Functional Pointer in VBScript Abstract Function pointers are used in C and C++ to enable callbacks and as a result, more generic

and parsimonious coding. This article shows how to implement function pointers in VB Script using the Scripting.Dictionary object and the Command Wrapper design pattern. It is also shown how to implement “constructor” functions, callbacks land event handlers.

Keywords: function pointer, callback, delegate, scripting dictionary, hash table, design pattern, command wrapper, event handler. Introduction In the following sections I shall explain several basic concepts that are essential to understand the rationale behind the suggested method. The list might be nonexhaustive and the reader is hereby encouraged to recur to complementary sources to fill-in any eventual knowledge gaps. Function Pointers A function pointer is a variable that stores the memory address of a block of code that is programmed to fulfill a specific function. Function pointers are useful to avoid complex switch case structures; instead, they support direct access in run-time to previously loaded functions or class methods. This enables the construction of callback functions. A callback is, in essence, executable code that is passed as argument to a function. This enables more generic coding, by having lower-level modules calling higher-level functions or subroutines. Function pointers are supported by programming languages like C and C++. A recent good introductory text on the subject with implementation examples is Lars Haendel (2005). This article shows how to implement function pointers in VB script. Design Patterns Design patterns are well established generic design techniques whose aim is to make software design and the derived code more parsimonious, scalable, and reusable, as well as more easily maintainable. This article will delve only into the command wrapper design pattern, which is important for its purpose. For a more extensive explanation of

design patterns, the reader is encouraged to read the literature in the field (e.g., Gamma, Helm, Johnson, & Vlissides, 1997). The command-wrapper design pattern enables to load blocks of code dynamically to memory as well as other uses as will be reviewed below. The singleton design pattern is used to restrict instantiation of a class to one object. This is useful when there is no need for more than one object within a system (e.g., a reporter/logger object). For example, the factory design pattern usually implements the singleton design pattern to have a single interface from which to retrieve references to a specific class object or to a variety of class objects. The Command Wrapper Design Pattern The command wrapper design pattern is a technique that implements a function as a class, usually with a single public method. During run-time, it is possible to instantiate the class and thus to get a reference (or pointer) to the object. This way, the code embodied in such methods can be dynamically loaded, a feature that can be of great value in systems that are poor in resources. This is especially true when the to-be called function is rarely used. After creating the object, it is possible to execute its method according to need. For example, let us assume a function that adds numbers (note: this example is, of course, trivial. It is done just to illustrate the point.)
1: Public Function Sum(ByVal arrNumbers) 2: Dim ix 3: 4: 'Check if an array was passed. If so, there's nothing to do – exit function 5: If (Not IsArray(arrNumbers)) Then 6: 'Add your error handling code here 7: Exit Function 8: End If 9: 10: Sum = 0 11: For ix = LBound(arrNumbers) To UBound(arrNumbers) 12: If (IsNumeric(arrNumbers(ix))) Then 13: Sum = Sum + arrNumbers(ix) 14: Else 15: 'Add your error handling code here 16: End If 17: Next 18: End Function 19: 20: 'Test the function and display result returned by the Sum function 21: MsgBox Sum(Array(23, 56, 78, 95, 114)), vbOKOnly, "Result from simple function" 22: 23: 24: 'Now, let us convert this function into an object using the Command Wrapper design pattern:

25: Class Sum 26: Private m_arrVarNumbers 27: Private m_varResult 28: 29: Private Sub Class_Initialize() 30: 'Initialize the Numbers member as an empty array 31: ReDim m_arrVarNumbers(-1) 32: End Sub 33: 34: Public Function Init(ByVal arrVarNumbers) 35: Numbers = arrVarNumbers 36: End Function 37: 38: Public Default Function Exec() 39: Dim ix, arrNumbers 40: 41: 'Check if an array was passed. If so, there's nothing to do – exit function 42: If (Not IsArray(Numbers)) Then 43: 'Add your error handling code here 44: Exec = "Invalid data type was supplied to perform the operation." 45: Exit Function 46: ElseIf (UBound(Numbers) - LBound(Numbers) + 1 <= 1) Then 47: 'Array is empty or has single item - Add your error handling code here 48: Exec = "Not enough data was supplied to perform the operation." 49: Exit Function 50: Else 51: arrNumbers = Numbers 52: End If 53: 54: Result = 0 55: For ix = LBound(arrNumbers) To UBound(arrNumbers) 56: If (IsNumeric(arrNumbers(ix))) Then 57: Result = Result + arrNumbers(ix) 58: Else 59: 'Add your error handling code here 60: End If 61: Next 62: Exec = Result 63: End Function 64: 65: Public Property Get Numbers() 66: Numbers = m_arrVarNumbers 67: End Property 68: 69: Private Property Let Numbers(ByVal arrVarNumbers) 70: m_arrVarNumbers = arrVarNumbers 71: End Property 72: 73: Public Property Get Result() 74: Result = m_varResult 75: End Property 76: 77: Private Property Let Result(ByVal varResult) 78: m_varResult = varResult

79: End Property 80: End Class 81: 82: 'This function behaves as a constructor and returns an initialized instance of the class 83: Public Function GetSum(ByVal arrNumbers) 84: Set GetSum = New Sum 85: GetSum.Init(arrNumbers) 86: End Function 87: 88: 'Test the class 89: Dim objSum, arrNumbers 90: 91: 'Assign an array of numbers 92: arrNumbers = Array(23, 56, 78, 95, 114) 93: 94: 'Get an instance of class Sum and initialize it with numbers array 95: Set objSum = GetSum(arrNumbers) 96: 97: 'Execute Sum (Exec method) 98: objSum.Exec() 99: 100: 'Display result stored in Result property 101: MsgBox objSum.Result, vbOKOnly, "Result from Result Property" 102: 103: 'Alternatively, display the result returned by the Exec method: 104: MsgBox GetSum(Array(23, 56, 78, 95, 114)).Exec, vbOKOnly, "Result from Exec method"

The Scripting Dictionary In VB Script, the scripting dictionary is an object that stores key-item pairs using a hashing algorithm. The items can be accessed using their corresponding keys. Basically, the dictionary is useful to store variant type, non-structured data, such as the catalog number of an item in an inventory list, or the authors of a list of books, and to retrieve the data using the unique keys supplied, as in the following example:
1: Dim dicBooksAuthors 2: 3: 'Create an instance of the scripting dictionary class 4: Set dicBookAuthors = CreateObject("Scripting.Dictionary") 5: 6: With dicBookAuthors 7: 'Add some books (keys) and authors (items) 8: .Add "The Selfish Gene", "Richard Dawkins" 9: .Add "The Mismeasure of Man", "Stephen J. Gould" 10: .Add "The Da Vinci Code", "Dan Brown" 11: 12: 'Answer the question: who wrote "The Selfish Gene"? 13: strBook = "The Selfish Gene" 14: MsgBox .item(strBook), vbOKOnly+vbInformation, "Author query result for book: " & strBook 15: End With

In the next chapter. For instance. it is actually possible to store references to objects of different types. arrNumbers 16: 17: If (Not IsArray(Numbers)) Then 'Not an array. so nothing to do – exit function 18: 'Add your error handling code here . Recall that a pointer is a variable that stores the memory address of a function or instantiated class object. we shall delve into a method to exploit this feature of the scripting dictionary object to implement function pointers in VBScript.The example above demonstrates the power of a dictionary as a data storage and data retrieval device. is out of the scope of the current paper). Thanks to its capacity to store variant data types. it has been indicated that the VBScript Scripting Dictionary object is equipped with the capacity of storing variant data types. Here below. The Addition Operation 1: Class Sum 2: Private m_arrVarNumbers 3: Private m_varResult 4: 5: Private Sub Class_Initialize() 6: 'Initialize the Numbers member as an empty array 7: ReDim m_arrVarNumbers(-1) 8: End Sub 9: 10: Public Function Init(ByVal arrVarNumbers) 11: Numbers = arrVarNumbers 12: End Function 13: 14: Public Default Function Exec() 15: Dim ix. however. we shall now show how it is possible to build function pointers using a Scripting Dictionary. In addition. In what follows a method will be described to combine these two features into a powerful design and programming technique. For instance. it eliminates the need for item search functions (direct access) and avoids duplicate key creation. Implementing Function Pointers in VB Script Earlier in this article. including references to objects. we shall demonstrate the case of the four basic arithmetic operations. However. Following the example given above in the Command Wrapper section. the dictionary can be much more than just a data storage and retrieval device. we have shown that it is possible to encapsulate any function as a class by using the command wrapper design pattern. it is possible to store complex data structures by using nested dictionaries (this issue.

