Thank you for your response.
Let me be more specific. I have a Keyword testcase that tests that I want to be more flexible in having different types of tester input WITHOUT having to change the testcase. For example, assume you that are testing the window of a tax application in which has has the ability to show the monthly tax periods for the last 5 years. Since there 60 tax periods in 5 years, the window is designed to show the most recent 12 tax periods, However, if an older tax period must be edited, there is another button that allows use to display the next 12 older tax periods. At the bottom of this window, there is a EDIT button which allows you to edit the data associated with a tax period. Also, assume that depending on the taxpayer account you are logged in under, some tax periods may be available for editing and others may not. Since you do not know before the testcase is executed, which tax periods will be available for editing, you place a breakpoint in the testcase to PAUSE the testcase on the statement which clicks on the EDIT button. (For the sake of this example, also assume that it is NOT possible for Test Complete to determine the INDEX of the selected tax period automatically. The tester must place the CURSOR on the line of the tax period to be edited, and then press the EDIT button.)
In order to help the tester know what to do next, you want to post a message, NOT to the TestComplete Log, but to the system on which the tax application is running. I was hoping that Test Complete would have a POP-UP WINDOW feature that would allow this message, which would remind the tester what should be done next, to be posted. However, as I stated in my original post, I tried the INDICATOR feature but that message is displayed too quickly and the message associated with the testcase executing the BREAKPOINT overwrites that message as well. I want this message to remain visible until the tester manually closes the window.
As a workaround, I just created a simple Notepad keyword test that posts the desired message in a Notepad window and I just call that Notepad keyword test right before the breakpoint is executed in the main application test.
I realize that leaving the breakpoint in the application test makes that test semi-automatic since manual intervention is now required when running that test. However, not REQUIRING the tester to know all possible testcase inputs PRIOR to creating the test, gives the tester more flexibility in dealing with applications where the actual, specific situation that may occur while running a test is NOT KNOWN beforehand. It also allows the tester more flexibility in selecting the tax accounts that need to be worked on, which, in turn, makes the testcase more flexible in dealing with different tax accounts.
Does this description help describe what I am looking for?
