Hello,

I found that a big report crushes the report viewer. The report we can download from https://drive.google.com/open?id=0B0EHXWVdmSJFc3Zub3JaT2ZqNW8 (CrashReport.zip). One crashes the viewer each time.

As I understood there is a problem with the viewer's system memory size, it is very big.

I need to solve the crash problem, it is above all.

I took a small investigation of the reason for the big report size.

In two attached test projects TestProjectForTelerik_BigReport and TestProjectForTelerik_SmallReport the same test actions are executed (tests set is the same too but there are some differences in test configurations).

1. Load www.telerik.com page

2. Come back to the main page

3. Click on a link to the products. The link is found by content (product's name) from data source (TelerikProducts.xls file, Products sheet)

But different ways are used to run tests in the projects.

In TestProjectForTelerik_SmallReport we start from SmallReportTest (use SmallReport test list from the project) and bind ClickProduct test to the data (the data source) directly.

In TestProjectForTelerik_BigReport we start from BigReportTest (use BigReport test list from the project) and bind IntermediateStep test to the data directly. Data binding is removed from ClickProduct test but this one must inherit the parent data source.

In IntermediateStep test we choose which test will be run next according to data from data source. In IntermediateStep we always run ClickProduct test. Please see comments in the IntermediateStep test for more information.

This way is very needed for my real project.

So, these two projects have very different report size for the small report it is 249674 bytes, for another 989670 bytes. The difference is about 4 times.

If steps count in ClickProduct test or row count in data source increases, size difference will be more.

As I can see in the report from TestProjectForTelerik_BigReport project, there are a lot of duplicated information in the report. The information from StepResults section of ClickProduct test is duplicated in DataIterationResults section of IntermediateStep test.

Please check this. Maybe there is a way to decrease the report size.

And as I said earlier Telerik.TestStudio.ResultsViewer crush must be fixed.

I need your help, thanks.

Right now, the way I handle it is creating a local data grid in the parent test, and using a single row with multiple columns to send the proper data to a Test as Step (usually no more than 5-10 columns), which inherits the parents data source.

Example: There is an individual parent test for individual data tables (lets say I have 10 tables, and 10 parent tests), which will export the data to an MDB file. Test as Step will be used to handle the actual export functionality which requires a file path and table name. It should get these values from the parent test, as this will be used in a bunch of different tests.

I'd like to avoid using coded steps (in each parent) or an external code file because if I need to change an element reference for any reason, it takes forever to update them all.

I know I could also create a single coded step in each parent and use a handful of SetExtractedValue(); and get those from the Test as Steps, instead of using the local data grid.

Any other solutions that I'm overlooking?

Have a simple typeahead box on a web site and I am able to record the actions to enter the text and select from the dropdown list that appears one of the selections. When playing this back it appears that the DOM does not refresh during the test, so I converted the step to a code step and added ActiveBrowser.RefreshDomTree() call. This did not help.

So I added a break point on the failing step to capture the DOM. I then opened the DOM explorer using that option right next to the capture option and hit refresh. Captured the results again (which now has the list items in it). But the test still fails.

There has to be someway to click an option in the list that appears (I could modify the input text that drive the list and get more or less options, but I only want to select a specific one form that list). Thoughts?

Hello,

I am trying to print out the name of the test executing another test as a test as step and was wondering if this can be done? For example lets say i have two tests "A" and "B". They are both Wpf tests and "B" just contains a coded step. Test "A" has a step where it executes "B" as a Test As Step. Is it possible to get the name of Test "A" through the code?

 

The closest thing I've found was this 

http://www.telerik.com/forums/how-to-get-the-test-name-which-i-am-executing-during-run-time

but would only get me the name of "B".

 

Thanks,

Hi,

    In my form I have two drop-downs. When i Select a value from first drop-down the second one should get enabled.

While playback the first drop-down selection is done but it is throwing a exception.