19: 20: 21: 22: 23: 24: 25: 26: 27: 28: 29: 30: 31: 32: 33: 34: 35: 36: 37: 38: 39: 40: 41: 42: 43: 44: 45: 46: 47: 48: 49: 50: 51: 52: 53: 54: 55: 56: 57: 58: 59: 60: 61: Exec = "Invalid data type was supplied to perform the operation.Init(arrNumbers) End Function The Subtraction Operation 1: Class Subtract 2: Private m_arrVarNumbers 3: Private m_varResult 4: 5: Private Sub Class_Initialize() 6: 'Initialize the Numbers member as an empty array ." Exit Function Else arrNumbers = Numbers End If Result = 0 For ix = LBound(arrNumbers) To UBound(arrNumbers) If (IsNumeric(arrNumbers(ix))) Then Result = Result + arrNumbers(ix) Else 'Add your error handling code here End If Next Exec = Result End Function Public Property Get Numbers() Numbers = m_arrVarNumbers End Property Private Property Let Numbers(ByVal arrVarNumbers) m_arrVarNumbers = arrVarNumbers End Property Public Property Get Result() Result = m_varResult End Property Private Property Let Result(ByVal varResult) m_varResult = varResult End Property End Class 'This function behaves as a constructor and returns an initialized instance of the class Public Function GetSum(ByVal arrNumbers) Set GetSum = New Sum GetSum.LBound(Numbers) + 1 <= 1) Then 'Array is empty or has single item .Add your error handling code here Exec = "Not enough data was supplied to perform the operation." Exit Function ElseIf (UBound(Numbers) .

arrNumbers(ix) 33: Else 34: 'Add your error handling code here 35: End If 36: Next 37: Exec = Result 38: End Function 39: 40: Public Property Get Numbers() 41: Numbers = m_arrVarNumbers 42: End Property 43: 44: Private Property Let Numbers(ByVal arrVarNumbers) 45: m_arrVarNumbers = arrVarNumbers 46: End Property 47: 48: Public Property Get Result() 49: Result = m_varResult 50: End Property 51: 52: Private Property Let Result(ByVal varResult) 53: m_varResult = varResult 54: End Property 55: End Class 56: 57: 'This function behaves as a constructor and returns an initialized instance of the class 58: Public Function GetSubtract(ByVal arrNumbers) 59: Set GetSubtract = New Subtract 60: GetSubtract.Init(arrNumbers) ." 24: Exit Function 25: Else 26: arrNumbers = Numbers 27: End If 28: 29: Result = arrNumbers(LBound(arrNumbers)).LBound(Numbers) + 1 <= 1) Then 22: 'Array is empty or has single item .arrNumbers(LBound(arrNumbers)+1) 30: For ix = LBound(arrNumbers)+2 To UBound(arrNumbers) 31: If (IsNumeric(arrNumbers(ix))) Then 32: Result = Result . arrNumbers 16: 17: If (Not IsArray(Numbers)) Then 'Not an array. so nothing to do – exit function 18: 'Add your error handling code here 19: Exec = "Invalid data type was supplied to perform the operation." 20: Exit Function 21: ElseIf (UBound(Numbers) .Add your error handling code here 23: Exec = "Not enough data was supplied to perform the operation.7: ReDim m_arrVarNumbers(-1) 8: End Sub 9: 10: Public Function Init(ByVal arrVarNumbers) 11: Numbers = arrVarNumbers 12: End Function 13: 14: Public Default Function Exec() 15: Dim ix.

" 24: Exit Function 25: Else 26: arrNumbers = Numbers 27: End If 28: 29: Result = arrNumbers(LBound(arrNumbers)) * arrNumbers(LBound(arrNumbers)+1) 30: For ix = LBound(arrNumbers)+2 To UBound(arrNumbers) 31: If (IsNumeric(arrNumbers(ix))) Then 32: Result = Result * arrNumbers(ix) 33: Else 34: 'Add your error handling code here 35: End If 36: Next 37: Exec = Result 38: End Function 39: 40: Public Property Get Numbers() 41: Numbers = m_arrVarNumbers 42: End Property 43: 44: Private Property Let Numbers(ByVal arrVarNumbers) 45: m_arrVarNumbers = arrVarNumbers 46: End Property 47: 48: Public Property Get Result() . so nothing to do – exit function 18: 'Add your error handling code here 19: Exec = "Invalid data type was supplied to perform the operation.61: End Function The Multiplication Operation 1: Class Multiply 2: Private m_arrVarNumbers 3: Private m_varResult 4: 5: Private Sub Class_Initialize() 6: 'Initialize the Numbers member as an empty array 7: ReDim m_arrVarNumbers(-1) 8: End Sub 9: 10: Public Function Init(ByVal arrVarNumbers) 11: Numbers = arrVarNumbers 12: End Function 13: 14: Public Default Function Exec() 15: Dim ix.LBound(Numbers) + 1 <= 1) Then 22: 'Array is empty or has single item .Add your error handling code here 23: Exec = "Not enough data was supplied to perform the operation. arrNumbers 16: 17: If (Not IsArray(Numbers)) Then 'Not an array." 20: Exit Function 21: ElseIf (UBound(Numbers) .

arrNumbers 16: 17: If (Not IsArray(Numbers)) Then 'Not an array.Add your error handling code here 23: Exec = "Not enough data was supplied to perform the operation.LBound(Numbers) + 1 <= 1) Then 22: 'Array is empty or has single item ." 24: Exit Function 25: Else 26: arrNumbers = Numbers 27: End If 28: 29: If (IsNumeric(arrNumbers(LBound(arrNumbers))) And IsNumeric(arrNumbers(LBound(arrNumbers)+1)) And (arrNumbers(LBound(arrNumbers)+1) <> 0)) Then 30: Result = arrNumbers(LBound(arrNumbers)) / arrNumbers(LBound(arrNumbers)+1) 31: Else 32: 'Add your error handling code here 33: Exec = "Invalid data was supplied to perform the operation.49: 50: 51: 52: 53: 54: 55: 56: 57: 58: 59: 60: 61: Result = m_varResult End Property Private Property Let Result(ByVal varResult) m_varResult = varResult End Property End Class 'This function behaves as a constructor and returns an initialized instance of the class Public Function GetMultiply(ByVal arrNumbers) Set GetMultiply = New Multiply GetMultiply." 34: Exit Function 35: End If . so nothing to do – exit function 18: 'Add your error handling code here 19: Exec = "Invalid data type was supplied to perform the operation." 20: Exit Function 21: ElseIf (UBound(Numbers) .Init(arrNumbers) End Function The Division Operation 1: Class Divide 2: Private m_arrVarNumbers 3: Private m_varResult 4: 5: Private Sub Class_Initialize() 6: 'Initialize the Numbers member as an empty array 7: ReDim m_arrVarNumbers(-1) 8: End Sub 9: 10: Public Function Init(ByVal arrVarNumbers) 11: Numbers = arrVarNumbers 12: End Function 13: 14: Public Default Function Exec() 15: Dim ix.

Add "/". New Divide 12: End With 13: 14: 'Execute the functions using the Function Handler 15: With dicFunctionHandler 16: With . New Subtract 10: .Add "*".Init(arrNumbers) End Function Using the Function Pointers to Perform the basic Arithmetic Operations 1: Dim dicFunctionHandler 2: 3: 'Create an instance of the scripting dictionary class 4: Set dicFunctionHandler = CreateObject("Scripting. New Sum 9: .Dictionary") 5: 6: 'Load some functions 7: With dicFunctionHandler 8: .Add "+".36: 37: 38: 39: 40: 41: 42: 43: 44: 45: 46: 47: 48: 49: 50: 51: 52: 53: 54: 55: 56: 57: 58: 59: 60: 61: 62: 63: 64: 65: 66: 67: 68: For ix = LBound(arrNumbers)+2 To UBound(arrNumbers) If (IsNumeric(arrNumbers(ix)) And (arrNumbers(ix) <> 0)) Then Result = Result / arrNumbers(ix) Else 'Add your error handling code here End If Next Exec = Result End Function Public Property Get Numbers() Numbers = m_arrVarNumbers End Property Private Property Let Numbers(ByVal arrVarNumbers) m_arrVarNumbers = arrVarNumbers End Property Public Property Get Result() Result = m_varResult End Property Private Property Let Result(ByVal varResult) m_varResult = varResult End Property End Class 'This function behaves as a constructor and returns an initialized instance of the class Public Function GetDivide(ByVal arrNumbers) Set GetDivide = New Divide GetDivide.Add "-". New Multiply 11: .item("+") .

