Squish for macOS Tutorials
Learn how to test native macOS applications.
- Tutorial: Starting to Test macOS Applications
- Tutorial: Designing Behavior Driven Development (BDD) Tests
- Tutorial: Migration of existing tests to BDD
Tutorial: Starting to Test macOS Applications
Squish comes with an IDE and command line tools. Using the squishide
is the easiest and best way to start, but once you build up lots of tests, you will want to automate them. For example, to do nightly runs of your regression test suite. Therefore, it is worth knowing how to use the command line tools that can be run from batch files or shell scripts.
Note: There is a 45-minute Online course about Squish Basic Usage at the if you desire some video guidance.
For this chapter, we will use a simple address book application as our AUT. The application is shipped with Squish in SQUISHDIR/examples/mac/addressbook
. This is a very basic Cocoa application that allows users to load an existing address book or create a new one, add, edit, and remove entries, and save (or save as), the new or modified addressbook. Despite the application's simplicity, it has all the key features that most standard applications have: a menu bar with pull down menus, a toolbar, and a central area—in this case showing a table. It supports in-place editing and also has a sheet for adding items. All the ideas and practices that you learn to test this application can easily be adapted to your own applications. For more examples of testing various Cocoa-specific features, see How to Create Test Scripts.
Note: Throughout the manual, we often refer to the SQUISHDIR
directory. This means the directory where Squish is installed, which might be C:\Squish
, /usr/local/squish
, /opt/local/squish
, or something else, depending on where you installed it. The exact location doesn't matter, so long as you mentally translate the SQUISHDIR
directory to whatever the directory really is when you see paths and filenames in this manual.
The screenshot shows the application with a newly created, empty address book.
Using the Examples
The first time you try running a test for one of the example AUTs, you might get a fatal error that begins Squish couldn't find the AUT to start.... To recover from the error, click the Test Suite Settings toolbar button, and in the Application Under Test (AUT) section, choose the AUT from the combobox if it is available, or click the Browse button and navigate to the AUT's executable in the finder dialog. Some versions of Squish automatically open this dialog if no AUT is specified. You need to do this only once per example AUT, and not at all when testing your own AUTs.
Squish Concepts
In the following sections we will create a test suite and then create some tests, but first we will very briefly review some key Squish concepts.
To perform testing, you need:
- An application to test, known as the Application Under Test (AUT).
- A test script that exercises the AUT.
One fundamental aspect of Squish's approach is that the AUT and the test script that exercises it are always executed in two separate processes. This ensures that even if the AUT crashes, it should not crash Squish. In such cases, the test script will fail gracefully and log an error message. In addition to insulating Squish and test scripts from AUT crashes, running the AUT and the test script in separate processes brings other benefits. For example, it makes it easier to store the test scripts in a central location and to perform remote testing on different machines and platforms. The ability to do remote testing is particularly useful for testing AUTs that run on multiple platforms and for testing AUTs that run on embedded devices.
Squish runs a small server, squishserver, that handles the communication between the AUT and the test script. The test script is executed by the squishrunner tool, which in turn connects to squishserver. squishserver starts the instrumented AUT on the device, which starts the Squish hook. The hook is a small library that makes the AUT's live running objects accessible, and allows communication with squishserver. With the hook in place, squishserver can query AUT objects regarding their state and can execute commands on behalf of squishrunner. squishrunner directs the AUT to perform whatever actions the test script specifies.
All the communication takes place using network sockets which means that everything can be done on a single machine, or the test script can be executed on one machine and the AUT can be tested over the network on another machine.
The following diagram illustrates how the individual Squish tools work together.
From the test engineer's perspective, this separation is not noticeable because all the communication is handled transparently behind the scenes.
Tests can be written and executed using the squishide
, in which case squishserver is started and stopped automatically, and the test results are displayed in the squishide
's Test Results view. The following diagram illustrates what happens behind the scenes when the squishide
is used.
The Squish tools can also be used from the command line without the squishide
. This is useful if you prefer to use you own tools, such as your favorite editor or want to perform automatic batch testing. For example, when running regression tests overnight. In these cases, squishserver must be started manually and stopped when all the testing is complete or started and stopped for each test.
Note: The Squish documentation mostly uses the term widget when referring to GUI objects, such as buttons, menus, menu items, labels, and table controls. Windows users might be more familiar with the terms control and container, but here we use the term widget for both. Similarly, macOS users may be used to the term view.
Creating a Test Suite
A test suite is a collection of one or more test cases (tests). Using a test suite is convenient since it makes it easy to share scripts and test data between a group of related tests.
Here, and throughout the tutorial, we will start by describing how to do things using the squishide
, with the information for command line users following.
Creating Test Suites from squishide
Start up the squishide
, by clicking or double-clicking the squishide icon, by launching squishide from the taskbar menu or by executing squishide on the command line, whichever you prefer and find suitable for the platform you are using. Once Squish starts up, you might be greeted with a Welcome Page. Click the Workbench button in the upper right to dismiss it. Then, the squishide will look similar to the screenshot, but probably slightly different depending on the windowing system, colors, fonts, and theme that you use.
Once Squish has started, click File > New Test Suite to pop-up the New Test Suite wizard shown below.
Enter a name for your test suite and choose the folder where you want the test suite to be stored. In the screenshot we have called the test suite suite_py
and will put it inside the addressbook
folder. (For your own tests you might use a more meaningful name such as "suite_addressbook"; we chose "suite_py" because for the sake of the tutorial we will create several suites, one for each scripting language that Squish supports.) Naturally, you can choose whatever name and folder you prefer. Once the details are complete, click Next to go on to the Toolkit (or Scripting Language) page.
If you get this wizard page, click the toolkit your AUT uses. For this example, we must click Mac since we are testing a macOS application. Then click Next to go to the Scripting Language page.
Choose whichever scripting language you want—the only constraint is that you can only use one scripting language per test suite. (So if you want to use multiple scripting languages, just create multiple test suites, one for each scripting language you want to use.) The functionality offered by Squish is the same for all languages. Having chosen a scripting language, click Next once more to get to the wizard's last page.
If you are creating a new test suite for an AUT that Squish already knows about, simply click the combobox to drop-down the list of AUTs and choose the one you want. If the combobox is empty or your AUT isn't listed, click the Browse button to the right of the combobox—this will pop-up a file open dialog from which you can choose your AUT. In the case of Cocoa programs, the AUT is the application's executable (e.g., SquishAddressBook
on macOS). Once you have chosen the AUT, click Finish and Squish will create a sub-folder with the same name as the test suite, and will create a file inside that folder called suite.conf
that contains the test suite's configuration details. Squish will also register the AUT with the squishserver. The wizard will then close and squishide
will look similar to the screenshot below.
We are now ready to start creating tests. Read on to learn how to create test suites without using the squishide
, or skip ahead to Recording Tests and Verification Points.
Creating Test Suites from Command Line
To create a new test suite from the command line:
- Create a new directory to hold the test suite—the directory's name should begin with
suite_
. In this example we have created theSQUISHDIR/examples/mac/addressbook/suite_py
directory for Python tests. (We also have similar subdirectories for other languages but this is purely for the sake of example, since normally we only use one language for all our tests.) - Register the AUT with the squishserver.
