QTP Actions and Action Properties
Actions divide a test into logical units. This aids in highly readable tests and ensures that modularity is achieved.
There are 3 types of QTP Actions:
- Reusable – Only these can be called multiple times within the same or different tests.
- Non-Reusable – These cannot be called by any other actions.
- External Actions – A reusable action when called from another action becomes an external action in the called action.
What You Will Learn:
How to Use QTP Actions in Tests?
#1) By default all new actions are reusable.
#2) Every action has its own “Action” datasheet or a local sheet.
#3) Call to an existing action in another test can be inserted by using the menu option: “Insert -> Call to an existing action -> (Choose test/action)”
#4) The statement or action call inserted in the current test will be as follows:
RunAction “Action2[Testname]”, oneiteration
#5) The action will be called as many times as the action called properties is specified for the parent action.
Action2 is a reusable action in Test2. It is called on every iteration in the data table of Test2.
Action1 is the action that calls Action2 in Test1.
Action1 runs only once.
Then, Action2 will run only once irrespective of how many times it was called in Test2.
#6) A previously reusable action can be changed to become non-usable and vice versa.
#7) When a previously reusable action is marked as non-reusable, then the test that calls the action fails and a warning message is displayed to the user.
#8) In the previous article, we have seen how parameters can be passed “To” and “From” actions. To define this, you can do one of the following:
- Choose the action from the keyword view, right-click and choose Action Properties.
- While in the action in the expert view, select Edit ->Action Properties from the menu.
#9) The dialog box appears with General (name, description, reusable or not), Parameters(i/p or o/p), associated repositories (can associate another action’s OR), and Used by tabs.
#10) To send parameters to action, the general syntax is:
RunAction “Action Name”. Oneiteration, i/p parameters, o/p parameters
Given below are some of the ways in which you can utilize the output value of an action in your test depending on your requirement:
a) Action o/p value (value returned by a called action) can be stored in a variable.
RunAction “Action1”,OneIteration, i/pvalues, x, x being the variable
b) Action o/p value can be stored in an environment variable.
To create an environment variable choose “File -> Settings -> Environment tab”, in the ‘Variable Type’ drop-down choose user-defined and click + and ‘Add new environment Parameter’ – enter the name (env_Var) and an empty value and click OK.
To assign the o/p value to this variable,
RunAction “Action1”,OneIteration, i/pvalues, Environment(“env_var”)
c) Action o/p values can be stored in the data table column.
RunAction “Action1”, OneIteration, i/pvalue, DataTable(“A”, dtGlobalsheet)
QTP Functions and Function Libraries
Though there are many built-in functions available in the QTP for the tester, there might still be instances when you might want to perform a specific operation and make those steps reusable.
You can do that by creating user-defined functions, putting them in a separate file, associating it with the test, and having the function name work as a keyword for that particular test.
So by definition, a function is nothing but a piece of code that performs a certain task.
Let’s create a function library now:
Use the menu option “File -> New -> Function Library” or “Shift+Alt+N” or click on the “New” drop-down option on the menu and choose “Function library”.
QTP will open up an editor, very similar to the keyword view of a test. You can write your code here. I am going to include statements that will log a user into the gmail.com page.
Function gmailLogin(uname, pwd) SystemUtil.Run "iexplore.exe", "http://www.gmail.com" Browser("Gmail: Email from Google").page("Gmail: Email from Google").Sync Browser("Gmail: Email from Google").Page("Gmail: Email from Google").WebEdit("Email").Set uname Browser("Gmail: Email from Google").Page("Gmail: Email from Google").WebEdit("Passwd").SetSecure pwd Browser("Gmail: Email from Google").Page("Gmail: Email from Google").WebButton("Sign in").Click End Function
Once you are done with writing the code, “Save” the function library by choosing the required name and location on your computer.
Note: The function library does not get auto-included in any of your tests. You are going to have to do it explicitly.
Given below are the steps to associate a Function Library into your test:
Step #1: Open your test or create a new test.
Step #2: Depending on the settings you have on your IDE, you might have the resources pane already available to you. If not, choose the menu option “View -> Resources” and this displays the pane.
Step #3: Right-click on the “Associate Function libraries”, click the “Associate Function library” option, and choose the library file you created earlier.
Step #4: There will be a message that asks for the tester’s confirmation before converting the path to a relative path. This is done to make sure that the test does not fail if the files are moved as long as the same hierarchy is followed.
Click yes on this dialog box:
Step #5: It will appear under this node from now on and that is how you verify if your association has worked or not.
Step #6: Once you have associated the library, make sure you save your test.
From now on if there is a statement in my test that is as follows, then it will invoke the function in the library and execute the steps in there.
In the statement above, you are passing constant username and password values. But if you choose to, you can pass on the values from the data table too.
gmailLogin(DataTable("SignInName", dtGlobalSheet), DataTable("GPassword", dtGlobalSheet))
You can have more than one function in your library files and you can have more than one library file associated with your test.
This concludes our Function Libraries and Actions. These two topics are crucial for a tester as they make all your tests more modular and reusable.
The next topic is going to be on Object repositories (OR). We will discuss how to create, associate, and work with shared ORs and OR managers.
Please feel free to post your questions and comments.