To be honest, it looks worse than it is. The whole method is dedicated to
one purpose, executing a test and catching an exception if required, which
is the behaviour you're looking to modify. We could probably break it up
into a couple of overridable methods, but I'm loath to undo all Jeff's good
work in reducing the stack depth.

On 2 October 2010 20:09, Mark Kharitonov <[email protected]> wrote:

> Yep, the more I look at the Execute method the less I want to override
> it. Too much internal logic to copy/paste.
>
> On Oct 2, 9:03 pm, Mark Kharitonov <[email protected]> wrote:
> > Yes, you are right. Execute is the right place, however, this function
> > is not trivial. At least all that code around the failure logging plus
> > the various checks. I wish I could make a more simple override.
> >
> > On Oct 2, 4:59 pm, Graham Hay <[email protected]> wrote:
> >
> > > You just need to override Execute, and replace the existing
> implementation
> > > (which handles ExpectedExceptions) with your own. It might be worth us
> > > pulling the exception handling code out into another hook method, to
> make
> > > this easier in future.
> >
> > > On 2 October 2010 15:35, Mark Kharitonov <[email protected]>
> wrote:
> >
> > > > We already subclass the TestAttribute class (amongst others). The
> most
> > > > suitable place is the Consume override, but it seems complex. I hope
> > > > there is a better solution.
> >
> > > > On Oct 2, 3:34 pm, Graham Hay <[email protected]> wrote:
> > > > > I thought you could assert the InnerException as well, but that
> doesn't
> > > > > appear to have survived the transition from MbUnit
> > > > > v2<
> > > >
> http://www.gallio.org/api-v2/html/P_MbUnit_Framework_ExpectedExceptio...>.
> > > > > The easiest thing to do is probably to subclass/replace the
> TestAttribute
> > > > > with a custom version that satisfies your requirements, either by
> > > > stripping
> > > > > the outer exception or adding inner exception metadata to check.
> >
> > > > > On 2 October 2010 12:33, Mark Kharitonov <
> [email protected]>
> > > > wrote:
> >
> > > > > > Dear sirs.
> > > > > > We have a situation, where tests throw exceptions wrapped in a
> general
> > > > > > purpose exception. So that one has to specify this general
> purpose
> > > > > > exception in the [ExpectedException] attribute, rather than
> actual
> > > > > > business exception.
> >
> > > > > > Only there is another complication. We have build a framework,
> where
> > > > > > individual tests are combined in scenarios using XML files. The
> same
> > > > > > test may appear several times in one scenario and the test input
> > > > > > specifies whether the test is expected to fail or not. So, there
> is
> > > > > > neither hard coded [ExpectedException] attribute nor
> Assert.Throws
> > > > > > statements in the code. (We dynamically add the expected
> exception
> > > > > > metadata when needed).
> >
> > > > > > Anyway, the person creating the scenario (i.e. writing the XML
> file)
> > > > > > now has to indicate that the particular test is expected to fail
> with
> > > > > > some artificial exception (like ShunraDataPortalException)
> instead of
> > > > > > a specific business exception (like
> LastChangedMismatchException).
> >
> > > > > > I was wondering if any of the following possible:
> > > > > >  - Automatically and transparently wrap each test in a wrapper
> with
> > > > > > the same signature, that would try-execute-catch-strip-throw.
> > > > > >  - Somehow modify the exception checking, to extract the actual
> > > > > > business exception before asserting the expected exception.
> >
> > > > > > Thanks.
> >
> > > > > > --
> > > > > > You received this message because you are subscribed to the
> Google
> > > > Groups
> > > > > > "MbUnit.User" group.
> > > > > > To post to this group, send email to [email protected]
> .
> > > > > > To unsubscribe from this group, send email to
> > > > > > [email protected]<mbunituser%[email protected]>
> <mbunituser%[email protected]<mbunituser%[email protected]>
> >
> > > > <mbunituser%[email protected]<mbunituser%[email protected]>
> <mbunituser%[email protected]<mbunituser%[email protected]>
> >
> >
> > > > > > .
> > > > > > For more options, visit this group at
> > > > > >http://groups.google.com/group/mbunituser?hl=en.
> >
> > > > --
> > > > You received this message because you are subscribed to the Google
> Groups
> > > > "MbUnit.User" group.
> > > > To post to this group, send email to [email protected].
> > > > To unsubscribe from this group, send email to
> > > > [email protected]<mbunituser%[email protected]>
> <mbunituser%[email protected]<mbunituser%[email protected]>
> >
> > > > .
> > > > For more options, visit this group at
> > > >http://groups.google.com/group/mbunituser?hl=en.
>
> --
> You received this message because you are subscribed to the Google Groups
> "MbUnit.User" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to
> [email protected]<mbunituser%[email protected]>
> .
> For more options, visit this group at
> http://groups.google.com/group/mbunituser?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"MbUnit.User" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/mbunituser?hl=en.

Reply via email to