A large part of me agrees with you.. but there is a tiny part that questions
why have mocks at all then?
On Fri, Nov 7, 2008 at 7:57 AM, Ayende Rahien <[EMAIL PROTECTED]> wrote:
> Shane,I think that a good guidance would be to tell anyone who doesn't
> really care about the difference to just use stubs, and assert on the
> results.
> And then we can direct them to Single assert per test rule.
>
>
> On Fri, Nov 7, 2008 at 4:54 PM, Shane Courtrille <
> [EMAIL PROTECTED]> wrote:
>
>> I was originally responding to Ayende.. then I realized my original
>> response didn't clarify what I was saying No about.
>>
>> The difference for me is in interaction based testing.
>>
>> The only place I use a mock is when I need to verify behavior but the only
>> way to do that is to verify that interaction occured with another class (aka
>> Interaction based testing). For me the canonical example is a class calling
>> a Repository (data access provider) to save something. In a unit test I
>> don't want to call the live Repository since that would access the live
>> database. BUT it's very important that I verify that the save occurs. In
>> this instance I would create a SINGLE test that uses a mock to verify this
>> one interaction.
>>
>> Now instead let's say I have a Foo that calls IBar for a piece of
>> information and without that information the class blows up (the result
>> can't be null for instance). Now this may be a case where I care about the
>> interaction in which case again I create a SINGLE test that uses a mock.
>> BUT for every other test I would use a Stub to provide the value that Foo
>> requires to do it's job.
>>
>> This I believe is keeping inline with the single assertion principle
>> (which I'm sure has a real name I'm forgetting). In my world I use a mock
>> to create an assertion and a stub to provide information a class requires to
>> do it's job.
>>
>> Hopefully that clears up my beliefs a little bit..?
>>
>>
>>
>> On Fri, Nov 7, 2008 at 7:44 AM, Tim Barcz <[EMAIL PROTECTED]> wrote:
>>
>>> Shane,
>>>
>>> Not following who you were replying to...can you clarify?
>>>
>>> Also, summarize your stance on Mocks vs. Stubs.
>>>
>>> I *can* assert interaction with a stub, see the test case that was
>>> provided in the first post.
>>>
>>>
>>> On Fri, Nov 7, 2008 at 8:42 AM, Shane Courtrille <
>>> [EMAIL PROTECTED]> wrote:
>>>
>>>> This email should have said "No.. you are correct that there is not
>>>> really that much of a difference"
>>>>
>>>>
>>>> On Fri, Nov 7, 2008 at 7:42 AM, Shane Courtrille <
>>>> [EMAIL PROTECTED]> wrote:
>>>>
>>>>> No.. and that is the problem. The difference is very much a conceptual
>>>>> thing so it's very important that it be protected. By allowing the
>>>>> assertion of stub calls the usage of stub and mock is diluted.
>>>>>
>>>>> Maybe you can help us understand why someone want to be able to assert
>>>>> stub calls and your view on why this was a good feature to add? Isn't
>>>>> that
>>>>> what a mock is for?
>>>>>
>>>>> Shane
>>>>>
>>>>>
>>>>> On Fri, Nov 7, 2008 at 7:08 AM, Ayende Rahien <[EMAIL PROTECTED]>wrote:
>>>>>
>>>>>> Because there isn't really much of a difference between the two, and I
>>>>>> got feature request to be able to assert of stub calls.
>>>>>>
>>>>>>
>>>>>> On Fri, Nov 7, 2008 at 4:03 PM, Shane Courtrille <
>>>>>> [EMAIL PROTECTED]> wrote:
>>>>>>
>>>>>>> Which to me is the standard difference between the two.
>>>>>>>
>>>>>>> The question is why does the interface for a stub have to support
>>>>>>> operations which are the domain of a mock object? When a new comer
>>>>>>> looks at
>>>>>>> a Stub shouldn't they be able to tell the difference without having to
>>>>>>> deal
>>>>>>> with the mental baggage of "Hey wait.. a stub looks a lot like a mock to
>>>>>>> me".
>>>>>>>
>>>>>>> Shane
>>>>>>>
>>>>>>>
>>>>>>> On Fri, Nov 7, 2008 at 6:15 AM, Ayende Rahien <[EMAIL PROTECTED]>wrote:
>>>>>>>
>>>>>>>> The main thing is a concpetual difference.I would use stub if I
>>>>>>>> don't care about what is going to happen
>>>>>>>> I would use a mock if I do care.
>>>>>>>>
>>>>>>>>
>>>>>>>> On Fri, Nov 7, 2008 at 2:05 PM, Tim Barcz <[EMAIL PROTECTED]>wrote:
>>>>>>>>
>>>>>>>>> I am working on a presentation and was about to make the assertion,
>>>>>>>>> "stubs can't have expectations set on them". I decide to double check
>>>>>>>>> myself and run the following code and see that I can set expectations.
>>>>>>>>> (Removing the call to logger.Log in the Order method causes the test
>>>>>>>>> to
>>>>>>>>> fail).
>>>>>>>>>
>>>>>>>>> [TestFixture]
>>>>>>>>> public class Stubs
>>>>>>>>> {
>>>>>>>>> [Test]
>>>>>>>>> public void CreateStub()
>>>>>>>>> {
>>>>>>>>> var logger = MockRepository.GenerateStub<ILogger>();
>>>>>>>>>
>>>>>>>>> var svc = new OrderService(logger);
>>>>>>>>>
>>>>>>>>> svc.Order();
>>>>>>>>>
>>>>>>>>> logger.AssertWasCalled(x=>x.Log("hello"));
>>>>>>>>> }
>>>>>>>>>
>>>>>>>>> public class OrderService
>>>>>>>>> {
>>>>>>>>> private readonly ILogger logger;
>>>>>>>>>
>>>>>>>>> public OrderService(ILogger logger)
>>>>>>>>> {
>>>>>>>>> this.logger = logger;
>>>>>>>>> }
>>>>>>>>>
>>>>>>>>> public void Order()
>>>>>>>>> {
>>>>>>>>> logger.Log("hello");
>>>>>>>>> }
>>>>>>>>> }
>>>>>>>>> }
>>>>>>>>>
>>>>>>>>> public interface ILogger
>>>>>>>>> {
>>>>>>>>> void Log(string message);
>>>>>>>>> }
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
>
> >
>
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Rhino.Mocks" 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/RhinoMocks?hl=en
-~----------~----~----~----~------~----~------~--~---