Init(Array(23. vbOKOnly+vbInformation. GetSubtract(Array(117. 56. 78. . "Result (/)" 'Display result returned by the Exec method 50: End With In the above example we have shown how to: 1. Implement the Command Wrapper design pattern in VBScript. "Result (*)" 'Display result returned by the Exec method 49: MsgBox .Exec().Exec(). using the "constructors": 35: 'First. 95.item("*") 25: Call . 12)) 42: End With 43: 44: 'Second.item("/") 29: Call . "Result (/)" 'Display result returned by the Exec method 31: End With 32: End With 33: 34: 'Or. GetSum(Array(23. 114)) 39: . GetMultiply(Array(7. vbOKOnly+vbInformation. Call the loaded function via the dictionary key and retrieve the result. vbOKOnly+vbInformation.item("+").Exec(). 114)) 18: MsgBox . load some functions 36: With dicFunctionHandler 37: . Implement a “constructor” for a class in VBScript. 2. 12)) 30: MsgBox . "Result (+)" 'Display result returned by the Exec method 19: End With 20: With . vbOKOnly+vbInformation. 3.Add "-". "Result (*)" 'Display result returned by the Exec method 27: End With 28: With . vbOKOnly+vbInformation.item("-"). 78.item("*").Exec(). vbOKOnly+vbInformation.item("-") 21: Call . 5)) 41: .Exec(). vbOKOnly+vbInformation.17: Call .Init(Array(7. 95.Init(Array(84.Add "+".Add "*". 56. Implement a function handler using a scripting dictionary. vbOKOnly+vbInformation. "Result (-)" 'Display result returned by the Exec method 23: End With 24: With . Instantiate such custom classes and load them to a scripting dictionary. 23)) 40: . execute the functions using the Function Handler 45: With dicFunctionHandler 46: MsgBox .Init(Array(117.Exec().Exec(). 5)) 26: MsgBox . 23)) 22: MsgBox .item("/"). 4. "Result (+)" 'Display result returned by the Exec method 47: MsgBox .RemoveAll 38: . 5. GetDivide(Array(84.Add "/". "Result (-)" 'Display result returned by the Exec method 48: MsgBox .Exec().

In a similar fashion. "Result "Result "Result "Result (+)" (-)" (*)" (/)" and this is because the item property is the scripting dictionary‟s default property.Exec(). dicFunctionHandler(“-”).0674 4. vbOKOnly+vbInformation.1219 Discussion We have seen so far how to implement the Command Wrapper design pattern in VBScript and a “constructor” for a class in VBScript.Exec(). "Result "Result "Result "Result (+)" (-)" (*)" (/)" The readers are encouraged to try to execute the sample code shown in this paper. the more worthwhile it is to use function pointers. vbOKOnly+vbInformation. vbOKOnly+vbInformation. as follows: 1: 2: 3: 4: 5: 'Execute the functions using the Function Handler MsgBox MsgBox MsgBox MsgBox dicFunctionHandler(“+”). together with different ways to call the loaded . as well as a function handler using a scripting dictionary. dicFunctionHandler(“/”). it is possible to define the Exec methods of the above mentioned classes as Default (by declaring it: Public Default Function) and then the code above can be further reduced to: 1: 2: 3: 4: 5: 6: 'Execute the functions using the Function Handler MsgBox MsgBox MsgBox MsgBox dicFunctionHandler(“+”).0635 0. dicFunctionHandler(“-”). vbOKOnly+vbInformation.0943 Function Pointers 0.0922 100 5.This is the method suggested in this paper to implement a function pointer in VBScript. vbOKOnly+vbInformation. vbOKOnly+vbInformation. dicFunctionHandler(“/”). There is also another possible way to call a function implemented with this method. We have also shown how to instantiate such custom classes and load them to a scripting dictionary. vbOKOnly+vbInformation.1561 10 0.Exec().5806 2. Number of Calls Single Regular calls 0. vbOKOnly+vbInformation.9212 50 2. dicFunctionHandler(“*”).5630 0.Exec().5308 20 1. Performance of Function Pointers A small case study comparing the performance of regular function calls with that of function pointers shows that the more a function is used. dicFunctionHandler(“*”). as well as to try the method to implement their own functionality.

The Function Pointer Tutorials. 2005).. such as an XML file. it would be possible to build a generic controller component that would enable execution of real keyword-driven scripts. in fact. equivalent to the implementation of a function pointer in C or C++.function via the dictionary key and to retrieve the result. References Haendel.newty. which is a central feature of function pointers. Recall that common QTP usage requires the automation developer to add every single function library to the test resources. First. Lars Haendel. for it makes the need for decision structures to analyze the return code of a function obsolete. Fourth. the method can be used to emulate callbacks. when the system is low in RAM. Conclusion This paper attempted to demonstrate how to implement function pointers in VBScript. to define the flow of the different code blocks in an external file. by implementing a function handler as illustrated above. This can be easily done by passing the function handler entry (dictionary item) to another function. might be of benefit. Please notice that this technique would yield a highly parsimonious coding style. With such a component it would be by means of dynamic loading of code blocks. and with Quicktest Pro (QTP) in particular. Source: http://www..e. Callbacks and Functors. An example for this can be found in Appendix 2. This can be achieved by returning a string with the name of the function to be called. and hence they will not be covered here.g. The general uses and benefits of function pointers are explained elsewhere (e. In what remains I shall attempt to convey in which cases the implementation of this design pattern in VBScript in general.pdf . for example. L. the method presented in this paper should be of great value when the basic hardware configuration is poor. We have also indicated that this method is. (2005). and pinpointed the possible uses and advantages of the method in general and particularly for QTP. i. It is concluded that the technique can help developers to achieve more efficient and generic design and code. and better run-time resources management. Second. Future forthcoming articles will further expand on several topics mentioned throughout this paper. the method can be used to emulate event handling. Third. Introduction to C and C++ Function Pointers.

Gamma. Helm. in case of error the Login. & Vlissides. In addition. R. This action is then performed by means of the statement: strNextAction = dicFunctionHandler(strNextAction)(arrArgs). Within the TestFlow Sub. vbOKOnly+vbInformation. Appendix 1: Event Handling Example We shall show here how to implement an event handler in VBScript for a typical login flow in QTP.. J.Exec function returns the name of the action to be performed (AbortTest). vbOKOnly+vbInformation. E. FirstTest and Logout. vbOKOnly+vbCritical."). As can be seen in the code. ".. two additional functions (AbortTest and Cleanup) are included for error handling and orderly exiting the test flow. "Login" 10: Exec = "" 11: End If 12: End Function 13: End Class 14: 15: Class Logout 16: Public Default Function Exec(ByVal arrArgs) 17: MsgBox "Exec". _ 6: vbOKOnly+vbCritical. 1: Class Login 2: Public Default Function Exec(ByVal arrArgs) 3: 'Add your code here 4: If (Not IsArray(arrArgs)) Then 5: MsgBox "Error: username & password are mandatory parameters. R. the functions are loaded using a generic function (class) loader function (GetFunction). "Logout" 18: 'Add your code here 19: End Function 20: End Class 21: 22: Class AbortTest 23: Public Default Function Exec(ByVal arrArgs) 24: MsgBox "Exec". Design Patterns: Elements of Reusable Object Oriented Software. and since the for loop continues to advance. the normal course of the test flow is altered such that the FirstTest action is never executed. "AbortTest" 25: Exec = "Cleanup" 26: 'Add your code here 27: End Function 28: End Class 29: 30: Class FirstTest . "Login" 7: Exec = "AbortTest" 8: Else 9: MsgBox Join(arrArgs. and then called according to their order in the arrRunActions.. Addison-Wesley Publishing. The flow includes three functions: Login. (1997). Johnson.". with their corresponding arguments from arrActionArgs.

