Reading an ExcelSheet

The only way you have to do what you want is to create some sort of unit of code that is shared across all your other tests and test cases where you read the excel sheet and basically rebuild, from scratch, on your own, the equivalent of NameMapping. That's basically what you're trying to do... have all the object identification factors preloaded in your test project.

Based upon what you're trying to do, it seems you're trying to basically replicate what happens in Selenium... you're using TestComplete now and there is no need to do XPath identification in code. Map your objects with NameMapping and the object repository is immediately available.

The reason why things are running slow is not necessarily because of reading the Excel... it has EVERYTHING to do with the fact that, for every object, you're doing a scan of the HTML document tree for the XPath... a long process to begin with. Using NameMapping that scan doesn't happen and the object identification is MUCH faster. I'd suggest you look into re-designing your framework to use NameMapping/Aliases instead.

I would suggest, As a first step of driver script create table type project variable(s) for each sheets which you want to access the data in one go.

 

Then wherever you can just getting from project variable. I have been using this approach for long time and i'm getting faster execution and without any hang.

 

Sample code which i'm using,

function fn_getTestCaseObject(gTestDataPath,strSheetName)
{
      try
      {
            var  tempTCExcelObject;
    
            tempTCExcelObject = DDT.ExcelDriver(gTestDataPath,strSheetName,false);
    
            Project.Variables.AddVariable("tblTestData","Table");
    
            for (var i = 0;i< tempTCExcelObject.ColumnCount;i++) 
            {
                  Project.Variables.tblTestData.AddColumn(tempTCExcelObject.ColumnName(i));
            }
            
            tempTCExcelObject.DriveMethod("<UnitName>.fn_driveTestData");
            DDT.CloseDriver(tempTCExcelObject.Name);
      }
      catch(ex)
      {
            Log.Error("Error occured in fn_getTestCaseObject function , Error Description: " + ex.stack);
      }
	
}
function fn_driveTestData()
{
      var intRowCount=1,i,tempValue;
      try
      {
            intRowCount = Project.Variables.tblTestData.RowCount;
            Project.Variables.tblTestData.RowCount = intRowCount + 1;
            
      	for(i = 0 ; i < DDT.CurrentDriver.ColumnCount ; i++)
      	{
                 Project.Variables.tblTestData.$set(DDT.CurrentDriver.ColumnName(i),intRowCount,aqConvert.VarToStr(DDT.CurrentDriver.Value(i)));
      	}
      }
      catch(ex)
      {
            Log.Error("Error occured in fn_driveTestData function , Error Description: " + ex.stack);
      }	
      
}

Are you opening and closing the Excel driver inside your test case loop?  We open it before the loop and close it after the loop and don't seem to be having any timing issues.

 

Why not use name mapping, at least a high level one, for your objects?  It's much faster for TestComplete to access the object from there.