Re: [TFUI] The TFUI Principles

> Dot.Net sure didn't get in the way, huh?
well, actually the implementation was in java.  but it would work much
the same in .net, as i'm sure you can imagine.  what i haven't done
yet is to hook into the swing or winforms event queue.  has anyone
tried this?

> I can't tell if your blog specifies that you frequently comment your
> reveal() out. 
yes.  comment out or (preferrably) delete.  part of what i was saying
in the blog entry is that we use a jvm parameter to ensure that a
modal dialog won't pop up and cause the build to freeze (we're using
cruisecontrol) just in case someone forgets to comment out/delete the
reveal() method call.

Follow two rules:
> 
>     - tune your reveal() so you can call it from any test case in any
>       suite of tests, on the GUI, at whim. You should have to think of
>       nothing between realizing you need to see a given GUI, writing
>       reveal(), and seeing it.

we're not quite there with that as we are testing the client layer in
its entirety.  in setting up the tests we need to stub out all server
calls, so they're only "unit" tests.

>     - the goal is to never need to call reveal(). The goal is to TDD
>       GUIs for as long as possible without even realizing you haven't
>       seen a window in a long time.

for the most part, reveal is not needed.  it is just instructive to be
able to see and interact with the gui in order to visually validate
certain behaviour.

> So, as you attain the first goal, you stand on a slippery slope. You _could_
> write a GUI very rapidly by writing the beginnings of test cases, calling
> reveal(), finishing the tested feature, and never writing any assertions.

actually, we don't really have that problem yet.  but we are generally
writing tests around  screens that already exist.  we are only writing
tests for the functionality that we intend to add.

> (BTW about that nasty Spam infestation, fight it quickly, even if you must
> lock your blog down. Spam works by letting Google spider your page, even if
> nobody who reads your page is stupid enough to click on a link. Those links
> now have a higher pagerank in Google, which is what the spammers' clients
> are buying.)

thanks.  i know.  it's a pain.  my blog provider only provides
protection by turning comments off, which is something that i don't
want to do (yet).

cheers,
owen.

-- 
Owen Rogers | http://dotnetjunkies.com/weblog/exortech |
CruiseControl.NET - http://ccnet.thoughtworks.com


To unsubscribe, email:
TestFirstUserInterfaces-unsubscribe-hHKSG33TihhbjbujkaE4pw@public.gmane.org
 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/TestFirstUserInterfaces/

<*> To unsubscribe from this group, send an email to:
    TestFirstUserInterfaces-unsubscribe-hHKSG33TihhbjbujkaE4pw@public.gmane.org

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/