Note: Each AUT must be registered with the squishserver so that test scripts do not need to include the AUT's path, thus making the tests platform-independent. Another benefit of registering is that AUTs can be tested without the
squishide
— for example, when doing regression testing.This is done by executing the squishserver on the command line with the
--config
option and theaddAUT
command. For example, assuming we are in theSQUISHDIR
directory on macOS:squishserver --config addAUT SquishAddressBook \ SQUISHDIR/examples/mac/addressbook
We must give the
addAUT
command the name of the AUT's executable and—separately—the AUT's path. In this case the path is to the executable that was added as the AUT in the test suite configuration file. For more information about application paths, see AUTs and Settings, and about the squishserver's command line options see squishserver. - Create a plain text file (ASCII or UTF-8 encoding) called
suite.conf
in the suite subdirectory. This is the test suite's configuration file, and at the minimum it must identify the AUT, the scripting language used for the tests, and the wrappers (i.e., the GUI toolkit or library) that the AUT uses. The format of the file iskey
=
value
, with one key–value pair per line. For example:AUT = SquishAddressBook LANGUAGE = Python WRAPPERS = Mac OBJECTMAPSTYLE = script
The AUT is the Cocoa executable registered during the previous step. The LANGUAGE can be set to whichever one you prefer—currently Squish is capable of supporting JavaScript, Python, Perl, Ruby, and Tcl. The WRAPPERS should be set to Mac.
Recording Tests and Verification Points
Squish records tests using the scripting language that was specified for the test suite. Once a test has been recorded, we can run the test and Squish will faithfully repeat all the actions that we performed when recording the test, but without the pauses that humans are prone to but which computers don't need. It is also possible and very common to edit recorded tests, or to copy parts of recorded tests into manually created tests, as we will see later on in the tutorial.
Recordings are made into existing test cases. You can create a New Script Test Case in the following ways:
- Select File > New Test Case to open the New Squish Test Case wizard, enter the name for the test case, and select Finish.
- Click the New Script Test Case ( ) toolbar button to the right of the Test Cases label in the Test Suites view. This creates a new test case with a default name, which you can easily change.
Give the new test case the name "tst_general".
Squish automatically creates a sub-folder inside the test suite's folder with this name and also a test file, for example test.py
. If you choose JavaScript as the scripting language, the file is called test.js
, and correspondingly for Perl, Ruby, or Tcl.
If you get a sample .feature
file instead of a "Hello World" script, click the arrow left of the Run Test Suite ( ) and select New Script Test Case ( ).
To make the test script file (such as, test.js
or test.py
) appear in an Editor view, click or double-click the test case, depending on the Preferences > General > Open mode setting. This selects the Script as the active one and makes visible its corresponding Record ( ) and Run Test ( ) buttons.
The checkboxes are used to control which test cases are run when the Run Test Suite ( ) toolbar button is clicked. We can also run a single test case by clicking its Run Test ( ) button. If the test case is not currently active, the button may be invisible until the mouse is hovered over it.
Initially, the script's main()
logs Hello World to the test results. To create a test manually, as we will do later on in the tutorial, we must create a main
function, and we should import the same imports at the top. The name main
is special to Squish. Tests may contain as many functions and other code as you like, as supported by the scripting language, but when the test is executed (that is, run), Squish always executes the main
function. You can share commonly used code between test scripts, as described in How to Create and Use Shared Data and Shared Scripts.
Two other function names are also special to Squish: cleanup
and init
. For more information, see Tester-Created Special Functions.
Once the new test case has been created, we are free to write test code manually or to record a test. Clicking on the test case's Record ( ) button replaces the test's code with a new recording. Alternatively, you can record snippets and insert them into existing test cases, as instructed in How to Edit and Debug Test Scripts.
Creating Tests from Command Line
To create a new test case from the command line:
- Create a new subdirectory inside the test suite directory. For example, inside the
SQUISHDIR/examples/mac/addressbook/suite_py
directory, create thetst_general
directory. - Inside the test case's directory create an file called
test.py
(ortest.js
if you are using the JavaScript scripting language, and similarly for the other languages).
Recording Our First Test
Before we dive into recording let's briefly review our very simple test scenario:
- Open the
MyAddresses.adr
address file. - Navigate to the second address and then add a new name and address.
- Navigate to the fourth address (that was the third address) and change the surname field.
- Navigate to the first address and remove it.
- Verify that the first address is now the new one that was added.
We are now ready to record our first test. Click the Record ( ) to the right of the tst_general
test case shown in the Test Suites view's Test Cases list. This will cause Squish to run the AUT so that we can interact with it. Once the AUT is running perform the following actions—and don't worry about how long it takes since Squish doesn't record idle time:
- Click File > Open, and once the file dialog appears, press Shift+ +Command+g and enter the
MyAddresses.adr
filename in the line edit that appears, then click the Go button, then click the Open button. - Click the first row, then click the Add toolbar button, then in the Add sheet's first line edit type in "Jane". Now click (or tab to) the second line edit and type in "Doe". Continue similarly, to set an email address of "jane.doe@nowhere.com" and a phone number of "555 123 4567". Don't worry about typing mistakes—just backspace delete as normal and fix them. Finally, click the Add button. There should now be a new second address with the details you typed in.
- Double-click the fourth row's second (surname) column, delete its text and replace it with "Doe". (You can do this simply by overtyping, then pressing Enter.)
- Click the first row, then click the Remove toolbar button and then click the Remove button in the message box. The first row should be gone, so your "Jane Doe" entry should now be the first one.
- In the Squish Control Bar, click the Verify toolbar button and select Properties from the drop-down menu.
This will make the
squishide
appear, and the central widget should show you a Verification Point Creator view. Before selecting any properties, make sure the Type of the VP is Scriptified Property, as shown in the following screenshots.In the Application Objects view, expand the
Address Book - MyAddresses.adr_NSWindow_0
object, then theNSView_0
object, then theNSScrollView_0
object, then theNSClipView_0
object, and then theNSTableView_0
object. Click theJane_NSCFString_0
object to make its properties appear in the Properties view, and then check thestringValue
property's checkbox. Now scroll down and click theDoe_NSCFString_127
object and check itsstringValue
property. Finally, click the Save and Insert Verifications button (at the bottom of the Verification Point Creator) to have the forename and surname verifications for the first row inserted into the recorded test script. (See the screenshot below.) Once the verification points are inserted, thesquishide
's window will be hidden again and the Control Bar window and the AUT will be back in view. - We've now completed the test, so in the AUT click SquishAddressBook > Quit, then click Don't Save in the message box, since we don't want to save any changes.
Once we quit the AUT, the recorded test will appear in squishide
as the screenshot illustrates. (Note that the exact code that is recorded will vary depending on how you interact. For example, you might invoke menu options by clicking them or by using key sequences—it doesn't matter which you use, but since they are different, Squish will record them differently.)
If the recorded test doesn't appear, click (or double-click depending on your platform and settings) the tst_general
test case; this will make Squish show the test's test.py
file in an editor window as shown in the screenshot.
Now that we've recorded the test we are able to play it back, i.e., run it. This in itself is useful in that if the play back failed it might mean that the application has been broken. Furthermore, the two verifications we put in will be checked on play back as the screenshot shows.
Inserting verification points during test recording is very convenient. Here we inserted two in one go, but we can insert as many as we like as often as we like during the test recording process. However, sometimes we might forget to insert a verification, or later on we might want to insert a new verification. We can easily insert additional verifications into a recorded test script as we will see in the next section, Inserting Additional Verification Points.
Before going further we will look at how to record a test from the command line. Then we will see how to run a test, and we will also look at some of the code that Squish generated to record the test and discuss some of its features.
Recording Tests from Command Line
The squishserver must always be running when recording a test. This is handled automatically by the squishide
, but command line users must start the squishserver manually, as instructed in squishserver.
