On Tue, Aug 17, 2010 at 09:27, Frans Bouma <[email protected]> wrote:

> > On Tue, Aug 17, 2010 at 04:02, Frans Bouma <[email protected]> wrote:
> >       > Since there seems to be some confusion, this is the main test I
> was
> >       > referring to:
> >       >
> >       > [Fact]
> >       > public void ConfigurationIsValid()
> >       > {
> >       >     //call the method that creates the configuration and builds
> the
> >       session
> >       > factory }
> >
> >              useless. Even if this test succeeds, you have no idea if
> your
> > xml is
> >       valid or not.
> >
> > If NH compiles it, it's valid. That's the only purpose of this test. It
> can
> > check well-formedness, schema, structure, names and semantic.
>
>         gee, then why did I got crashing queries when I ran them, but the
> xml was valid? some magical ball hovering over my office, which influenced
> the queries at runtime?
>

When did I claim otherwise?


> > ...Which is something FNH won't do by itself, and that was the whole
> point.
>
>         I think what was claimed was that FNH lets you catch crap you can
> only check for by running it. I don't see how you can magically get that
> with text-based XML.
>

With a one-line test. Are we really going to keep arguing about that?


>
> >              Both are results of work done by a human which can be better
> done by
> >       a machine, but heck, some people like the idea of doing work a
> machine could
> >       do better...
> >
> > The machine is xunit. Why would that be better that csc?
>
>         ah, so xunit creates your tests? Out of the blue? Or did you write
> extra code to get that working?
>

It's the one-liner shown above. Unless you also believe I need a machine to
write that too...


>
> >              What I don't understand is that you find the 'run the tests
> to
> > see
> >       if what I wrote is correct' is a valuable approach: it takes time,
> > and if
> >       you screwed up during test writing, you're not going to be happy at
> > runtime.
> >
> > Time? It's one click away. Just like the C# compiler.
>
>         So is a debugger, that doesn't make it the same thing.
>

The debugger is interactive and requires a lot of time in my part, the test
runner, in my view, is just a build step.
Not only that: in dynamic languages, it's THE ONLY step.

  Diego

-- 
You received this message because you are subscribed to the Google Groups 
"nhusers" 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/nhusers?hl=en.

Reply via email to