On Tue, Jun 11, 2013 at 2:45 AM, Dan Haywood <d...@haywood-associates.co.uk>wrote:
> > > > On 10 June 2013 20:49, Bhargav Golla <bhargav.go...@gmail.com> wrote: > >> >> 1. Delete_thenResponseCode_205_bad: Should I check for a DELETE request >> and then check for response code? >> > > Not sure what you mean by "check for" a DELETE request. You need to > submit an HTTP DELETE in the client, and then assert in the client that you > get a 205 response. > > There's a good chance, though, that this test will fail - ie that the > server does not correctly implement this behaviour. You are very welcome > to dig into it if so.... will probably involve some rooting around in > RestEasy and the JAX-RS. However, if you get nowhere, then skip this test > and find something easier to work on (at least, while you are coming up to > speed). Such as... > I will try to look into it. If I get nowhere, I will skip it as per your suggestion and concentrate on quick and easy ones. > > >> >> 2. Get_thenResponseHeaders_ContentLength_ok: What is the content length I >> should check for? >> > > With these, the easiest thing is to just assert on a nonsense value, have > the test fail, and then (so long as you've manually verified that the > response looks ok), take this content length as the one to check for. > > What we're after here is a regression test; there's no easy way - other > than manual inspection - to figure out the initial value. > > For a slightly more robust test, you might want to check that the content > length is larger than some value. After all, if there are changes it'll > probably involve the addition of new members to the class; which would > break this content length test. All we want is a "reasonableness" check > here, ie "yes, a content length was returned, and yes, it looks about > right". > I will check for content length greater than 1000 in all tests for now. > > Thx > Dan > > > >>
