[concurrency-interest] Request for feedback -- Bill/Brian/Cliff JavaOne talk

Hanson Char hanson.char at gmail.com
Fri Apr 6 18:00:22 EDT 2007


On MTC, it mentions it uses the same ideas as JUnit to run initialize() if
it exists, run threadXxxx() methods concurrently, etc.

Well that's true for JUnit 3 but since JUnit 4 they've moved on to an
annotation driven approach.  So instead of relying on method naming
convention, they use annotation such as @Test, @BeforeCass, etc.

Ideally MTS could use the similar approach.  So there will be method
annotations like

@BeforeClass or @Before // (same as JUnit 4)
@Concurrent // to run methods concurrently
@Concurrent(threadPoolSize=3)  // to run methods concurrently with the
specified thread pool size
@After // to check result after the test (same as JUnit 4)

My 2 cents.

Hanson Char

On 4/5/07, Brian Goetz <brian at quiotix.com> wrote:
>
> Bill Pugh, Cliff Click, and myself are doing a talk at JavaOne on
> Testing Concurrent Code.
>
> I've attached a (commentable) PDF of the slides.  We'd love it if anyone
> had feedback, which you can send in text, or use Acrobat Reader to apply
> comments directly to the PDF.  (I think you can even save a .fdf file
> from Reader, which just saves the comments and which we can merge back
> into the PDF.)
>
> If you're going to comment, please do so by Monday -- as we're already
> weeks late getting our slides in!
>
> _______________________________________________
> Concurrency-interest mailing list
> Concurrency-interest at altair.cs.oswego.edu
> http://altair.cs.oswego.edu/mailman/listinfo/concurrency-interest
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20070406/61b5be8a/attachment.html 


More information about the Concurrency-interest mailing list