To record a test from the command line, execute the squishrunner program and specify the test suite to record inside and the name of the test case. For example, assuming we are in the directory that contains the test suite's directory:
squishrunner --testsuite suite_py --record tst_general --useWaitFor
It is always best to record using the --useWaitFor
option that records calls to the Object waitForObject(objectOrName) function, which is more reliable than using the snooze(seconds) function, even though it is the default for historical reasons. The squishide
automatically uses the Object waitForObject(objectOrName) function.
When you have multiple devices and/or emulators attached, then you need to specify the target using --device some-device
Running Tests from IDE
To run a test case in the squishide
, click the Run Test ( ) that appears when the test case is hovered or selected in the Test Suites view.
To run two or more test cases one after another or to run only the selected test cases, clic Run Test Suite ( ).
Running Tests from Command Line
The squishserver must always be running when running a test, or the --local
option must be provided to squishrunner. For more information, see squishserver.
To play back a recorded test from the command line, we execute the squishrunner program and specify the test suite our recorded script is in and the test case we want to play. For example, assuming we are in the directory that contains the test suite's directory:
squishrunner --testsuite suite_py --testcase tst_general --local
Examining the Generated Code
If you look at the code in the screenshot (or the code snippet shown below) you will see that it consists of lots of Object waitForObject(objectOrName) calls as parameters to various other calls such as activateItem(objectOrName), mouseClick(objectOrName), and type(objectOrName, text). The Object waitForObject(objectOrName) function waits until a GUI object is ready to be interacted with (i.e., becomes visible and enabled), and is then followed by some function that interacts with the object. The typical interactions are activate (pop-up) a menu, click a menu option or a button, or type in some text.
For a complete overview of Squish's script commands see How to Create Test Scripts, How to Test Applications - Specifics, API Reference, and Tools Reference.
Objects are identified by names that Squish generates. See How to Identify and Access Objects for full details.
The generated code is about 35 lines of code. Here's an extract that just shows how Squish records clicking the Add toolbar button, typing in Jane Doe's details into the Add sheet, and clicking Add at the end to close the sheet and update the table.
Note: Although the screenshots only show the Python test suite in action, for the code snippets quoted here and throughout the tutorial, we show the code for all the scripting languages that Squish supports. In practice you would normally only use one of them of course, so feel free to just look at the snippets in the language you are interested in and skip the others.
mouseClick(waitForObject(names.address_Book_MyAddresses_adr_Add_NSToolbarItem)) type(waitForObject(names.add_Address_NSTextField), "Jane") type(waitForObject(names.add_Address_NSTextField), "<Tab>") type(waitForObject(names.add_Address_NSTextField_2), "Doe") type(waitForObject(names.add_Address_NSTextField_2), "<Tab>") type(waitForObject(names.add_Address_NSTextField_3), "jane.doe@nowhere.com") type(waitForObject(names.add_Address_NSTextField_3), "<Tab>") type(waitForObject(names.add_Address_NSTextField_4), "555 123 4567") mouseClick(waitForObject(names.add_Address_Add_NSButton))
mouseClick(waitForObject(names.addressBookMyAddressesAdrAddNSToolbarItem)); type(waitForObject(names.addAddressNSTextField), "Jane"); type(waitForObject(names.addAddressNSTextField), "<Tab>"); type(waitForObject(names.addAddressNSTextField_2), "Doe"); type(waitForObject(names.addAddressNSTextField_2), "<Tab>"); type(waitForObject(names.addAddressNSTextField_3), "jane.doe@nowhere.com"); type(waitForObject(names.addAddressNSTextField_3), "<Tab>"); type(waitForObject(names.addAddressNSTextField_4), "555 123 4567"); mouseClick(waitForObject(names.addAddressAddNSButton));
mouseClick(waitForObject($Names::address_book_myaddresses_adr_add_nstoolbaritem)); type(waitForObject($Names::add_address_nstextfield), "Jane"); type(waitForObject($Names::add_address_nstextfield), "<Tab>"); type(waitForObject($Names::add_address_nstextfield_2), "Doe"); type(waitForObject($Names::add_address_nstextfield_2), "<Tab>"); type(waitForObject($Names::add_address_nstextfield_3), "jane.doe\@nowhere.com"); type(waitForObject($Names::add_address_nstextfield_3), "<Tab>"); type(waitForObject($Names::add_address_nstextfield_4), "555 123 4567"); mouseClick(waitForObject($Names::add_address_add_nsbutton));
mouseClick(waitForObject(Names::Address_Book_MyAddresses_adr_Add_NSToolbarItem)) type(waitForObject(Names::Add_Address_NSTextField), "Jane") type(waitForObject(Names::Add_Address_NSTextField), "<Tab>") type(waitForObject(Names::Add_Address_NSTextField_2), "Doe") type(waitForObject(Names::Add_Address_NSTextField_2), "<Tab>") type(waitForObject(Names::Add_Address_NSTextField_3), "jane.doe@nowhere.com") type(waitForObject(Names::Add_Address_NSTextField_3), "<Tab>") type(waitForObject(Names::Add_Address_NSTextField_4), "555 123 4567") mouseClick(waitForObject(Names::Add_Address_Add_NSButton))
As you can see the tester used the keyboard to tab from one text field to another and clicked the Add button using the mouse rather than with a key press. If the tester had clicked the button any other way (for example, by pressing Enter), the outcome would be the same, but of course Squish will have recorded the actual actions that were taken.
Squish recordings refer to objects using variables that begin with a names.
prefix, which identifies them as Symbolic Names. Each variable contains, as a value, the corresponding Real Name, which can be string-based, or implemented as a key-value mapping of properties to values. Squish supports several naming schemes, all of which can be used—and mixed—in scripts. The advantage of using symbolic names is that if the application changes in a way that results in different names being needed, we can simply update Squish's Object Map (which relates symbolic names to real names), and thereby avoid the need to change our test scripts. See Object Map and Object Map view for more about the Object Map.
When a Symbolic Name is under the cursor, the editor's context menu allows you to Open Symbolic Name, showing its entry in the Object Map, or Convert to Real Name, which places an inline mapping in your script language at the cursor, allowing you to hand-edit the properties in the script itself.
Now that we have seen how to record and play back a test and have seen the code that Squish generates, let's go a step further and make sure that at particular points in the test's execution certain conditions hold.
Inserting Additional Verification Points
In the previous section we saw how easy it is to insert scriptified property verification points during the recording of test scripts. Verification points can also be inserted into existing test scripts, either by setting a breakpoint and using the squishide
, or simply by editing a test script and putting in calls to Squish's test functions such as Boolean test.compare(value1, value2) and Boolean test.verify(condition).
Squish supports many kinds of verification points: those that verify that object properties have particular values—known as "Object Property Verifications"; those that verify that an entire table has the contents we expect—known as "Table Verifications"; those that verify that two images match—known as "Screenshot Verifications"; and a hybrid verification type that includes properties and screenshots from multiple objects, known as "Visual Verifications". In addition, it is possible to verify that a search image exists somewhere on the screen, or that certain text is found by OCR. The most commonly used kind is object property verifications, and it is these that we will cover in the tutorial. For further reading, see How to Create and Use Verification Points).
Regular (non-scriptified) property verification points are stored as XML files in the test case or test suite resources, and contain the value(s) that need to be passed to test.compare()
. These verification points can be reused across test cases, and can verify many values in a single line of script code.
