As an HtmlUnit user, I thought I'd pipe in a bit too. Dan, please correct me if I'm blatantly wrong on anything here.
HtmlUnit, despite its name, tries to do a lot more than just HTML. So, the framework is general, which seems like a great idea until you actually go to use it and are forever casting things to their HTML variants, which, I would imagine is the majority use case. It got to be such an eyesore and time sink that we actually wrote another wrapper on top of HtmlUnit. The point here being that HtmlUnit, while simple, is really cumbersome and you're going to end up writing plumbing code you probably don't want to and probably shouldn't have to. Additionally, since it is integration testing, I've found it really hard to say X percent of the code is covered by tests. I suppose use cases can be covered, but actual units seem difficult, if not impossible. It'd be interesting to see how you guys are calculating this metric. -- Kevin On 12/21/05 1:32 PM, "Dan Adams" <[EMAIL PROTECTED]> wrote: > Yeah I'm the only developer on this and I've found that using htmlunit > with tdd has helped a lot to keep development speed high. without tdd it > would take much longer. > > On Wed, 2005-12-21 at 18:46 +0100, Pedro Abelleira Seco wrote: >> We also have high-level methods for the particular components of our >> application, but the development cycle is slow and we haven't achieved >> to finish the test cases. We have little resources, but without testing >> we can't get very far as any change takes longer and longer. Testing >> really makes a difference on that. >> >> Thanks. >> >> Pedro > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