vbOKOnly+vbInformation. ByVal arrActionArgs) Dim dicFunctionHandler. "")) .Add arrLoadActions (ix). "Cleanup"). pass an empty string arrArgs = "" End If 'If the next command is empty If (strNextAction = "") Then 'Run the next planned action strNextAction = dicFunctionHandler(arrRunActions(ix))(arrArgs) Else 'Run the action returned by the previously call action strNextAction = dicFunctionHandler(strNextAction)(arrArgs) End If Next End Sub Call TestFlow(Array("Login". "FirstTest". GetFunction(arrLoadActions (ix)) Next End With 'Run the required flow strNextAction = "" For ix = LBound(arrRunActions) To UBound(arrRunActions) 'Get the action arguments arrArgs = split(arrActionArgs(ix). ix.31: 32: 33: 34: 35: 36: 37: 38: 39: 40: 41: 42: 43: 44: 45: 46: 47: 48: 49: 50: 51: 52: 53: 54: 55: 56: 57: 58: 59: 60: 61: 62: 63: 64: 65: 66: 67: 68: 69: 70: 71: 72: 73: 74: 75: 76: 77: 78: 79: 80: 81: 82: 83: 84: Public Default Function Exec(ByVal arrArgs) MsgBox "Exec". "AbortTest". "FirstTest" 'Add your code here End Function End Class Class Cleanup Public Default Function Exec(ByVal arrArgs) MsgBox "Exec". _ Array("Login".LBound(arrArgs) + 1 = 0) Then 'If no args are found.123456". "FirstTest". arrArgs. "Logout"). strNextAction Set dicFunctionHandler = CreateObject("Scripting. Array("User.Dictionary") 'Load the required functions With dicFunctionHandler For ix = LBound(arrLoadActions) To UBound(arrLoadActions) . "". vbOKOnly+vbInformation. ByVal arrRunActions. "Logout". ".Dictionary") Set arrArgs = CreateObject("Scripting. "Cleanup" 'Add your code here End Function End Class 'This generic function will load any class Public Function GetFunction(ByVal strFunctionName) Execute "Set GetFunction = New " & strFunctionName End Function 'This sub will run the flow Public Sub TestFlow(ByVal arrLoadActions.") If (UBound(arrArgs) .

while not having any testing value by themselves. messing with the application‟s RO objects has no testing value (not emulating user behavior. QTP doesn‟t recognize this tab-control. it could cause serious bugs. that calls for serious discussion.g. CONST_DBL_VAT)). On the other hand. On the one hand. 1: 'Define a constant with the tax value multiplier 2: Const CONST_DBL_VAT = 1. I think this opinion is quite easy to understand. 78. it becomes obvious that the only method that will work is TabControl. this sucks.). here).Appendix 2: Implementing a Callback with Function Pointers We shall show here how to implement a callback with the classes defined in Appendix 1 and used in this paper. slow system behavior etc. RO Vs THE QTP SHELL Update: this is an article from the old site that wasn‟t transferred to the new one. 114)). 56. by dynamically loading the functions without the function handler (as shown earlier in this article): 5: 'Execute the (*) function on the result of the (+) function to calculate the tax on an item 6: MsgBox GetMultiply(Array(GetSum(Array(23. The example used here will be the calculation of the total amount plus the tax on the sum of several items prices. Thus. the question whether it is “Kosher” to change the run-time properties of the AUT (application under test) was given new life in SQAForums (e. that could otherwise fail our script. I side with the latter opinion.155 3: 4: 'Now. but it is very useful. and I think an example will illustrate why: Say our application has some input fields. "Result (Tax)" 8: True. Now. and it is very . when done carefully and selectively. _ 7: vbOKOnly.Click X Y. for two main reasons (beside being an ugly solution): It tends to break (Click events are often block by focus shifting. and could miss UI bugs a user would have encountered. and after many attempts to solve the problem elegantly. setting RO properties can give us a robust way of doing certian tasks. residing within a tab-control. backstage behavior should be tested via unit-tests). It‟s a tough question. the syntax looks complicated. So here it is: In past few days. 95. these task could safely get us to a position where we could do some real testing. divided among various tabs.

and switch. we ask ourselves if we could miss UI bugs due to working backstage. add a new tab. different tab layout) We paid for it with: A. resolutions etc. the tab control‟s inner object change its ActiveTab property to the new tab‟s index.Object. VERY small possibility for missing tab-switching UI bugs due to not actually pressing the tab control. Here comes the “carefully and selectively” part (which is REALLY important): we make damn sure that setting the object‟s RO property doesn‟t break anything. different tab ordering. Moreover. and after a while. we gained: A. VERY small possibility for creating new tab-switching bugs due to messing around with the RO objects.) C. this is unacceptable. It suddenly seems very tempting to just set the RO property whenever we want to switch tabs. we become confident that no new bugs will be caused by this.Click breaking the script) B. we learn that the process works the other way as well – changing the property via QTP. We can run through the tab collection. so we try thinking outside the box. B. Zero-maintenance growth and change (handles “out of the box” New tabs. and the script breaks). good-old-fashion QTP style. makes the control bring the relevant tab to the front. OK. run a few tests of ourselves. We ask the programmers. A generic script (serves multiple systems with various fonts. do a littlie investigating. It doesn‟t test the tab control.Tabs collection). So. So. A much more robust script (no . NEVER on a data-grid control). . get its index. where we could rip the application apart. but it makes sure we‟ll get to the right tab. At the end of the day.VERY demanding from a maintenance point of view (just change the tab-order. find the tab with the relevant name (By looping the TabControl. Just to be clear – I agree we do need to be selective and careful about applying this methond (e. And this seems like a pretty good balance to me. After a little creative fiddling. but we shouldn‟t abandon it all together. change the resolution or the font. we can then decide that tab navigation will be done via RO manipulation. to sum-up. and answer a reasonable NO (since any bugs will cause the next step of the script to fail).g. and find out that when the user switches the active-tab.

A test object stores the description of the referenced object. Test Objects and Run-Time Objects A test object encapsulates a reference to a run-time object – the real world object that is instantiated in the application under test (AUT). Introduction A central issue in automation development is how to reduce the effort spent on maintenance. I shall attempt to show that QTP‟s Object Repository does not fit well with such principles and what is the impact of this on automation development viz-a-viz costeffectiveness and maintainability. Implementing a GUI Layer with Classes Abstract This article describes a powerful technique that exploits Object Oriented Design Patterns. then what is the optimal way to implement it?” are quite ubiquitous and the answers given can vary depending on the particular characteristics of the project. This concept has been adopted by SOLMAR‟s automation experts based on the observation that it is possible to maximize code reusability (and thus boosting maintainability) by undertaking an OO approach that breaks an automation project into several abstraction levels.You can read more about RO manipulating techniques in our article about automating custom controls. The article includes a valuable bonus: a highly effective tip that enables to exit a test smoothly. In this article I will analyze the concept of Test Objects in the context of Object Oriented principles. QTP descriptive programming (DP) and the Dictionary object to pack together GUI objects with their corresponding business oriented functions. Subsequently. Questions such as: “Should we use an Object Repository (OR) or Descriptive Programming (DP)? if an OR is chosen. but many times also on the individual personalities involved. then should we use a shared OR or a local OR for each Action? If DP is selected. I will describe an expansion of the OR concept that is compatible with OO principles – the GUI Layer. which is a set of attributes and values that serve as criteria to . preventing QTP from getting stuck when it fails to identify GUI objects during runtime.

the one that returns the record. Test Complete. this record contains a single field that stores a reference to the actual runtime object. What do I mean by this? Based on a description composed of a set of properties and values. Similarly. Test Objects and the Object Repository The above account of test objects makes it seem only natural to think of them as data entities. as you would with the brunette. Rational Robot. etc. Moreover. This is similar to the retrieval of a particular record from a database table based on a SQL WHERE clause. if used properly. reflects a best practice according to which we ought to store the description of each GUI (Test) Object only once to reduce . What.). QTP uses such a description to query the OS if there‟s an object currently loaded to the machine‟s RAM. and will be wearing a red dress and shoes”. If the answer is positive. insomuch that the above account leaves us with an incomplete picture. not the DB. if you noticed the bdb suffixed files in QTP tests. seems to be the problem with the OR approach? Tests Objects are really data entities. It is true that the OR. since the private methods and properties are hidden from view by definition (so QTP never makes progress to the private domain. basically it is this process that enables us to interact with the GUI (and other) objects using QTP or other automation tools (Test Partner. it is the OS (Windows). you‟ve been told on the phone). then note that bdb actually stands for Berkeley Data Base – an Oracle product). QTP is able to identify a GUI object among all the objects currently loaded to the machine‟s RAM. The difference is that in this case. Though this description may seem overly simplified. slim and tall. which essentially is exactly what the Object Repository is (a side note. AutoIt.locate the right object during the test run session. but at the end of the day we need to perform actions on Runtime objects. if lucky!). Following this line of thought. This description is pretty much like the one we keep in mind to identify our blind date partner at the meeting time (“I‟m brunette. A more detailed account of the distinction between Test Objects and Run-Time Objects is available in Yaron Assa‟s article “Differences and Connections between Runtime Objects and Test Objects” (2009). then. the OS returns the address of the object (a reference or pointer) and hence QTP gains access to the object‟s public methods and properties. I will explain below why. it seems only natural to have these entities – Test Objects – stored in a database.