Below are the Exception Details :
ExecuteCommand failed!
InError set by the client. Client Error:
System.InvalidOperationException: Javascript call [function fireChanged(){var el=document.getElementsByTagName(\'select\')[0]; if(el._events == null || el._events.change == null){return;} var hndlr = el._events.change.length;for(i=0; i<hndlr;i++){el._events.change[i].handler();}}fireChanged()] failed! Please make the function exists and the call is using the correct prototype signature. Javascript error: TypeError: Unable to get value of the property 'target': object is null or undefined 
   at ArtOfTest.InternetExplorer.IECommandProcessor.InvokeFunction(String functionCall, Boolean useEval, Boolean returnJSON)
   at ArtOfTest.InternetExplorer.IECommandProcessor.ProcessActionCommands(BrowserCommand request)
   at ArtOfTest.InternetExplorer.IECommandProcessor.ProcessCommand(WebBrowserClass ieInstance, BrowserCommand request, IHTMLDocument2 document)
BrowserCommand (Type:'Action',Info:'NotSet',Action:'InvokeJsFunction',Target:'ElementId (tagName: '',occurrenceIndex: '-1')',Data:'function fireChanged(){var el=document.getElementsByTagName(\'select\')[0]; if(el._events == null || el._events.change == null){return;} var hndlr = el._events.change.length;for(i=0; i<hndlr;i++){el._events.change[i].handler();}}fireChanged()',ClientId:'Client_4296c26d-43a3-4f94-aca4-9e154a7ca4db',HasFrames:'False',FramesInfo:'',TargetFrameIndex:'0',InError:'True',Response:'System.InvalidOperationException: Javascript call [function fireChanged(){var el=document.getElementsByTagName(\'select\')[0]; if(el._events == null || el._events.change == null){return;} var hndlr = el._events.change.length;for(i=0; i<hndlr;i++){el._events.change[i].handler();}}fireChanged()] failed! Please make the function exists and the call is using the correct prototype signature. Javascript error: TypeError: Unable to get value of the property 'target': object is null or undefined 
   at ArtOfTest.InternetExplorer.IECommandProcessor.InvokeFunction(String functionCall, Boolean useEval, Boolean returnJSON)
   at ArtOfTest.InternetExplorer.IECommandProcessor.ProcessActionCommands(BrowserCommand request)
   at ArtOfTest.InternetExplorer.IECommandProcessor.ProcessCommand(WebBrowserClass ieInstance, BrowserCommand request, IHTMLDocument2 document)')
InnerException: none.


 I tried to fire Javascript/jquery events using coded step but nothing solved my problem.

Can you please help me with this issue?

Thanks
Bhaskar

Hello,

we want to verify that a certain dynamic price is in a certain price range. How can we test situations like this?

 

Example:     153,12 Dollar  -> We want to verify that the price is between 150 to 160 Dollar. Otherwise we want to receive an error message.

 

Thank you very much.

Is it possible to set a timeout on a test case while executing test list ? if a test has not finished executing in defined time period, we should kill the test and active browser (if there is one) and move to next test.

We run our test lists on Jenkins over night, some times due to in stability of tests, test cases are getting stuck on some screens and and execution sits there all night ( ex: open an upload file dialogue but provide in-valid file name) . I know we need to make our tests better, which we are doing. To avoid these kind of issues we are thinking to set a TimeOut on testcase so that if a test case is not finished in that time time period, we should kill that test and kill the active browser session( if there is) and move on to next test. is that possible ? if so how can we achieve this ? any information is appreciated. 

we are using Test Studio Version 2016.3.928.0. 

I have been trying to explore the integrating Test Studio with our current CI (Team City 7.x).

I have been reviewing the existing documentation here.

The type of testing we are wanting to perform would be automated testing of a WPF application.

My question is, is there a way I can set this up to run headless. i.e. Not have a user logged in/remoted in to the machine with the build runner on it?

Thanks in advance for any help.

Josh

Hi Team,

 

I am using Telerik Test Studio 2017.1 and having an issue, whenever I update an element's properties(which is already saved in repository and used by multiple scripts, now I am updating its property) and save it it gives me attached screen. Telerik gets crashed everytime, I have to restart it everytime I am forced to add new element for same verifications.

 

Please look into this.

 

Thanks,

Nishant

Goal: We want to load test our application (url: https://myhealth.alberta.ca/). For now, we just want to do it for 10,000 users. We already have one Test Studio Ultimate Licence with 100 users and also there is one other order under process for 10,000 VUs.

As per my understanding because we need 10,000 users, it wont be possible to support that many users using one machine. So I think we need more run-time licenses to create network of machines (see attachment, I got to know from one of the webinars i attended). Without this, having just 10,000 users wont give us I guess. Please confirm if I am thinking the right way or not.

Also please elaborate a bit on how ultimate and run-time licenses work together. How many users one machine can support. Is it going to be concurrent execution of scripts on different machines ?? Is it only one script per machine we can have or it could be more?

 

Thanks and Regards

Satpreet Birah