I have some problems using EasyMockNET with C#, trying the samples from Astel's book. When I try to link the view and the controller in GUI using a call to a...
I also have problems in using EasyMockNET when I uses ArrayList as parameters of the methods. If anybody wants, I can send him a demo project to see the...
... I have bootstrapped the TFUI Principles for an Exalted Dialog Box in C#. To use this, start a Windows app, paint a single dialog called Form1, and paint on...
... Let's go at this from first principles: - Inspecting a window, instead of writing a test, sucks. - mocking a window won't test. The mock will tell you what...
I am having problems, doing TDD with C#, when using mocks. I have try with EasyMockNET and with NMock, and I have similar problems with methods with ArrayList...
Hi Juan, let me jump in here and send me your zip file via private mail. Please describe exactly what you expect your code to do and what it doesn't do. Maybe...
... Please send it to me, then start a new project using this technique: http://groups.yahoo.com/group/TestFirstUserInterfaces/message/426 Don't throw your...
... Sounds righteous. ... Below my sig is the first pass - a tiny Win32 screen-cap program. The second pass would be ImageMagick to convert a sequence of GIFs...
Has somebody know of a framework like Jemmy for TDD of GUIs in C#? Something that let me go from the test to the controls in the user interface without writing...
... Make them public. Testing a GUI by driving it thru its form breaks the rule "the test should be close to the testee". Think of it this way: When a control...
... That's for finishing a GUI. My rule is for starting one. Can you test-first thru Jemmy, meaning tell it to expect controls, fault when they are not there,...
... Just checking that I understand your point (please confirm or correct if I got it wrong) because my first hasty - and now deleted - reply missed the subtle...
... Caveat: I have never tested a GUI via capture/playback, and have never used Jemmy. I'm not really certain what it is. Anyone here feel free to rescue the...
Thank you all for your replies. I was looking for something like Jemmy because it is used in the book that I am studying. ("Test-driven development. A...
... Make it public. Using a tool that searches for the button and is the same thing as making its click method public and clicking on it. If I wrote a...
... I use Visual Basic .NET. There the controls are protected. This means that if the form is in the same assembly as the tests then you already have public...
... Firstly Dave Astels uses NUnit, but you have switched to csUnit. I had to replace that with this: using csUnit; Also, I had to add csunit.dll to the...
... We've tried this approach and rejected it because it doesn't test the user's actual experience. Rather, it tests instances of GUI objects on the assumption...
... I came to the same conclusion, and decided that it was too hard to do. I remain interested. Ron Jeffries www.XProgramming.com Just because we learned...
... My technique makes refactoring in the GUI Layer safer. You describe the "acceptance test" need. ... That is an example of a Naive Convenience - thanks for ...
... In MFC you can say: bool windowWouldBeVisible(CWnd& window) { return (window.GetStyle()&WS_VISIBLE)!=0; } I guess the equivalent is available for C# and...
... Could you get around this by simply adding an assert in front of the PerformClick call to check the enabled state of the menu item? That way the test would...
... Also, in my previous experience testing forms in Delphi and J++, it was usually necessary to call form.show, form.update (sometimes), to help get the...
... It depends on how the lib accesses the widget, and I actually have no experience with the perl/ruby libs yet, so I'm not sure what limitations may exist....