Scriptified property verification points are direct calls to the Boolean test.compare(value1, value2) function, with two arguments—the value of a particular property for a particular object, and an expected value. We can manually insert calls to the Boolean test.compare(value1, value2) function in a recorded or hand written script, or we can get Squish to insert them for us using scriptified verification points. In the previous section we showed how to use the squishide
to insert verifications during recording. Here we will first show how to use the squishide
to insert verifications into an existing test script, and then we will show how to insert a verification by hand.
Before asking Squish to insert verification points, it is best to make sure that we have a list of what we want to verify and when. There are many potential verifications we could add to the tst_general
test case, but since our concern here is simply to show how to do it, we will only do two—we will verify that the "Jane Doe" entry's email address and phone number match the ones entered, and put the verifications immediately after the ones we inserted during recording.
To insert a verification point using the squishide
we start by putting a break point in the script (whether recorded or manually written—it does not matter to Squish), at the point where we want to verify.
As the above screenshot shows, we have set a breakpoint near the end of the script. This is done simply by double-clicking, or selecting Add Breakpoint after bringing up the context menu from the gutter (next to the line number in the editor). We chose this line because it follows the script lines where the first address is removed, so at this point (just before invoking the File menu to close the application), the first address should be that of "Jane Doe". The screenshot shows the verifications that were entered using the squishide
during recording. Our additional verifications will follow them. (Note that your line number may be different if you recorded the test in a different way, for example, using keyboard shortcuts rather than clicking menu items.)
Having set the breakpoint, we now run the test as usual by clicking the Run Test ( ) button, or by clicking the Run > Run Test Case menu option. Unlike a normal test run the test will stop when the breakpoint is reached, and Squish's main window will reappear (which will probably obscure the AUT). At this point the squishide
will automatically switch to the Test Debugging Perspective.
Perspectives and Views
The squishide
works just like the Eclipse IDE. If you aren't used to Eclipse, it is crucial to understand the following key concepts: Views and Perspectives. In Eclipse, and therefore in the squishide
, a View is essentially a child window, such as a dock window or a tab in an existing window. A Perspective is a collection of views arranged together. Both are accessible through the Window menu.
The squishide
is supplied with the following perspectives:
- Test Management Perspective that the
squishide
starts with, and that is shown in all previous screenshots - Test Debugging Perspective
- Spy Perspective
You can change these perspectives to show additional views or to hide views that you don't want, or create your own perspectives with exactly the views you want. So if your windows change dramatically, it just means that the perspective changed. Use the Window menu to change back to the perspective you want. However, Squish automatically changes perspective to reflect the current situation, so you should not need to change perspective manually.
Inserting Verification Points
As the screenshot below shows, when Squish stops at a breakpoint the squishide
automatically changes to the Test Debugging Perspective. The perspective shows the Variables view, the Editor view, the Debug view, the Application Objects view, and the Properties view, Methods view, and Test Results view.
To insert a verification point we can expand items in the Application Objects view until we find the object we want to verify, or we can use the Object Picker ( ) to visually select it. In this example we want to verify the NSTableView
's first row's texts, so we expand the "Address Book - MyAddresses_adr_NSWindow_0" item, and its child items until we find the NSTableView
, and within that the item we are interested in. Once we click the item object its properties are shown in the Properties view as the screenshot shows.
The normal Test Management Perspective can be returned to at any time by choosing it from the Window menu (or by clicking its toolbar button), although the squishide
will automatically return to it if you stop the script or run it to completion.
Here, we can see that the stringValue
property of the first item in the table has the value "Jane"; we already have a verification for this that we inserted during recording. Scroll down so that you can see the corresponding (i.e., first) email address (the jane.doe@nowhere.com_NSCFString_250
item). To make sure that this is verified every time the test is run, select (without checking the box) this object in the Application Objects view to make its properties appear in the Properties view, and then click the stringValue
property to check its check box. After it is checked, the Verification Point Creator view appears as shown in the screenshot.
After the property is selected, the verification point has not yet been added to the test script. We could easily add it by clicking the Save and Insert Verifications button, but before doing that, set the Type of the verification to Property, since the combobox remembers the previously selected VP type. Choosing a meaningful name for the VP is also a good idea at this time. Otherwise we get names like VP1 and VP2, etc.
We'll add one more thing to be verified before we insert it into the script.
Scroll down and click the first phone number item (the 555 123 4567_NSCFString_375
item in the Application Objects view; then check its stringValue
property in the Properties view. Now both verifications will appear in the Verification Point Creator view as the next screenshot shows.
We have now said that we expect these properties to have the values shown, that is, an email address of "jane.doe@nowhere.com" and phone number of "555 123 4567". We must click the Save and Insert Verifications button to actually insert the verification point, so do that now.
We don't need to continue running the test now, so we can either stop running the test at this point (by clicking the Stop toolbar button), or we can continue (by clicking the Resume button).
Once we have finished inserting verifications and stopped or finished running the test we should now disable the break point. Just Ctrl+Click the break point and click the Disable Breakpoint menu option in the context menu. We are now ready to run the test without any breakpoints but with the verification points in place. Click the Run Test ( ) button. This time we will get some test results—as the screenshot shows—one of which we have expanded to show its details. (We have also selected the line of code that Squish inserted to perform the verification—notice that the name you picked for the VP is used here).
This particular verification point is stored as an XML file, containing two property comparisons for email, and phone number of the newly inserted entry. You can see it in the Test Case Resources window in the VPs tab, as well as open it in an editor from there.
Another way to insert verification points is to insert them in code form. In theory we can just add our own calls to Squish's test functions such as Boolean test.compare(value1, value2) and Boolean test.verify(condition) anywhere we like in an existing script. In practice it is best to make sure that Squish knows about the objects we want to verify first so that it can find them when the test is run. This involves a very similar procedure as using the squishide
. First we set a breakpoint where we intend adding our verifications. Then we run the test script until it stops. Next we navigate in the Application Objects view until we find the object we want to verify. At this point it is wise to Ctrl+Click the object we are interested in and click the Add to Object Map context menu option. This will ensure that Squish can access the object. Then Ctrl+Click again and click the Copy Symbolic Name context menu option—this gives us the name of the object that Squish will use to identify it. Now we can edit the test script to add in our own verification and finish or stop the execution. (Don't forget to disable the break point once it isn't needed any more.)
Although we can write our test script code to be exactly the same style as the automatically generated code, it is usually clearer and easier to do things in a slightly different style, as we will explain in a moment.
For our manual verifications we want to check the number of addresses present in the NSTableView
after reading in the MyAddresses.adr
file, then after the new address is added, and finally after the first address is removed. The screenshot shows two of the lines of code we entered to get one of these three verifications, plus the results of running the test script.
When writing scripts by hand, we use Squish's test
module's functions to verify conditions at certain points during our test script's execution. As the screenshot (and the code snippets below) show, we begin by retrieving a reference to the object we are interested in. Using the Object waitForObject(objectOrName) function is standard practice for manually written test scripts. This function waits for the object to be available (i.e., visible and enabled), and then returns a reference to it. (Otherwise it times out and raises a catchable exception.) We then use this reference to access the item's properties and methods—in this case the NSTableView
's numberOfRows
method—and verify that the value is what we expect it to be using the Boolean test.verify(condition) function. (Incidentally, we got the name for the object from a later line so we didn't need to set a breakpoint and manually add the table's name to the Object Map to ensure that Squish would remember it in this particular case because Squish had already added it during the test recording.)