because ad-hoc requests and resource limitations can endanger the integrity of such a precious resource. However. classes mostly are representations of . inheritanceand polymorphism. this requires good project management. So. GUI changes reflect modifications in the way the AUT actually functions. as aforesaid. And this is a relatively good result. changes in the AUT would require matching changes in both the OR and the scripts that refer to its test objects. but one is maintaining the scripts for the last version and the other is developing new scripts for the newest one. Moreover. redundant. in order to achieve this a great effort must be invested also into managing the automation project properly. Because automated tests are. The Object Oriented View The Object Oriented software paradigm is based on the basic concepts of encapsulation. which is not always available. the aforementioned line of analysis lead me to the conclusion that something in the OR concept is faulty regarding the manageability of a large-scale automation project. as well as in the data sources that feed these scripts. For instance. and the package through which we achieve this is called a class. then using an OR would be an excellent solution. as well as the input data and expected results data. not developed. for example. regressive by nature. in general. But. However. more often than not. it will not be too long until the OR becomes cluttered with new. test objects required by the developer working on the newest version. keeping versions of the OR leaves one issue still open. Of course. Given that GUI changes are made quite regularly from one AUT version to the next. such as keeping separate versions of the OR that match each AUT version and using configuration management software in the automation project. which are many times also accompanied by even deeper changes – in the database structure. If the common situation would have been that only the object descriptions changed from one AUT version to the next. In fact.maintenance costs. The worst case would be one where each developer would change in turn the descriptions of existing objects to match their requirements. suppose that two automation developers work simultaneously with a shared repository. damaging irrevocably the ROI on the automation project as a whole. there are solutions to such situations. Encapsulation basically means to pack together functionality and data structures that belong together. then maintenance would certainly pertain not only to the OR but also that the code that implements the actual GUI manipulations and verifications. Taking into account the fact that scripts are most of the time maintained.

scalable.e. we shall see later that this is not a reason to refrain from using classes in testing automation.. Hence in our code that uses these functions we would not need to use casting. yes. getCustomerId. phoneNo. The difference between a fully featured class and a mere data structure (as in the C programming language) is that the class also wraps the functions available for use with the class data fields. maintainable. For example. These are commonly referred to as the classmethods.) together with functions such assetCustomerId. It is recognized as one approach that makes the resulting code more reusable. etc. extensible and. lastName. the class fields and methods). readable. However in QTP these two last concepts of inheritance and polymorphism cannot be implemented based on the VB Script language which provides very rudimentary support for working with classes. also more . as well as others like getCustomerAge and getCustomerBalance which would perform a calculation based on the class fields current values and then return the result. firstName. The public and private notions mentioned above are of great importance with regard to encapsulation and entities (like Customer. Polymorphism is yet another powerful concept that enables one to design more than one version of a function (with the same name but different number or types of arguments) to be able to handle transparently different situations within the application context. reusability comes into play by means of the ability to device a new class that “inherits” the fields and methods of a previously defined class (also called a base class) and expands on these according to the specific requirements that raised the need for the new class. we‟ll have several functions with the same name but different signatures (different type of arguments). The OO methodology is not new and has been widely put into practice in the software design and development industry. with the correct process ultimately invoked by the runtime environment. For instance. a typical class for a Customer would pack the data fields that define the customer as a data entity (customerId. we would use the same interface in both cases. we might need to handle numbers of different types (float. etc. Inheritance is deeply related to one of the major goals of the OO approach – reusability. which assign and retrieve the value of the respective field. In OO programming languages like C++ and Java. Nevertheless. and which will remain for internal (private) use of the class members (i. int) and so instead of having a single function with conditional statements that check for the value type that was passed.). They enable the code developer to determine which fields (also called properties or attributes of the class) and methods will be actually accessible to the world outside the class (public). Product.

Scripts that carry out operations on GUI objects – such as QTP tests – are not different than any other piece of software. This reflects a misunderstanding of the role of the Automation Developer and. Generally speaking. First. Put it in other words. As with any software product. then an automation project should be treated and managed as any other development project and. it is not rare to find automation professionals that think likewise since many times the role of automation. even more so. Testing Automation and Software Development One thing I have witnessed throughout my career is that QA professionals of all levels treat testing automation as just another testing activity. Members of the AUT‟s development team do . or should approach. or robot. self-contained capsules or packages).testable (as huge pieces of functionality are broken down into smaller. as I shall explain in what follows. computer programs that perform operations instead of humans implement automation. How this relates to the way we approach. an automation project also has its own SRS (Software Requirements Specifications) document: the STD (Software Test Design) “document” (or tests design in a quality management tool such as HP‟s Quality Center or Orcanos‟ QPack). Some of the main challenges are the following: 1. a testing automation project is. If so. the automation developer must have an overall view of the AUT. in fact. as the automation scripts may cover a large part of the AUT‟s main functionality. Not surprisingly. the testing automation challenge? Let us delve into it below. which should guide the automation developer in the implementation of the required code. The task of automating tests should not be treated differently than any other content specific automation task. So basically any block of code is a kind of automation device. Sometimes this is even reflected by the job title given to the automation staff: “automatic tester” is one of the absurd titles I have encountered several years ago. indeed. Of course. This is because the automation developer faces challenges that are usually not felt by the developers of the AUT. this belief is a true misconception of the essence of testing automation. a specific kind of software development project. making good use of the methodology requires deep analytical skills that enable the code designer to infer correctly from the requirements of the software under construction which are the appropriate entities involved and how they are interrelated.

I will illustrate below how such a layer can be built. In a sense. Many times solutions that extend the basic capacities of the tools are required. the automation scripts should be mapped to the test design. So a developer can focus on the part assigned to him and do it well even without having in depth knowledge of the whole system. to keep it short. with scripts spread all around the place. why should one fall again and again victim to the fallacy of treating an automation project as a trivial task (does “record-replay” remind you of something?). In other words. But. is far from being the optimal approach. let us take a look at the concept of developing code in layers. Third. as they should emulate the steps done by a human tester. then. and also the AUT‟s behavior may pose a real technological challenge regarding the identification of the objects by QTP (or other tools). as the automation scripts do not possess the flexibility and ingenuity of a human tester during runtime. Following the above exposition. automation developers find out that the test design leaves many open issues that need to be resolved beforehand. In general..not necessarily need this. the test objects proper) and the Business or Application Layer. Second. and what are the benefits gained from adopting such an approach to testing automation. 2. which is based on solid OO principles. I shall define a GUI Layer as a set of classes that pack (encapsulate) together the interfaces required to manipulate the AUT‟s GUI objects in each application context. where the opposite is true? Why not. of which we shall have more to say later. Implementing the GUI Layer . but it is already accustomed among the experts to call it a GUI Layer. the GUI controls used by the development team. The Concept of Layers I hope to have managed so far to make clear why the OR approach. especially the GUI Layer concept. quite often. Now. may be it would be more appropriate to call it the GUI-Business Bridge Layer.e. as is the case with third-party and custom controls. layers are useful to maximize reusability (recall the previous discussion on OO principles). it is a set of classes that bridge between the AUT‟s GUI (i. If this is so. approach to do it using the most widely accepted development method – Object Oriented – to attain the best possible results? In what follows I will describe a method to implement automation “scripts” based on an extension to the OR concept – the GUI Layer. following the above discussion. I think we may conclude that an automation project should be definitely managed as a development project. as the division of labor among the different teams is typically coordinated by the team leaders and a project manager. 3.

