Issues designing and running tasks

[Superstar]

We just purchased a license of AA Enterprise 4.5 in a large government organization.

We do massive testing of Commercial Off The Shelf (COTS) applications.
For example, we test Word 2003, Project 2003, visio 2003, Adobe Acrobat 8 to name a few on a variety of baseline platforms.

I am currently designing a testing methodology based on AA 4.5 and I'm having some conceptual issues. I personnally recommended it based on the trial version but I'm starting to panic because, although our testing seems to be a "text book" case for AA, I'm starting to doubt it will do the job.

Basically, I want to deliver a solution that is "modularized" and relatively easy to maintain.


Issues/questions:

1) I need to use a multitude of tasks, basically, one for each tested apps. But I realized that if you launch an .exe, it will be slow overall because the runtime is loaded everytime, say 20 times. Is there a way to load a first exe and then call up addidional separate tasks that are not compiled (.atmn)? (we cannot install AA on each tested computer)

2) As a workaround for my issue described above, I set to copy paste all my separate tasks into one main one. Well, is it possible that AA gets confused with the windows names? For exemple, there a lot of "Save as" windows and sometimes AA stops because it can't seem to find the right one. So is is a bad idea to "cut and paste" task code?

3) Also, I recall (in the trial version) having seen "screenshots" of the various windows used in a recorded task. I cant seem to find those anymore...

4) Is there a way to list or view everything that is being done by AA in the background? I noticed AA mimics keyboard input to the letter, including delays etc where are those delays? They're not in the source code.

5) Often, some windows are being shifted (re-positioned) slightly during a task execution - why is that? I can't figure this one out.

6) Is there a way to "branch" in the task code? Like goto's? Sometimes it's convenient to bypass sections of code. I haven't found the debugger usefull in that regard.


Thanks for your help

[forumsupport]

Hello,

Regarding your queries,

1) You can create separate task for each tested apps and can chain them together using the Run Task command in one main task. You can find Run Task command under ‘Tasks/Scripts” command category.

2) If you insert enough delay after every command in task and ensure that the keystrokes are being played in correct window then it should not create any problem. In addition, you can work with standard windows controls using ‘Manage Windows Controls’ command, which is found under ‘Programs/Files/Windows’ command category. For e.g. to insert the file name in 'Save As' window, you can use ‘Set Text of Textbox’ command. Text format would appear as follows,
Set Text of Textbox in Window : "Save As" Text : "C:\...."

3) You would be able to see the screenshots on the same machine and in same user account in which you recorded the task.

4) You can see every action that is recorded in a Task Editor after saving the task. Regarding Delay command after every command, they are the delays associated with the command and are required for errorless execution of that command. Recorder automatically inserts the delay associated with each action or command. Insufficient delay can skip any command in task and results into task failure.

5) The position where the window was, when it was closed last time, it opens again at the same position. However, when you run the task again, the window moves to the position where it was at the time when you created the task. This is because the task records size and position of windows the moment when you insert any keystroke or mouse action in that window to ensure all the activities are performed reliably.

So, what you can do is keep the window in one size and position where you want to when you enter any keystroke/mouse actions in it so that when the task is run again, it will be in the same size and position as it was at the time it was created.

6) You can make use of ‘If-Else-EndIf’ statement to perform actions based on some condition. If condition were satisfied then commands after ‘If’ would be executed otherwise commands after ‘Else’ would be executed. You can find ‘If’ command under ‘Conditions/Loops’ command category.

Hope that helps.