Hi,

Our application team recently updated one of the screens in our application to replace RadTreeView with RadTreeListView to display a hierarchical set of data.  While updating our automated tests to react to this change, I ran into several issues in the translator for the TreeListView control:

a) Attached error message showed up when I tried to record a step to verify root node count on the TreeListView control.  Step was recorded using the Build Validation dialog - error occurred when I clicked OK after changing expected value (originally "value?") to the expected value.

b) When I ran the test, that same step failed with the following exception (note:  same exception also occurs if the type of comparison is anything other than Equals such as GreaterThan):

c) Same exception occurred when I attempted to get the root node count in a coded step (using the API RadTreeListView.TreeListItems.Count)
d) Same exception occurred when I attempted to get all the root nodes in the tree (using RadTreeListView.TreeListItems).

Upon reviewing past posts in the forum, I came across the following threads that (I believe) are reporting the same or similar issues:
http://www.telerik.com/automated-testing-tools/community/forums/test-studio-express/automation-framework/bug-with-radtreelistview.aspx
http://www.telerik.com/automated-testing-tools/community/forums/test-studio-express/automation-framework/time-outs-when-evaluating-properties-in-radtreelistview.aspx
http://www.telerik.com/automated-testing-tools/community/forums/test-studio-express/automation-framework/treelistviewtests-bugs.aspx

The bug referenced in the above threads (PITS 10006) remains unscheduled - and there is no corresponding issue (that I could find) in the new Issues and Feature Requests site.

Telerik:  Could you update the status of the above issue?   If it is fixed, which version of Test Studio has the fix? (Note, we are on 2012.2.1420 - upgrade is currently not an option for us).  If it is not fixed, could you raise priority on it (there are already 5 votes on the PITS issue - I will add my vote as well).

I have worked around the above issues using the workarounds suggested in the above threads as well as some I developed on my own - but I do have a couple of questions/unresolved issues:

a) What is the real object type of the FrameworkElement objects in the list returned by the following code?  it would be nice if I could get a list of higher-order objects (such as RadTreeListItem).

b)  Due to the issues with TreeListItems API, I am unable to perform tree traversal and validation functions (for example, verify row B is a child of row A, or get all child rows of a given root node row).  I believe that the TreeListView is still a tree view.  I welcome any ideas/code for doing that in my version of Test Studio (in my signature block below).

Thanks for any help you can provide.

Shashi
Telerik Test Studio 2012.2.1420.0 (VS plugin).

Hi!

I'm testing a Silverlight application.
I have a step on my test that's to type something in a textbox.
The recording works fine, but in the execution that step fails, there is no text typed into the textbox.

I've found that if I set "SimulateRealUser" to False, in the execution the text is typed in the textboxes, and no errors occurs.

Could you explain me what's the difference between having "SimulateRealUser" to True or False, and why when it's set to True my step fails?

Thanks,
Paulo

When I made execute for a simple test case the result was "Fail - 0 passed of 0 executed"
and on viewing the text view there was an error

--------------------------------------------------
'4/30/2010 2:22:23 AM' - Error encountered during execution.
--------------------------------------------------
'4/30/2010 2:22:23 AM' - Error attempting to ping runner. Error: System.IO.IOException: Pipe is broken.
   at System.IO.Pipes.PipeStream.WinIOError(Int32 errorCode)
   at System.IO.Pipes.PipeStream.WriteCore(Byte[] buffer, Int32 offset, Int32 count)
   at System.IO.Pipes.PipeStream.Write(Byte[] buffer, Int32 offset, Int32 count)
   at ArtOfTest.WebAii.Messaging.Process.PipeCommunication.WriteCommandToPipe(PipeCommand command, PipeStream pipe, Boolean waitForDrain)
   at ArtOfTest.WebAii.Design.Execution.RunnerController.Ping()
--------------------------------------------------
'4/30/2010 2:22:23 AM' - Aborting....

