Hi there,

I am totally new in case of RadControls and at beginner level at the moment.
 I am just doing the first simple webpage i which I want to put a RadComboBox and with few items already in there.

When I run it I need to know at client side, how many items are there in a combo box and need to show the count in a msgbox or alert.

Can any one help me please... :D

Thank you in advance..

Khyati.

We got some problems running Silverlight tests on Firefox (16.0.2). The test stops as soon as the first element in the Silverlight app is accessed. This is the log message:


A complete log is attached.
The tests run successfully on IE and Chrome. Firefox is configured manually according to the documentation and the Telerik Extension is installed. We are using Test Studio Vesion 2012.2.920.0.

Hi, my suite of tests runs against IE8 and before I move to IE9 I would like to know if the file download dialogues 'Save', 'Open', 'Cancel' I have recorded in IE8 will simply work in IE9 or will I have to re-record them all? Thanks

We encountered a problem along the following test scenario for our Silverlight application:

TestA: test the creation-process of an object which automatically receives an unique ID. Extract this ID.
TestB: find the newly created object out of a list of other objects using the extracted ID from TestA. Here we test the search and filter routine.
TestC: test editing-process of this object, part 1
TestD: test editing-process of this object, part 2
...(other tests to test the editing of this object)..

(We tried to modularize the Test Studio project as much as possible to improve exchangeability and reduce redundancy.)
Now to run all tests we are using a Test List containing all tests. Unfortunately it doesn't seem possible to pass the extracted ID from test to test within a Test List. According to the documentation, this and this threads, the only solution would be

1) write the extracted data to a file (or another data source) and read from it for every test
2) use some kind of parent test which includes TestA, TestB, TestC, ... as steps and inherit the extracted data from it.

Solution 1) would be acceptable but needs some minor coding effort. Solution 2) has some disadvantages too. For example it somehow confilcts with the idea of Test Lists. (Another negative aspect is that you have to combine the storyboards from all sub-tests into a single one if you wish to have screenshots included in the test documentation.)

So, it would be really nice to avoid these "work-arounds" if the extracted data is globally available during the execution of tests in Test Lists. From our point of view, this would be the cleanest, easiest and fastest solution.

Are we missing something?
Is this feature planned for future releases?

I'm getting a red reporter icon and when I try to run tests I get "Cannot run because the reporter cannot establish its database"

How do I give it a DB?

Hi, just a suggestion.  I really like the fact that you can drive the tests from a CSV but given the fact that the client is often consuming JSON wouldn't it make more sense if you could actually drive the tests using JSON data.  That way I can make a restful API call from our web site and drive the test using the JSON result set.  If I could do that I would be a happy man indeed.

Is it possible to create a test that has a test step at the start with a data bind?

For example: I had a test that used an XLS dataset for the navigation URL at the start.  This worked successfully: it navigated to the first URL in the data, logged in and continued through the test successfully before doing the next iteration of the dataset.

I wanted to try a test step, so I made two copies of the original test - one that did the navigate and login, and one that continued on the test AFTER the login was successful.

However, the 2nd iteration fails and looking at the log it looks like it is doing the first navigate and login, then switches and does the 2nd navigate and login, then continues with the test.

How do I get it to wait to the end of the test before doing the next iteration.  Is it even possible?

Thanks in advance

I am quite frustrated at this point.  It appears that Test Studio has everything that I need to create a good web UI test.  Unfortunately the website that I am testing is very heavy jQuery (dynamically changing DOM) and has very few elements with ID's.  

This is the FRUSTRATING thing - I know how to find the exact element by its xpath (I like this TS feature) and I also know exactly what the innerText should contain in order to validate the test, however it appears that I cannot edit a test step without being connected to the DOM.  When I am connected to the DOM and I select the element that I want to validate during the recording session, the DOM element that I want to validate disappears and what I actually get is some other element that has nothing to do with my test.  The reason it disappears is because it was created dynamically and by clicking on the targeting dot to target the element it actually removes focus from the element selected.  Then what I get is some underlying more static DOM element, not what I need.

Is there any way to edit the test or create a test step without actually having a live DOM connection?

This would probably not be that big of deal but running the tool and connecting to my browser is incredibly slow, sometimes completely freezing the browser.  Currently I am running this on a VM with over 8gb of memory and plenty of disk space.  Perhaps I should run this on my host machine as it has 24gb memory.

This is satyam, I am facing one problem in test studio . In the below I have mentioned the steps what I have to do using telerik test studio.

Steps:

1. 1. In Telerik test studio I have to create a web test , Using the record function and I have recorded some steps on web page.
2. 2. In the same project I have created a WPF test and I recorded some steps on one of the WPF application which is under test.

Problem:

1. 1.  I want to execute WPF test by calling methods(which are defined in WPF) in web test.
2. 2.  In one of the test file i have to call the functions which is availabel in another test file.

 

Thanks
satyam

Hi,

We are playing around with installations of Test Studio for our developers as we are hoping that they will be able to use the run-time licences to run scripts against their own code-base before checking in, not only to see if what they have done is breaking the system they are working on, but also to see how much work they are going to cause me to fix any of the automation scripts.

On the other hand, they have also agreed to help out whenever they can to fix automation scripts, so I have trained them up on the automation tool and they will mainly be working with the coded steps areas of the automation code.

I was hoping to install both versions on each of their machines and for them to check in and out the licences as and when they need them.  However when attempting to install both on a single machine, the installation of the 2nd ( run-time ) executable failed. 

Please could you let me know if this is even possible, and if not, what is the best practice for what I am trying to achieve?

Many thanks
Shane