Here is the code we entered manually for the first verification for all the scripting languages that Squish supports. Naturally, you only need to look at the code for the language that you will be using for your own tests. (For the other verifications we just did calls to the Boolean test.verify(condition) function—or to the Boolean test.compare(value1, value2) function for Tcl since it's more convenient—reusing the table
object reference we obtained in the code shown below.)
table = waitForObject(names.address_Book_MyAddresses_adr_NSTableView) test.compare(table.numberOfRows(), 125)
var table = waitForObject(names.addressBookMyAddressesAdrNSTableView); test.compare(table.numberOfRows(), 125);
my $table = waitForObject($Names::address_book_myaddresses_adr_nstableview); test::compare($table->numberOfRows(), 125);
table = waitForObject(Names::Address_Book_MyAddresses_adr_NSTableView) Test.compare(table.numberOfRows(), 125)
The coding pattern is very simple: we retrieve a reference to the object we are interested in and then verify its properties using one of Squish's verification functions. And we can, of course, call methods on the object to interact with it if we wish.
For more examples of manually written tests, see Creating Tests by Hand, How to Create Test Scripts, and How to Test Applications - Specifics.
For complete coverage of verification points, see How to Create and Use Verification Points.
Test Results
After each test run finishes, the test results—including those for the verification points—are shown in the Test Results view at the bottom of the squishide
.
This is a detailed report of the test run and would also contain details of any failures or errors, etc. If you click on a Test Results item, the squishide
highlights the script line which generated the test result. And if you expand a Test Results item, you can see additional details of the test.
Squish's interface for test results is very flexible. By implementing custom report generators it is possible to process test results in many different ways, for example to store them in a database, or to output them as HTML files. The default report generator simply prints the results to stdout
when Squish is run from the command line, or to the Test Results view when squishide
is being used. You can save the test results from the squishide
as XML by Ctrl+Clicking on the Test Results and choosing the Export Results menu option. For a list of report generators, see squishrunner –reportgen: Generating Reports. It is also possible to log test results directly to a database. See How to Access Databases from Squish Test Scripts.
If you run tests on the command line using squishrunner, you can also export the results in different formats and save them to files. See the sections Processing Test Results and How to Use Test Statements for more information.
Creating Tests by Hand
Now that we have seen how to record a test and modify it by inserting verification points, we are ready to see how to create tests manually. The easiest way to do this is to modify and refactor recorded tests, although it is also perfectly possible to create manual tests from scratch.
Potentially the most challenging part of writing manual tests is to use the right object names, but in practice, this is rarely a problem. We can either copy the symbolic names that Squish has already added to the Object Map when recording previous tests, or we can copy object names directly from recorded tests. And if we haven't recorded any tests and are starting from scratch we can use the Spy. We do this by clicking the Launch AUT toolbar button. This starts the AUT and switches to the Spy Perspective. We can then interact with the AUT until the object we are interested in is visible. Then, inside the squishide
we can navigate to the object in the Application Objects view and use the context menu to both add the object to the Object Map (so that Squish will remember it) and to the clipboard (so that we can paste it into our test script). And at the end we can click the Quit AUT toolbar button to terminate the AUT and return Squish to the Test Management Perspective. See How to Use the Spy and Spy Perspective for more details on using the Spy.
We can view the Object Map by clicking the Object Map toolbar button (see also, the Object Map view). Every application object that Squish interacts with is listed here, either as a top-level object, or as a child object (the view is a tree view). We can retrieve the symbolic name used by Squish in recorded scripts by Ctrl+Clicking the object we are interested in and then clicking the context menu's Copy item. This is useful for when we want to modify existing test scripts or when we want to create test scripts from scratch, as we will see later on in the tutorial.
Modifying and Refactoring Recorded Tests
Suppose we want to test the AUT's Add functionality by adding three new names and addresses. We could of course record such a test but it is just as easy to do everything in code. The steps we need the test script to do are: first click File > New to create a new address book, then for each new name and address, click the Add toolbar button, then fill in the details, and click Add. And finally, Quit without saving. We also want to verify at the start that there are no rows of data and at the end that there are three rows. We will also refactor as we go, to make our code as neat and modular as possible.
First we must create a new test case. Click File > New Test Case and set the test case's name to be tst_adding
. Squish will automatically create an test.py
(or test.js
, and so on) file.
Command line users can simply create a tst_adding
directory inside the test suite's directory and create and edit the test.py
file (or test.js
and so on) within that directory.
The first thing we need is a way to start the AUT and then invoke a menu option. Here are the first few lines from the recorded tst_general
script:
import names import os def main(): startApplication('"' + os.environ["SQUISH_PREFIX"] + '/examples/mac/addressbook/SquishAddressBook"') activateItem(waitForObject(names.file_NSMenuItem)) activateItem(waitForObject(names.open_NSMenuItem))
import * as names from 'names.js'; function main() { startApplication('"' + OS.getenv("SQUISH_PREFIX") + '/examples/mac/addressbook/SquishAddressBook"'); activateItem(waitForObject(names.fileNSMenuItem)); activateItem(waitForObject(names.openNSMenuItem));
require 'names.pl'; sub main { startApplication("\"$ENV{'SQUISH_PREFIX'}/examples/mac/addressbook/SquishAddressBook\""); activateItem(waitForObject($Names::file_nsmenuitem)); activateItem(waitForObject($Names::open_nsmenuitem));
require 'squish' require 'names' include Squish def main startApplication("\"#{ENV['SQUISH_PREFIX']}/examples/mac/addressbook/SquishAddressBook\"") activateItem(waitForObject(Names::File_NSMenuItem)) activateItem(waitForObject(Names::Open_NSMenuItem))
The pattern in the code is simple: start the AUT, then wait for the File menu to appear, then activate the File menu; wait for the menu item, then activate the menu item. In both cases we have used the Object waitForObject(objectOrName) and activateItem(objectOrName) functions that we've already seen.
Note: It may seem a waste to put our functions in tst_adding
because we could also use them in tst_general
and in other test cases. However, to keep the tutorial simple we will put the code in the tst_adding
test case. See How to Create and Use Shared Data and Shared Scripts for how to share scripts.
If you look at the recorded test (tst_general
) or in the Object Map you will see that Squish sometimes uses different names for the same things. For example, the window is identified in two different ways, initially as AddressBook_Untitled_NSWindow
, but if the user clicks File > Open and opens the MyAddresses.adr
file, the window is then identified as AddressBook_MyAddresses_adr_NSWindow
. The reason for this is that Squish needs to uniquely identify every object in a given context, and it uses whatever information it has to hand. So in the case of identifying Main Windows (and their children), Squish uses the window title text to give it some context. (For example, an application's File or Edit menus may have different options depending on whether a file is loaded and what state the application is in.)
Naturally, when we write test scripts we don't want to have to know or care which particular variation of a name to use, and Squish supports this need by providing alternative naming schemes, as we will see shortly.
If the AUT appears to freeze during test execution, wait for Squish to time out the AUT (about 20 seconds), and show the Object Not Found dialog, indicating an error like this:
This usually means that Squish doesn't have an object with the given name, or property values, in the Object Map. From here, we can Pick a new object, Debug, Throw Error or, after picking a new object, Retry.
Picking a new object will update the object map entry for the symbolic name. In addition to the Object Picker ( ) we can use the Spy's Application Objects view to locate the objects we are interested in and use the Add to the Object Map context menu action to to access their real or symbolic names.
Naming is important because it is probably the part of writing scripts that leads to the most error messages, usually of the object ... not found kind shown above. Once we have identified the objects to access in our tests, writing test scripts using Squish is very straightforward. Especially, as Squish most likely supports the scripting language you are most familiar with.