We just bought Test Studio, it seems great! I am currently going through our Silverlight application and adding automation ID's to all of our Telerik UI controls, since the full Xaml path seems volatile.

I noticed something I hope I don't have to manually edit every time-

I click on a textbox in our application, and type text into it.  2 elements are added to the repository, not 1.
Example:

• LeftClick on Item4Textboxview
• radwatermarktextbox: Type 'name' into rwtbLastNameRadwatermarkertextbox

I've attached two screenshots of the find logic for both elements.  Why does it want to also use the full Xaml path for the click? I'm happy I can remove that, and the extra element is removed from the element repository, but we have hundreds of UI controls.  I hope I don't have to edit every click element since the Xaml path changes sometimes. Is there a shortcut?

Thanks!
-Dan

Hi,

I jsut upgrade from old version(2012_2_1420) in which there is 1 result file for the list.

But in the latest version, all cases reulsts are separate within the list. And when I exported to Word/Excel/VS, I can not get the steps details.

Can I export all cases and their steps into one .aiiresult file? That's really helpful as the final report.

Thanks a lot.

Hi,

I just upgrade from "2012_2_1420" to latest "2013.2.1210".

Before my upgrade, all cases can passed both Standalone running and List running.

But after upgrade, cases were failed in list running. I checked the log that some exception as

'12/25/2013 5:27:43 PM' - LOG: Unexpected dialog encountered. Closing the dialog, and halting execution.
'12/25/2013 5:27:45 PM' - 'Pass' : 10. Wait for Exists 'MoveToNextPageButtonRadbutton'
'12/25/2013 5:27:45 PM' - 'Fail' : 11. Execute test 'Public_Waiting_PleaseWait.tstest'

and log details

ArtOfTest.Common.Design.Exceptions.ExecutionException: Unexpected dialog:

But some cases passed with this same step in this list.

And for this faild case, I also can pass it in standalone running

So could you help me?

My IE version 9.0.8112, OS Win7

Attached is the result file and my list settings screen shot

Hi,

I just upgrated my version to latest 2013.2.1210

But the debug mode is not work when Run - From Here or Run - Selected Steps. Seems all cases can not use these function.

Run - To Here is fine.

Attached is the error log. Please have a check. Thanks in advance.

Hey everyone,
I'm having trouble trying to figure out the most efficient way to build my suite of tests.  The issue is, I have tests that do environment specific things.  For example, in one of my tests, I query our qa environment database to get a specific value back which is then used to help datadrive the test.  Now what if I want to run this same test in UAT or Prod?  This means I need to change in all the tests which environment I am pointing to.  I want to be able to just have to change one value in a test and then everything will run in the correct environment.

My idea was creating a custom C# class that has functions that based on which environment you are in, returns the correct value.

Here's a quick example of what I want.

public String getUrl(Environment env)
{
if(env == Environment.QA)
{
return "qaurl";
}else if(env == Environment.UAT)
{
return "uatUrl"
}
}

Is this possible in Test Studio?  If so, where can I create my C# class to use and reference it in my project?  I am not familiar with the C# way since I'm coming from a Java background.

Just did my first check-in into TFS. Having problems because the character limit is being hit.

Why do you guys put so many extra characters in the file paths? 

I am doing a test of an ajax form that we have running, and one of the conditions it does is sets a buttons state using class based on some of the jquery.ui built-in classes.  Our form is running this very long running ajax routine (we are doing some complex file IO) is using signalr to send feedback back to the ui.  What I would like to do is check the state of a button html element on the screen, and wait while it has a specific class.  when that class disappears (the button becomes enabled) I would like to proceed.

Here is the buttons initial html:

<button class="jw-button-next ui-button ui-widget ui-state-default ui-corner-all ui-button-text-icon-secondary ui-state-hover ui-state-disabled" role="button" aria-disabled="false" type="button">

notice 'ui-state-disabled' there, that is what I want to be waiting to disappear.

Any suggestions?