Can automated Web testing tools like Test Director from Mercury (http://www.mercury.com/us/products/quality-center/testdirector/) be used to test K2.net workflow applications?
What recommends do you have for automated testing of K2.net workflow solutions (assuming fairly "traditional/simple" human workflows with no external app integration that uses SmartForms invoked from a K2.net Task List (ITB web part or a custom user control that uses K2.net web services)).
Michael Herman
Parallelspace Corporation
+2
Hi Michael
Tools that record interaction with a web client like Microsoft Internet Explorer can be used in conjunction with both the K2.net Workspace and SmartForms as well as custom pages that use K2ROM.
Depending on what you would like to test these tests can be fairly simple or rather complex. You could, for instance, start a process from a SmartForm and then enter the Workspace to see if the worklist item is available.
Things that might become tricky, depending on the aim of the test and the capabilities of the tool, are defining the success criteria. To run a successful test (as recognized by the tool) you would probably need to start your test cycle on either a clean database or at least one in the same state as when the tests are recorded. This is especially true if you are going to include the reports pages in your tests.
Something else to consider carefully is the configuration of latency settings between HTTP requests. Dev and QA systems are seldom as powerful as their production counterparts, and this should be taken into account when performing automated tests.
I hope this answers some of your questions.
Emile
Tools that record interaction with a web client like Microsoft Internet Explorer can be used in conjunction with both the K2.net Workspace and SmartForms as well as custom pages that use K2ROM.
Depending on what you would like to test these tests can be fairly simple or rather complex. You could, for instance, start a process from a SmartForm and then enter the Workspace to see if the worklist item is available.
Things that might become tricky, depending on the aim of the test and the capabilities of the tool, are defining the success criteria. To run a successful test (as recognized by the tool) you would probably need to start your test cycle on either a clean database or at least one in the same state as when the tests are recorded. This is especially true if you are going to include the reports pages in your tests.
Something else to consider carefully is the configuration of latency settings between HTTP requests. Dev and QA systems are seldom as powerful as their production counterparts, and this should be taken into account when performing automated tests.
I hope this answers some of your questions.
Emile
Reply
View our Community Site Map
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.