We are now almost ready to write our own test script. It is probably easiest to begin by recording a dummy test. So click File > New Test Case and set the test case's name to be tst_dummy
. Then click the dummy test case's Record ( ). Once the AUT starts, click File > New, then click the (empty) table, then click Add and add an item, then press Return or click OK. Finally, click SquishAddressBook > Quit Squish Address Book to finish, and say "No" to saving changes. Then replay this test just to confirm that everything works okay. The sole purpose of this is to make sure that Squish adds the necessary names to the Object Map since it is probably quicker to do it this way than to use the Spy for every object of interest. After replaying the dummy test you can delete it if you want to.
With all the object names we need in the Object Map we can now write our own test script completely from scratch. We will start with the main
function, and then we will look at the supporting functions that the main
function uses.
import names import os def main(): startApplication('"' + os.environ["SQUISH_PREFIX"] + '/examples/mac/addressbook/SquishAddressBook"') table = waitForObject(names.address_Book_Untitled_NSTableView) mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem)) test.verify(table.numberOfRows() == 0) data = [("Andy", "Beach", "andy.beach@nowhere.com", "555 123 6786"), ("Candy", "Deane", "candy.deane@nowhere.com", "555 234 8765"), ("Ed", "Fernleaf", "ed.fernleaf@nowhere.com", "555 876 4654")] for oneNameAndAddress in data: addNameAndAddress(oneNameAndAddress) test.compare(table.numberOfRows(), len(data)) closeWithoutSaving()
import * as names from 'names.js'; function main() { startApplication('"' + OS.getenv("SQUISH_PREFIX") + '/examples/mac/addressbook/SquishAddressBook"'); var table = waitForObject(names.addressBookUntitledNSTableView); mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem)); test.verify(table.numberOfRows() == 0); var data = new Array( new Array("Andy", "Beach", "andy.beach@nowhere.com", "555 123 6786"), new Array("Candy", "Deane", "candy.deane@nowhere.com", "555 234 8765"), new Array("Ed", "Fernleaf", "ed.fernleaf@nowhere.com", "555 876 4654")); for (var row = 0; row < data.length; ++row) addNameAndAddress(data[row]); test.compare(table.numberOfRows(), data.length); closeWithoutSaving(); }
require 'names.pl'; sub main { startApplication("\"$ENV{'SQUISH_PREFIX'}/examples/mac/addressbook/SquishAddressBook\""); my $table = waitForObject($Names::address_book_untitled_nstableview); mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem)); test::verify($table->numberOfRows() == 0); my @data = (["Andy", "Beach", "andy.beach\@nowhere.com", "555 123 6786"], ["Candy", "Deane", "candy.deane\@nowhere.com", "555 234 8765"], ["Ed", "Fernleaf", "ed.fernleaf\@nowhere.com", "555 876 4654"]); foreach $oneNameAndAddress (@data) { addNameAndAddress(@{$oneNameAndAddress}); } test::compare($table->numberOfRows(), scalar(@data)); closeWithoutSaving(); }
require 'squish' require 'names' include Squish def main startApplication("\"#{ENV['SQUISH_PREFIX']}/examples/mac/addressbook/SquishAddressBook\"") table = waitForObject(Names::Address_Book_Untitled_NSTableView) mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem)) Test.verify(table.numberOfRows() == 0) data = [["Andy", "Beach", "andy.beach@nowhere.com", "555 123 6786"], ["Candy", "Deane", "candy.deane@nowhere.com", "555 234 8765"], ["Ed", "Fernleaf", "ed.fernleaf@nowhere.com", "555 876 4654"]] data.each do |oneNameAndAddress| addNameAndAddress(oneNameAndAddress) end Test.compare(table.numberOfRows(), data.length) closeWithoutSaving end
We begin by starting the application with a call to the ApplicationContext startApplication(autName) function. The name we pass as a string is the name registered with Squish (normally the name of the executable). Then we obtain a reference to the NSTableView
. The object name we used was not put in the Object Map when the tst_general
test case was recorded, so we recorded a dummy test to make sure the name was added—we could just as easily have used the Spy of course. We then copied the name from the Object Map into our code. The Object waitForObject(objectOrName) function waits until an object is ready (visible and enabled) and returns a reference to it—or it times out and raises a catchable exception. We have used a symbolic name to access the table—these are the names that Squish uses when recording tests—rather than a real/multi-property name. It is best to use symbolic names where possible because if any AUT object name changes, with symbolic names we just have to update the Object Map, without needing to change our test code. Once we have the table
reference we can use it to access any of the NSTableView
's public methods and properties.
Once we have started the AUT and obtained a reference to the NSTableView
, we tell Squish to click the New toolbar button to create a new empty address book. Next, we verify that the table's row count is 0. The Boolean test.verify(condition) function is useful when we simply want to verify that a condition is true rather than compare two different values. (For Tcl we usually use the Boolean test.compare(value1, value2) function rather than the Boolean test.verify(condition) function simply because it is slightly simpler to use in Tcl.)
Next, we create some sample data and call a custom addNameAndAddress()
function to populate the table with the data using the AUT's Add sheet. Then we again compare the table's row count, this time to the number of rows in our sample data. And finally we call a custom closeWithoutSaving
function to terminate the application.
We will now review both of the supporting functions, so as to cover all the code in the tst_adding
test case, starting with addNameAndAddress()
.
def addNameAndAddress(oneNameAndAddress): mouseClick(waitForObject(names.address_Book_Untitled_Add_NSToolbarItem)) type(waitForObject(names.add_Address_NSTextField), oneNameAndAddress[0]) type(waitForObject(names.add_Address_NSTextField_2), oneNameAndAddress[1]) type(waitForObject(names.add_Address_NSTextField_3), oneNameAndAddress[2]) type(waitForObject(names.add_Address_NSTextField_4), oneNameAndAddress[3]) mouseClick(waitForObject(names.add_Address_Add_NSButton));
function addNameAndAddress(oneNameAndAddress) { mouseClick(waitForObject(names.addressBookUntitledAddNSToolbarItem)); type(waitForObject(names.addAddressNSTextField), oneNameAndAddress[0]); type(waitForObject(names.addAddressNSTextField_2), oneNameAndAddress[1]); type(waitForObject(names.addAddressNSTextField_3), oneNameAndAddress[2]); type(waitForObject(names.addAddressNSTextField_4), oneNameAndAddress[3]); mouseClick(waitForObject(names.addAddressAddNSButton)); }
sub addNameAndAddress { my (@oneNameAndAddress) = @_; mouseClick(waitForObject($Names::address_book_untitled_add_nstoolbaritem)); type(waitForObject($Names::add_address_nstextfield), $oneNameAndAddress[0]); type(waitForObject($Names::add_address_nstextfield_2), $oneNameAndAddress[1]); type(waitForObject($Names::add_address_nstextfield_3), $oneNameAndAddress[2]); type(waitForObject($Names::add_address_nstextfield_4), $oneNameAndAddress[3]); mouseClick(waitForObject($Names::add_address_add_nsbutton)); }
def addNameAndAddress(oneNameAndAddress) mouseClick(waitForObject(Names::Address_Book_Untitled_Add_NSToolbarItem)) type(waitForObject(Names::Add_Address_NSTextField), oneNameAndAddress[0]) type(waitForObject(Names::Add_Address_NSTextField_2), oneNameAndAddress[1]) type(waitForObject(Names::Add_Address_NSTextField_3), oneNameAndAddress[2]) type(waitForObject(Names::Add_Address_NSTextField_4), oneNameAndAddress[3]) mouseClick(waitForObject(Names::Add_Address_Add_NSButton)) end
For each set of name and address data we click Add to pop up the Add sheet. Then, for each value received, we populate the appropriate field by waiting for the relevant NSTextField
to be ready and then typing in the text using the type(objectOrName, text) function. Finally, we click the sheet's Add button. We got the line at the heart of the function by copying it from the recorded tst_general
test and simply parametrizing it by the field name and text. Similarly, we copied the code for clicking the Add button from the tst_general
test case's code.