Dictionary to store the references to the test objects contained in the window. to define the members of each class. obviously. The first step would be to make a simple list of all application GUI contexts – i. as follows: '------------------------------------------------------------------------------Public Function CreateLogin() '------------------------------------------------------------------------------'Function: CreateLogin 'Creates an instance of the Login class ' 'Remarks: ' 'Arguments: ' N/A ' 'Returns: ' Object . it is also required to define the following function that will return an instance of a GUI layer class (a kind of constructor function). we shall make use of the Scripting.Encapsulating Test Objects in Classes Let us take a testing automation project on a typical AUT and see how the solution should be designed according to the approach outlined above.. which are containers of other GUI objects.e. Now. dialogs. . as aforesaid. and popup messages. For each of these entities. we define a class. a container of other GUI objects.As Login ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------Dim objLogin Set objLogin = New Login Set CreateLogin = objLogin '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------- The second step is. Because QTP does not allow for direct instantiation of classes defined in external vbs files with the operator New. for example: Class Login End Class Class MainWindow End Class Class CreateCustomer End Class and so on for each of the application contexts. the windows (pages in a Web application). since each class is.

and I will define it as m_htChildObjects. I will define a public method within the class called Init. For this purpose. This property is defined as follows: '------------------------------------------------------------------------------'Property: ChildObjects 'Get and Set the m_htChildObjects member field ' 'Remarks: ' R/W ' 'Arguments: ' dic ' 'Returns: ' m_htChildObjects As HashTable ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------Public Property Get ChildObjects() '------------------------------------------------------------------------------Set ChildObjects = m_htChildObjects '------------------------------------------------------------------------------End Property '------------------------------------------------------------------------------'------------------------------------------------------------------------------Public Property Let ChildObjects(ByRef dic) '------------------------------------------------------------------------------Set m_htChildObjects = dic '------------------------------------------------------------------------------End Property '------------------------------------------------------------------------------- The third step is to define the objects contained within each context.Dictionary End Class Class MainWindow Private m_htChildObjects 'As Scripting. the first member I shall introduce here will be common to all GUI Layer classes. which is what the dictionary really is). So. for example: Class Login Private m_htChildObjects 'As Scripting.dialog or page.Dictionary End Class and so on for each of the application contexts (ht stands for HashTable. as follows: '------------------------------------------------------------------------------- . (the Dictionary object has been already extensively discussed in other articles published at AdvancedQTP„s knowledge base). The private member m_htChildObjects will be accessed through the class property ChildObjects.

Add "Submit". It is thanks to this encapsulation method that we ensure that GUI objects are always defined at a single place.Add "Page".Add "Browser". The test objects are added as entries to the ChildObjects Dictionary. and their identity is defined using Descriptive Programming (DP).Dictionary") With ChildObjects . ChildObjects("Page"). ChildObjects("Page"). The reader can easily infer the analogy to the OR.Add "Password". IsContextLoaded is defined in a separate common library. ChildObjects("Browser"). Browser("name:=My App") .Login") .WebButton("outertext:=Submit") End With 'IsContextLoaded is a function that iterates through the Dictionary and checks if the GUI objects "exist" Init = IsContextLoaded(ChildObjects) '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------- The code snippet above shows a typical Init method for a GUI Layer class for a Web application Login page. as follows: .Add "Username".WebEdit("html id:=Username") .Public Function Init() '------------------------------------------------------------------------------'Function: Init 'Initializes the context and child objects ' 'Dependencies: ' IsContextLoaded(htContext) ' 'Remarks: ' N/A ' 'Arguments: ' N/A ' 'Returns: ' True/False ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------ChildObjects = CreateObject("Scripting. ChildObjects("Page").Page("title:=My App \.WebEdit("html id:=Password") . At the end of the function body you will notice that it returns the result of a call to the function IsContextLoaded which accepts as argument the Dictionary that stores the ChildObjects.

keys. strDetails. strAdditionalRemarks '--------------------------------------------------------------------------items = htContext. SOLMAR Knowledge Networks Ltd." & strAdditionalRemarks Next '--------------------------------------------------------------------------Reporter. ' 'Remarks: ' N/A ' 'Arguments: ' ByRef htContext .ReportEvent intStatus.'------------------------------------------------------------------------------Public Function IsContextLoaded(ByRef htContext) '------------------------------------------------------------------------------'Function: IsContextLoaded 'Checks that the current GUI context is loaded ' 'Iterates through the htContext (HashTable) items and executes the Exist method with 0 (zero) as parameter.Exist(0) strDetails = strDetails & vbNewLine & "Object #" & ix+1 & ": '" & keys(ix) & "' was" If IsContextLoaded Then intStatus = micPass strDetails = strDetails & "" strAdditionalRemarks = "" Else intStatus = micWarning strDetails = strDetails & " not" strAdditionalRemarks = " Please check the object properties. ' 'Date: ' 11-Nov-2008 ' 'See Also: ' '------------------------------------------------------------------------------Dim ix. items.Keys For ix = 0 To htContext.Items keys = htContext." End If strDetails = strDetails & " found. strDetails '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------- And it returns True if all objects defined in the Dictionary are identified. This function is generic and it is used in the projects I .Count-1 IsContextLoaded = IsContextLoaded And items(ix).As HashTable ' 'Returns: ' True/False ' 'Owner: ' Meir Bar-Tal. or False if at least one object is not found. "IsContextLoaded".

manage to ensure that QTP does not get stuck while attempting to perform some operation on a non-existing GUI object. For example. Another benefit of this method is that it points exactly to the object we need to recheck and update. making maintenance much easier. Encapsulating Business Methods in Classes The next step after defining the child objects of the GUI context is to define the operations required to perform the application or business scenarios within the given context.Set GlobalDictionary("Username") '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------'------------------------------------------------------------------------------Public Function SetPassword() '------------------------------------------------------------------------------'Function: SetPassword 'Set the Password field ' 'Dependencies: ' N/A ' 'Remarks: . This is easily done by implementing class methods. the login class outlined above would need the following methods to begin with: SetUsername. SetPassword and Submit. These are shown below: '------------------------------------------------------------------------------Public Function SetUsername() '------------------------------------------------------------------------------'Function: SetUsername 'Set the Username field ' 'Dependencies: ' N/A ' 'Remarks: ' N/A ' 'Arguments: ' N/A ' 'Returns: ' N/A ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------ChildObjects("Username").

.' N/A ' 'Arguments: ' N/A ' 'Returns: ' N/A ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------ChildObjects("Password"). and the use of the ChildObjects property to retrieve through the test object a reference to the runtime object.Set GlobalDictionary("Password") '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------'------------------------------------------------------------------------------Public Function Submit() '------------------------------------------------------------------------------'Function: Submit 'Presses the Submit button ' 'Dependencies: ' N/A ' 'Remarks: ' N/A ' 'Arguments: ' N/A ' 'Returns: ' N/A ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------ChildObjects("Submit").Click 'TODO: Verify data submission performed successfully '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------- Notice the use of the GlobalDictionary to retrieve the values required by the functions.

As you can also see. the code within the function above is easy to understand.SetPassword() objLogin. which implements an application scenario building on the strong foundations of the GUI Layer.Submit() 'If login succeeds intStatus = micPass Else intStatus = micFail End If do_login = intStatus '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------- Notice the use the Login class outlined above and of its Init function as a precaution to ensure the GUI context is loaded. and not get stuck. as mentioned above. If changes are made to the GUI objects of a given context. objLogin Set objLogin = CreateLogin() If objLogin. both with respect to the object properties and .Init() Then objLogin.SetUsername() objLogin. we could wrap it with the following function: '------------------------------------------------------------------------------Public Function do_login() '------------------------------------------------------------------------------'Function: do_login 'Implements the business logic of the do_login Action. and is not cluttered with the references to both the OR test objects and data sources as quite often is the case.The next step would be to move on to the Business Layer. ' 'Remarks: ' 'Arguments: ' None ' 'Returns: ' Status ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------Dim intStatus. For instance to perform the Login function based on the above example. then the changes will be concentrated within a single package.

to the functionality required to manipulate the context‟s child objects. For example: 'VB Script Document Option Explicit '------------------------------------------------------------------------------Class do_login '------------------------------------------------------------------------------'Class: do_login 'Encapsulates the do_login Action. By implementing code this way we achieve a high degree of homogeneity in the code written by different developers. as outlined in my article Function Pointers in VB Script (revised). An advanced alternative to the last example is to pack such a Business Layer function using the Command Wrapper Design Pattern. ' 'Remarks: ' 'Arguments: ' None ' 'Returns: ' Status ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------Dim intStatus Set objLogin = CreateLogin() . and thus enhancing the manageability of the automation project. Yet another gain from this method is standardization. ' 'Remarks: ' 'Owner: ' John Doe ' 'Date: ' dd-MMM-yyyy ' '------------------------------------------------------------------------------'------------------------------------------------------------------------------'Methods '------------------------------------------------------------------------------'------------------------------------------------------------------------------Public Default Function Run() '------------------------------------------------------------------------------'Function: Run 'Implements the business logic of the do_login Action.

