builds and runs in Flash CS3! (As long as we build everythin…

    robertpenner committed
    …g from source. SWC linkage was added in Flash CS4.)
  11. @robertpenner
  12. @robertpenner
  13. @robertpenner

    Small cleanups: AsyncTest' unused dispatcher, AllTests' references to…

    robertpenner committed
    … deleted tests, RunnerFactory formatting & Injector mapping.
  14. @robertpenner
  15. @robertpenner

    Changed SuiteRunnerTest back to use [Before] method to instantiate th…

    robertpenner committed
    …e unit under test. This necessitated restoring a way to pass an IResult to the SuiteRunner and verify it.
    
    SuiteRunner also now extends EventDispatcher to be consistent with the rest (AsUnitCore, TestRunner, etc.).
  16. @robertpenner

    AsUnitCore now extends EventDispatcher instead of delegating. Several…

    robertpenner committed
    … other classes were already doing that, might as well be consistent and have less code.
  17. @robertpenner
Commits on Nov 6, 2010
  1. @Stray

    Fixed hole in assertArraysEqualIgnoringOrder where the assert passed …

    Stray committed
    …incorrectly where the arrays contain repeated items, ie:
    
    [item1, item1, item2] equals [item1, item2, item2]
    
    These are now found to be not equal.
    
    Added tests for this to the asunit3 and asunit4 suites (which were failing before the change to Assert), and updated the Assert in both suites so that they now pass.
    
    Refactored logic for bothNull, eitherNull and differentLengths tests within asserts to use helper functions.  Refactored the null defense in assertEquals to give a cleaner early exit where expected is null.
Something went wrong with that request. Please try again.