def closeWithoutSaving(): type(waitForObject(names.address_Book_Untitled_NSTableView), "<Command+q>") mouseClick(waitForObject(names.save_Changes_Don_t_Save_NSButton))
function closeWithoutSaving() { type(waitForObject(names.addressBookUntitledNSTableView), "<Command+q>"); mouseClick(waitForObject(names.saveChangesDonTSaveNSButton)); }
sub closeWithoutSaving { type(waitForObject($Names::address_book_untitled_nstableview), "<Command+q>"); mouseClick(waitForObject($Names::save_changes_don_t_save_nsbutton)); }
def closeWithoutSaving type(waitForObject(Names::Address_Book_Untitled_NSTableView), "<Command+q>") mouseClick(waitForObject(Names::Save_Changes_Don_t_Save_NSButton)) end
Here we quit the application by pressing Command+Q and then click the "save changes" dialog's Don't Save button. The last line was copied from the recorded test.
The entire test is about 25 lines of code—and would be even less if we put some of the common functions (such as closeWithoutSaving()
) in a shared script. And much of the code was copied directly from the recorded test, and in some cases parametrized.
This should be sufficient to give a flavor of writing test scripts for an AUT. Keep in mind that Squish provides far more functionality than we used here. (All of which is covered in the API Reference and the Tools Reference.) And Squish also provides access to the entire public APIs of the AUT's objects.
However, one aspect of the test case is not very satisfactory. Although embedding test data as we did here is sensible for small amounts, it is rather limiting, especially when we want to use a lot of test data. Also, we didn't test any of the data that was added to see if it correctly ended up in the NSTableView
. In the next section, we will create a new version of this test, only this time we will pull in the data from an external data source, and check that the data we add to the NSTableView
is correct.
Creating Data Driven Tests
In the previous section we put three hard-coded names and addresses in our test. But what if we want to test lots of data? Or what if we want to change the data without having to change our test script's source code? One approach is to import a dataset into Squish and use the dataset as the source of the values we insert into our tests. Squish can import data in .tsv
(tab-separated values format), .csv
(comma-separated values format), .xls
, or .xlsx
(Microsoft Excel spreadsheet formats).
Note: Both .csv
and .tsv
files are assumed to use the Unicode UTF-8 encoding—the same encoding used for all test scripts.
Test data can either be imported using the squishide
, or manually using a file manager or console commands. We will describe both approaches, starting with using the squishide
.
For the SquishAddressBook application we want to import the MyAddresses.tsv
data file. To do this we must start by clicking File > Import Test Resource to pop-up the Import Squish Resource dialog. Inside the dialog click the Browse button to choose the file to import—in this case MyAddresses.tsv
. Make sure that the Import As combobox is set to "TestData". By default the squishide
will import the test data just for the current test case, but we want the test data to be available to all the test suite's test cases: to do this check the Copy to Test Suite for Sharing radio button. Now click the Finish button. You can now see the file listed in the Test Suite Resources view (in the Test Data tab), and if you click the file's name it will be shown in an Editor view. The screenshot shows Squish after the test data has been added.
To import test data from outside the squishide
, use a file manager, such as File Explorer or Finder, or console commands. Create a directory called shared
inside the test suite's directory. Then, create a directory called testdata
inside the shared
directory. Copy the data file (in this example, MyAddresses.tsv
) into the shared\testdata
directory.
Restart the squishide
if it is running. If you click the Test Suite Resources view's Test Data tab, you should see the data file. Click the file name to see the file in an Editor view.
Although in real life we would modify our tst_adding
test case to use the test data, for the purpose of the tutorial we will make a new test case called tst_adding_data
that is a copy of tst_adding
and which we will modify to make use of the test data.
The only function we have to change is main
, where instead of iterating over hard-coded items of data, we iterate over all the records in the dataset. We also need to update the expected row count at the end since we are adding a lot more records now, and we will also add a function to verify each record that's added.
import names import os def main(): startApplication('"' + os.environ["SQUISH_PREFIX"] + '/examples/mac/addressbook/SquishAddressBook"') table = waitForObject(names.address_Book_Untitled_NSTableView) mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem)) test.verify(table.numberOfRows() == 0) limit = 10 # To avoid testing 100s of rows for row, record in enumerate(testData.dataset("MyAddresses.tsv")): forename = testData.field(record, "Forename") surname = testData.field(record, "Surname") email = testData.field(record, "Email") phone = testData.field(record, "Phone") addNameAndAddress((forename, surname, email, phone)) # pass a single tuple checkNameAndAddress(row, table, record) if row > limit: break test.compare(table.numberOfRows(), row + 1) closeWithoutSaving()
import * as names from 'names.js'; function main() { startApplication('"' + OS.getenv("SQUISH_PREFIX") + '/examples/mac/addressbook/SquishAddressBook"'); var table = waitForObject(names.addressBookUntitledNSTableView); mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem)); test.verify(table.numberOfRows() == 0); var limit = 10; // To avoid testing 100s of rows since that would be boring var records = testData.dataset("MyAddresses.tsv"); for (var row = 0; row < records.length; ++row) { var record = records[row]; var forename = testData.field(record, "Forename"); var surname = testData.field(record, "Surname"); var email = testData.field(record, "Email"); var phone = testData.field(record, "Phone"); addNameAndAddress(new Array(forename, surname, email, phone)); checkNameAndAddress(row, table, record); if (row > limit) break; } test.compare(table.numberOfRows(), row + 1); closeWithoutSaving(); }
require 'names.pl'; sub main { startApplication("\"$ENV{'SQUISH_PREFIX'}/examples/mac/addressbook/SquishAddressBook\""); my $table = waitForObject($Names::address_book_untitled_nstableview); mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem)); test::verify($table->numberOfRows() == 0); my $limit = 10; # To avoid testing 100s of rows since that would be boring my @records = testData::dataset("MyAddresses.tsv"); my $row = 0; for (; $row < scalar(@records); ++$row) { my $record = $records[$row]; my $forename = testData::field($record, "Forename"); my $surname = testData::field($record, "Surname"); my $email = testData::field($record, "Email"); my $phone = testData::field($record, "Phone"); addNameAndAddress($forename, $surname, $email, $phone); checkNameAndAddress($row, $table, $record); if ($row > $limit) { last; } } test::compare($table->numberOfRows(), $row + 1); closeWithoutSaving(); }
require 'squish' require 'names' include Squish def main startApplication("\"#{ENV['SQUISH_PREFIX']}/examples/mac/addressbook/SquishAddressBook\"") table = waitForObject(Names::Address_Book_Untitled_NSTableView) mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem)) Test.verify(table.numberOfRows() == 0) limit = 10 # To avoid testing 100s of rows rows = 0 TestData.dataset("MyAddresses.tsv").each_with_index do |record, row| forename = TestData.field(record, "Forename") surname = TestData.field(record, "Surname") email = TestData.field(record, "Email") phone = TestData.field(record, "Phone") addNameAndAddress([forename, surname, email, phone]) # pass as a single Array checkNameAndAddress(row, table, record) break if row > limit rows += 1 end Test.verify(table.numberOfRows() == rows + 1) closeWithoutSaving end
Squish provides access to test data through its testData
module's functions—here we used the Dataset testData.dataset(filename) function to access the data file and make its records available, and the String testData.field(record, fieldName) function to retrieve each record's individual fields.
