[Mono-devel-list] NUnit Regression test suites.

Nick Drochak ndrochak at gol.com
Fri Jan 14 09:38:01 EST 2005


SP,

>| -----Original Message-----
>| > I planned to go
>| > through the tests
>| > again and put [Ignore("Bugzilla 39932")] and such on tests that
>| > fail there.
>| 
>| I don't think using [Ignore] is a good idea because:
>| (a) the tests won't be executed under the MS runtime (with run-ondotnet).
>| The tests risk to bitrot a long time - to the point of not working
>| correctly on newer MS runtime by the time it is fixed on Mono (e.g.
>| rarely
>| used classes);
>| 
>| (b) there is (or was) no easy way from nunit-console to "try" to execute
>| an
>| ignored test. So this gets a little harder for people who intend to fix
>| the
>| bugs in that category (as they have to change and recompile the test
>| assembly).
>| 
>| Using a different [Category] for those tests doesn't have those
>| inconvenients.
>| 

You are absolutely correct and I agree. I overlooked the .NET side.

So, until someone gives us a more specific bug watching attribute, let's use
[Category("NotMono_1_1")] and [Category("NotMono_2_0")] to avoid tests that
shouldn't be run on a particular version of the framework.  Then we use
[Category("MonoBug")] to exclude known bugs to get the tests useful for
regression testing. 

I would suggest commenting the bug number, maybe like

[Category("MonoBug")] // http://bugzilla.ximian.com/show_bug.cgi?id=71236

This way we can identify those exclusions that can be removed once bugs are
fixed.

How's that sound?

Nick D.





More information about the Mono-devel-list mailing list