On 1/30/07, Nathan <[EMAIL PROTECTED]> wrote:
> Hmmm,
> So here is some output from a simpler test I've begun to write and I don't
> approve of this. If you can see that little dot, and notice the output, there
> is only one test.
Seriously, get a book pretty soon. The dot means that the test
passed. An "F" means failure, and an "E" means error.
Within your Test::Unit::TestCase, every method that starts with
the characters "test_" will be considered one pass or one fail,
regardless of how many assertions that method contains. You are free
to have one assertion per "test_" method, or as many as you like.
Now I want to break up my test into specific components, but I don't
see that browsing to the page every time is very efficient, nor is it
anything like the way that our commercial competitor (Watir's) QTP
does things.
This is generally considered a feature. :)
So I'm thinking I'm going ot need to write my own framework
ultimately, but anyway. So far I have two assertions in here, but they
don't count as tests!?
If you want them to be two different tests, put them into two
different "test_" methods. Just BTW, the behavior upon failure *is*
probably what you want. You should try it out. Also, my personal
advice is to relax for awhile about the framework, and get a good
solid understanding of the 19 (!) different assert() methods that Ruby
provides; then look into what the Rails people have done with custom
assert() methods, and investigate Rspec, before you get all
framework-wacky.
I don't mind answering these questions, you're definitely on the
right track, but you're moving fast enough now to need to read up.
-Chris
_______________________________________________
Wtr-general mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-general