Init() Then objLogin. scalability. . Summary This article has reviewed an alternative approach to code implementation in a testing automation project that is based on object-oriented principles. and suggested an expansion to the OR concept – which was shown as inadequate viz-a-viz the OO paradigm – to encapsulate the interfaces to the GUI objects: the GUI Layer. The article has provided a practical example of how to implement such a layer and how to invoke it using a Business Layer. I tried to convey that logically following this line of thought leads to the conclusion that an automation project must be approached as a software development project par excellence.Submit() 'If login succeeds intStatus = micPass Else intStatus = micFail End If Run = intStatus '------------------------------------------------------------------------------End Function '------------------------------------------------------------------------------'------------------------------------------------------------------------------End Class '------------------------------------------------------------------------------- Adopting this method would enable the implementation of an advanced generic controller that loads its flow from an external data source such as an XML file. extensibility and testability. and explained in detail the benefits of adopting this approach to obtain the maximum ROI from an automation project regarding maintainability.SetUsername() objLogin. Moreover. Such a controller device was developed by me together with my partners at SOLMAR Knowledge Networks as part of our generic Object Oriented comprehensive automation framework – My System.SetPassword() objLogin.If objLogin. I have shown that automation scripting should not be treated differently than software development. Future articles will expand on this theme and guide the readers into how to gain from the implementation of other design patterns within the framework laid out in this article. readability.

By using a window test object with a text property of “Notepad”. it also contains methods to operate it: Close. on top of these. but when it comes down to the actual details. WebTable. Maximize. as well as to perform standard operations on it and with it. Browser – these are all QTP constructed test objects. but are only derived from it. while the actual . VBCheckbox. This article will guide you through the general concepts. The general idea is usually clear enough. and explain all the nitpicking involved in all the commands and practices relevant to the matter. It is important to understand that the test object‟s properties and methods are not really “out there” in the application‟s runtime object.e. For example. a Window test object contains properties which allow QTP to identify it in the “real world”: text. Click. The Test Object contains QTP specific properties and methods which allow us both to identify the relevant runtime object during playback. we tell QTP to look for a runtime window object with text = “Notepad”. Test Objects roles Test objects play several roles in QTP. sometimes even the experts do not quite know if using the GetROProperty method on a TO will work. height. QTP builds the “all items” property out of the combobox runtime object‟s available items collection. Whenever we record a real-world object (also known as a Runtime object). etc. Test Objects refer to your everyday QTP entities. it is derived from it. so even though the property does not exist in the runtime object. a SwfComboBox Test Object has the property “all items”. QTP creates a corresponding Test Object and stores it in the Object application combobox runtime object has no such property. . We can easily see these methods and properties via the object spy‟s Test Object tab. So for example. nativeclass.. the most immediate one is object identification.Differences and Connections between Runtime Objects and Test Objects There is a lot of confusion about how the concepts of Runtime Objects andTest Objects are used in QTP. in our application under test). Window.. The QTP Test Object Shell Under their fancy name. which represent real-world application objects “out there” (i.

the corresponding SwfComboBox test object provides a much more simpleand standard “Select” method which does the trick for you. there are 4 special commands which allow us to access the core of QTP‟s test object mechanism: GetTOProperty. or if that value was updated in “the real world” since the object was recorded. Like GetTOProperty. This means that unlike the GetTOProperty and GetTOProperties commands. selecting a value in a . For example. we do it through the QTP test objects. and perform the relevant action on it. GetTOProperty allows us to retrieve the value of a test object property. only the test object which represents it. Working with Test Objects How can we work with test objects in our scripts? The most straight-forward way is to simply use them.).Net combobox usually involves changing several properties (SelectedIndex. events and properties. this command will print out “Notepad”: Msgbox Window("text:=Notepad"). QTP will touch the . select a value from a VBComboBox or click a WebElement. GetROProperty allows us to get the current value of a test object property. It does not matter if the corresponding runtime object exists. TextChanged. etc. Every time we maximize a window. or create instant test objects via Descriptive Programming. SetTOProperty and GetROProperty. and raising several events (SelectionChanged. the corresponding QTP test objects are standard and very simple in comparison. the GetTOProperties command has nothing to do with the actual runtime object in our application.GetTOProperty("text") The GetTOProperty command will retrieve the value as it was originally recorded (or created via DP). Aside from straightforwardly using our test objects to perform actions. For example. We can either use prerecorded test objects from the object repository. SelectedItem). However. GetROProperty requires the test object‟s corresponding runtime object to exist.Another important role is standardization. ItemSelected. QTP will then locate the corresponding “real” runtime object. QTP will translate the test object‟s simple commands to the runtime object‟s commands required to perform the task at hand. GetTOProperties returns a collection with all the test object properties. While the application runtime objects might be extremely complex. GetTOProperties. The test object‟s simple properties and methods act as a bridge. sparing us the need to nitpick our way through the runtime object‟s unknown commands.

So for example: Call Window("Notepad"). Even though it has RO (Runtime Object) in it.SwfCombobox("Y"). . From Test Objects to Runtime Objects Test objects are the cornerstone of all QTP object operations. Here it makes no difference if QTP succeeds to correctly recognize the object. the test objects it creates do not have the properties or methods that are required to operate the objects (like “all items” and Select).GetROProperty("all items") Last but not Least.runtime object. if the text of the Notepad window changed from when that object was recorded. So even though a . combo-boxes). Any changes made with SetTOProperty will be reset once the test-run had finished. as opposed to the QTP test objects: There are some objects (especially . our test object has only one property – nativeclass. It is important to note that changing a test object‟s properties might cause QTP not to identify its corresponding runtime object. Msgbox Window("nativeclass:=Notepad").Net custom controls). "incorrect value") Window("Notepad"). some combo-boxes might appear as SwfObjects. Another case is when we are required to perform some non-standard operation with an object. In the following example. As QTP fails to recognize the object for what they are (in this example.GetROProperty("text") Unlike GetTOProperty and GetTOProperties. For example. and retrieve the value of the property from it.SetTOProperty("nativeclass". but GetROProperty will still be able to retrieve the value of the window‟s text property (where GetTOProperty would return an empty result).Close 'this command will fail The second command will fail as QTP will search for an object with nativeclass = “incorrect value”. which QTP fails to recognize correctly.GetROProperty("text") It is important not to be confused by the “RO” in the command name. For example. it does not even matter if the test object originally recorded the relevant property. and it will not find any object matching that description. this command will retrieve the updated value: Msgbox Window("Notepad"). you could still execute the following command. the command still retrieves only test object properties. There are two situations in which we might prefer working with the actual runtime objects of the application. or generic ActiveX controls. SetTOProperty allows us to change a test object‟s property for the duration of the test-run.Net combobox runtime object does not have an “all items” property. or even when you go in and out of actions. and it will print out all the available items: Msgbox SwfWindow("X"). but they are sometimes too limited for the task at hand.

Object. Working with Runtime Objects You can see an object‟s runtime properties and methods in the RO tab of the object spy. we can perform almost any operation on our objects. by overwriting the runtime property of the selected item‟s index. it can be done via their . WebList ("z").Page("Y"). So in this case.Page("Y").CurrentStyle. For example: Print Browser("X"). Not all test objects support working with their corresponding runtime objects.Object.Page("Y"). WebList ("z"). we‟ll still be able to select an item from it. WebList ("z") combobox is identified as a SwfObject. The WebElement test object simply does not carry the information we need.Object. If it is a .as the corresponding test object will only provide us with a standard set of commands and properties. WebList ("z"). A good example for this case is trying to find the font style of a WebElement. only a derived property called “selected item index”.GetROProperty("selected item index") The true benefit of working with runtime objects lies with properties and methods that are not available through the QTP test object. and perform operations that are otherwise unavailable to us: Print Browser("X").SwfObject("Y"). access its selectedIndex property. and print its value. and so we have to resort to unusual measures such as working with the WebElement‟s runtime object. we can find out the color of a WebList‟s items text: Print Browser("X").Object. regardless of QTP‟s ability to recognize them for what they truly are.selectedIndex Print Browser("X"). these commands will print the same value: Print Browser("X").Object property. So for example.SelectedIndex = 1 . WebList ("z"). Of course that in order for this to work the runtime object must exist when we execute the command.selectedIndex This command will find the runtime object corresponding to the WebList‟s test object. but if they do. This property exposes all the properties and methods of the runtime object “hiding” behind our test object.Color We can even use the runtime object‟s methods.Page("Y"). It is important to note that WebList test object does not have a selectedIndex property.focus By accessing the runtime world. you can also use the . SwfWindow("X").Page("Y").Object.Net form spy to get a more detailed look of those properties that contain inner complex objects. So even if our .

