Forum Discussion

ex's avatar
ex
Occasional Contributor
9 years ago
Solved

Execute a group of tests in Network Suite in series

In a Master Network Suite, I can have a task that can run one task on a remote computer. I am unable to assign multiple tasks to the same Job, as they would try to executed on the same Slave machine ...
Network Suite
  • ex's avatar
    ex
    9 years ago

    I did something along those lines, 

     

    I made a helper called Run() 

     

    function Run(){
        for (var i = 0; i < arguments.length; i++)
        { 
            Log.AppendFolder("Folder 1");
            arguments[i]();
            Log.PopLogFolder();
        }
    }

     

    and I run the tests by calling Run(test1,test2,test3) ... 

     

    Thank you for the help!

mtsmith's avatar
mtsmith
Contributor

Good catch, try this, changes are in bold:

 

//USEUNIT Script_1

//USEUNIT Script_2

//USEUNIT Script_3

 

function Main() {
  
   var lResult = 1;
   try {
  
      var asScripts = ["Script_1.Main()","Script_2.Main()","Script_3.Main()"]
     
      for (var i = 0; i < asScripts.length; i++) {
        eval(asScripts[i]);
      }
     
   }     
   catch (ex)  {
      lResult = 0;
   }
   return lResult;

ex's avatar
ex
Occasional Contributor
9 years ago

I did something along those lines, 

 

I made a helper called Run() 

 

function Run(){
    for (var i = 0; i < arguments.length; i++)
    { 
        Log.AppendFolder("Folder 1");
        arguments[i]();
        Log.PopLogFolder();
    }
}

 

and I run the tests by calling Run(test1,test2,test3) ... 

 

Thank you for the help!

  • mjwatkins2's avatar
    mjwatkins2
    Occasional Contributor
    6 years ago

    Suppose your test1 fails or throws an exception. Wouldn't that cause all the other tests to be skipped or also be marked as failed? Or did you find a workaround for that issue?

    • tristaanogre's avatar
      tristaanogre
      Esteemed Contributor
      6 years ago

      If you write your tests with appropriate error/exception handling where they may fail and halt the execution of the one test without over-all impacting the entire test run, then no.

       

      Additionally, if you write your tests with "setup" and "tear down" code for each test within that handling so that, even if an error occurs, the test "tears down" in such a way as to make it possible to run the next test, then no.

       

      It's all in how you write your tests.  Applying good handling practices, making tests as "atomic" as possible, and other such things help to make your tests resiliant and robust.