All programming languages suffer from such problems. whereas in C++ it would trigger a compilation error. . we must always be aware of them. and exactly what properties and methods it has (CurrentStyle. In any language. calling the wrapping function. For example.Color was hard to come by). What usually follows is a series of trial-and-error breakpoint runs to determine which action the script was actually trying to perform when executing the function call that triggered the error. This is because in VB Script compilation is done on the basis of a line-after line interpretation during run-time. but nonetheless. in which compilation at least guarantees that the function calls are syntactically correct. without one having the slightest clue in its regard. a faulty call to a function (wrong function name or number of parameters) would pass unnoticed in VB Script when part of a function. it is important to remember that working directly with runtime objects can have significant downsides.With all their benefits. Even when we find the right properties or methods. that is. such as exploring custom controls. There are ways to overcome these shortcomings. Adding a Call-Chain Mechanism to QTP Background and motivation A common frustrating experience is having your script pop-up an error dialog in some long-forgotten function. but in late bound languages like VB Script this problem is really challenging. or by testing the inner function call on a standalone basis. in contrast to languages such as C/C++. it is not long until the calls to functions and object methods become too deeply nested for us humans to grasp. The only way to get the error message in such cases is by actually performing the code. it is unclear how to operate the object. Unlike the standard and simple test objects. using them “from behind the scenes” may cause application errors and invoke hidden bugs. and carefully mapping the consequences of changing runtime properties.

programmers do debug their code. They manage to do so via a dedicated mechanism: the call-chain (sometimes referred to as the call-stack). Or. Developed by our people at SOLMAR‟s R&D Division. My SystemTM brings the latest advances to boost ROI on testing automation (We shall have much more to say on that in future posts. Meanwhile. This allows us to analyze the error‟s path. how valuable would it been to know which parameters were sent into those function calls. Is the inevitable conclusion that there is no way to implement a call-chain in QTP effectively? Well. . Moreover. QTP reports are inaccessible during the run-session.Having said that. Figure 1: A Sample Visual Studio Error Call-Stack Just imagine how many debug runs you could have saved just by knowing the exact order of calls which lead you to the current function. Possible Implementations Now. However. An example of a Visual Studio call-stack for an error can be viewed in Figure 1 below. stay tuned!). There is a way. even when bugs and error occur within extremely nested function calls. though it‟s not trivial. which function called that function. The call-chain shows us exactly which function called the currently executing function. this will render your report cluttered and hence difficult to grasp. Please notice that it shows both the functions and code lines that precipitated the error. the answer is no. and see where things started to go wrong (usually 2-3 function calls before the function we are currently in). it is possible to implement a call-chain in QTP with intensive reporting – just by sending a line to the report from any function as you enter and exit it. and so on up to the root of the call hierarchy. indeed. and that is exactly why the call-chain data is crucial. I thought that I could share some of my techniques and thoughts on the subject. A call-chain mechanism is just one of the rich features provided by SOLMAR‟s flagship product: My SystemTM.

Storing the Call-Chain Data An excellent mechanism for storing call-chain data is a stack. This will make sure your call chain contains all the relevant data needed to understand what got you until this point. A mechanism for viewing the stored data mid-run (we won‟t get into this one – it is basically just a simple window to peek at the stored data) If you have a robust mechanism for updating the call-chain when you are exiting and entering contexts. 2. The Concept Developing an effective call-chain requires four basic components: 1. 4.. sub or action). Here is an example of how the stack mechanism would work: Let‟s say our code went though these functions: Function Func1 Call Func2 End Function Function Func2 Call Func3 End Function Function Func3 Dim x Call Func 4 x = 1/0 End Function . I do believe they can point you in the right direction for developing robust solutions of your own. 3. A place to store the call-chain data. the names of the contexts you happen to enter during the flow) in a FILO manner – First-In Last-Out. and nothing else. and an easy to use storage for the call chain.Though the examples given here are not meant to be used or to be an out-of-the-box solution. A mechanism for updating the call-chain when you exit a context. A mechanism for updating the call-chain when you enter a context (by context I mean a function.e. then implementing the call-chain will be quite straightforward. A stack can store data (i.

. “Func3”. once the code will exist Func4. and it would call Func4. returning the control to Func3.Function Func4 'Ended End Function If we call Func1. “Func4”). the fact the we have entered and exited Func4). “Func2”. Line 14 will result is an error.] 'Remove from stack oChainData. where we could debug. as the stack mechanism always know to remove the top level item. Since our call chain is a Stack. Now Func4 will end. we better start recording it. after we got our own storing mechanism for the call-chain data.Add "Func1" '[. ' Function body ' . it will remove the last item from our stack. Problems with the Naive Approach That is all fine and well.. “Func2”. our call-chain stack would look like this: (“Func1”. which would call Func3. you can take a look at Dani Vainstein‟s excellent article on the matter – here. but usually our code is more complex than this – usually our functions have multiple exit gates. and a removecommand every time we exit it. A Naive Approach to Updating the Call-Chain Data on the Run Well. make it: (“Func1”. it will call Func2. The naive approach would be to simply perform an add command once we enter a function. If you‟d like to know exactly how to implement a stack. which would hold just enough data to understand how we got into Func3.Remove End Function Notice that we don‟t have to specify the function name for our Remove command.. with add and remove commands): Function Func1 'Add to call chain oChainData. So it should look something like this (assuming our data is stored in an oChainData stack storage. and take a look at our stack. as shown in the following example: Function Func1 'Add to call chain oChainData. At this point. “Func3”)..Add "Func1" . and no irrelevant data (for example.

. 'If structures etc. but we can do the exit part quite easily. oChainData..'Code.. we could use the termination event of an object. oChainData.Remove Exit Function 'More Code. 'If structures etc. before letting us update the call-chain data.. anything we write within the Class_Terminate sub of the class will be executed when the class object is destroyed. So.. In order to achieve that. Implementing the whole package (monitor both entering and exiting a function) might be hard. And even if our code is perfect (and it never is)..Remove Exit Function 'Now the regular exit gate: oChainData. we still run the risk of a mid-function error that will throw us out. But such errors are exactly our motivation to be willing to put an effort in maintaining a call-chain mechanism in the first place! A More Robust Solution We have seen that our naive solution fails at some basic everyday cases. 'If structures etc. and that would just make our whole stack misleading and useless. What these examples show is that we need an automatic component that will “listen” to our function..... but perhaps we can overcome this with more complex scripting. oChainData.Remove Exit Function 'More Code. and tell us (or rather the call-chain) when we enter and exit it. it is just a matter of time before we forget to update the call-chain. When we write a class..Remove End Function When our function has multiple exit gates.. for example: Class KillSwitch Sub Class_Terminate Msgbox "Object terminated" End Sub End Class Set oKill = New KillSwitch Set oKill = Nothing .

to share with the community your own call-chain mechanism! .Remove End Sub End Class Function Func1 Dim oUpdater 'This will hold the exit kill-switch oChainData.'Now the message box will appear As you can see. I hope that it has successfully outlined how to build the main components that are required in order to implement such a mechanism. We can use this to our advantage – instead of popping up a message box. SOLMAR Knowledge Networks has developed an automation framework with a built-in call-chain mechanism. As a closing word. about this article. it builds upon the same basic principles demonstrated in this article. or due to an error). We examined an alternative approach by which call-chain update is achieved automatically through an object‟s kill-switch mechanism. It could look something like this: 'The blueprint for the updater Class ChainUpdater Sub Class_Terminate oChainData. the object “knows” when it is destroyed. let us make our object update the call-chain. The main concept explained how we can keep track of the chain-data. no matter when we exit the function. We have seen that the naive approach (manually update on entering and exiting a context) is impractical. Although the article did not provide a detailed example of a call-chain. Since the object will be destroyed whenever we exit the function (either purposefully.Add "Func1" 'Update the chain-data Set oUpdater = New ChainUpdater 'Create an instance of the kill switch 'Now. we don‟t have to worry about “forgetting” to update the call-chain. and if you happen to dare and design. the kill switch will be activated 'And the chain-data will be updated End Function Summary This article introduced the concept of a call-chain and provided substantial reasons for its centrality in the automation development effort. let me encourage you to share your thoughts and comments about the call-chain concept. While it is much more detailed and effective. As aforementioned.

Sign up to vote on this title
UsefulNot useful