Having used the test data to populate the NSTableView
we want to be confident that the data in the table is the same as what we have added, so that's why we added the checkNameAndAddress
function. We also added a limit to how many records we would compare, just to make the test run faster.
def checkNameAndAddress(row, table, record): for column in range(len(testData.fieldNames(record))): cellName = {'rowNumber': row, 'columnNumber': column, 'tableView': names.address_Book_Untitled_NSTableView, 'type': 'NSString'} cell = waitForObject(cellName) test.compare(cell.stringValue, testData.field(record, column))
function checkNameAndAddress(row, table, record) { for (var column = 0; column < testData.fieldNames(record).length; ++column) { var cellName = {'rowNumber': row, 'columnNumber': column, 'tableView': names.addressBookUntitledNSTableView, 'type': 'NSString'}; var cell = waitForObject(cellName); test.compare(cell.stringValue, testData.field(record, column)); } }
def checkNameAndAddress(row, table, record) for column in 0...TestData.fieldNames(record).length cellName = {:rowNumber => row, :columnNumber => column, :tableView => Names::Address_Book_Untitled_NSTableView, :type => 'NSString'} cell = waitForObject(cellName) Test.compare(cell.stringValue, TestData.field(record, column)) end end
This function accesses the give NSTableView
row and extracts each of its columns' values. We use Squish's SequenceOfStrings testData.fieldNames(record) function to get a column count and then use the Boolean test.compare(value1, value2) function to check that each value in the table is the same as the value in the test data we used.
As we mentioned earlier, the symbolic names Squish uses for menus and menu items (and other objects) can vary depending on the context, and often with the start of the name derived from the window's title. For applications that put the current filename in the title—such as the Address Book example—names will include the filename, and we must account for this.
In the case of the Address Book example, the main window's title is "Address Book" (at startup), or "Address Book - Unnamed" (after File > New , but before File > Save or File > Save As), or "Address Book - filename" where the filename can of course vary. Our code accounts for all these cases by making use of real (multi-property) names.
Symbolic names embed various bits of information about an object and its type. Real names are represented by a brace enclosed list of space-separated key–value pairs. Every real name must specify the type
property and at least one other property. Here we've used the rowNumber
, columnNumber
, tableView
, and type
properties to uniquely identify each cell in the table so that we can compare each cell's stringValue with the data used to populate it.
The screenshot show Squish's Test Summary log after the data-driven tests have been run.
Squish can also do keyword-driven testing. This is a bit more sophisticated than data-driven testing. See How to Do Keyword-Driven Testing.
Learning More
We have now completed the tutorial. Squish can do much more than we have shown here, but the aim has been to get you started with basic testing as quickly and easily as possible. The How to Create Test Scripts, and How to Test Applications - Specifics sections provide many more examples, including those that show how tests can interact with particular input elements, such as selects, select-ones, texts, and text-areas.
The API Reference and Tools Reference give full details of Squish's testing API and the numerous functions it offers to make testing as easy and efficient as possible. It is well worth reading the How to Create Test Scripts and How to Test Applications - Specifics, as well as browsing the API Reference and Tools Reference. The time you invested will be repaid because you'll know what functionality Squish provides out of the box and can avoid reinventing things that are already available.
If you are interested in testing iPhone Apps, see Tutorial: Starting to Test iOS Applications.
Tutorial: Designing Behavior Driven Development (BDD) Tests
This tutorial will show you how to create, run, and modify Behavior Driven Development (BDD) tests for an example application. You will learn about Squish's most frequently used features. By the end of the tutorial you will be able to write tests for your own applications.
For this chapter we will use a simple address book application as our Application Under Test (AUT). It is called SquishAddressBook.app in your examples/addressbook
folder. This is a very basic application that allows users to load an existing address book or create a new one, add, edit, and remove entries. The screenshot shows the application with a newly created, empty address book.
Introduction to Behavior Driven Development
Behavior-Driven Development (BDD) is an extension of the Test-Driven Development approach which puts the definition of acceptance criteria at the beginning of the development process, as opposed to writing tests after the software has been developed. This process also allows for possible cycles of code changes after testing.
Behavior Driven Tests are built out of a set of Feature
files, which describe product features through the expected application behavior in one or many Scenarios
. Each Scenario
is built out of a sequence of steps which represent actions or verifications that need to be tested for that Scenario
.
BDD focuses on expected application behavior, not on implementation details. Therefore BDD tests are described in a human-readable Domain Specific Language (DSL). As this language is not technical, such tests can be created not only by programmers, but also by product owners, testers or business analysts. Additionally, during the product development, such tests serve as living product documentation. For Squish usage, BDD tests shall be created using Gherkin syntax. The previously written product specification (BDD tests) can be turned into executable tests. This step by step tutorial presents automating BDD tests with squishide
support.
Gherkin Syntax
Gherkin files describe product features through the expected application behavior in one or many Scenarios. An example showing the "Filling of the address book" feature of the address book example application.
Feature: Filling of the address book As a user I want to fill the address book with entries Scenario: Initial state of the newly created address book Given the address book application is running When I create a new address book Then the address book should be empty Scenario: State after adding one entry Given the address book application is running When I create a new address book And I add a new person 'John','Doe','john@m.com','500600700' to the address book Then '1' entries should be present Scenario: State after adding two entries Given the address book application is running When I create a new address book And I add new persons to the address book | forename | surname | email | phone | | John | Smith | john@m.com | 123123 | | Alice | Thomson | alice@m.com | 234234 | Then '2' entries should be present Scenario: Forename and surname is added to table Given the address book application is running When I create a new address book And I add a new person 'Bob','Doe','Bob@m.com','123321231' to the address book Then previously entered forename and surname shall be at the top
Most of the above is free form text (does not have to be English). It's just the Feature
/Scenario
structure and the leading keywords like Given
, And
, When
and Then
that are fixed. Each of those keywords marks a step defining preconditions, user actions and expected results. The application behavior description above can be passed between software developers to implement the features, while at the same time, the same description can be passed to software testers to implement automated tests.
Test Implementation
Creating a Test Suite
First, we need to create a test suite, which is a container for all test cases. Start the squishide and select File > New Test Suite. Follow the New Test Suite wizard to provide a test suite name, choose the Mac toolkit and a scripting language of your choice and finally register the SquishAddressBook
application as AUT. See Creating a Test Suite for more details about creating new test suites.
Creating a Test Case
Squish offers two types of test cases: "Script Test Case" and "BDD Test Case". As "Script Test Case" is the default one, in order to create a new BDD Test Case we need to use the drop-down menu by clicking on the expander next to New Script Test Case ( ) and choosing New BDD Test Case. The squishide
will remember your choice and "BDD Test Case" will become the default when clicking on the button in the future.
The newly created BDD test case consists of a test.feature
file (filled with a Gherkin template while creating a new BDD test case), a file named test.(py|js|pl|rb|tcl)
which will drive the execution (there is no need to edit this file), and a test suite resources file named shared/steps/steps.(py|js|pl|rb|tcl)
where step implementation code will be placed.
We need to replace the Gherkin template with a Feature
for the example application. To do this, copy the Feature
description below and paste it into the Feature
file.
Feature: Filling of the address book
As a user I want to fill the address book with entries
Scenario: Initial state of the newly created address book
Given the address book application is running
When I create a new address book
Then the address book should be empty
When editing the test.feature
file, a Feature
file warning No implementation found is displayed for each undefined step. The implementations are in the steps
subdirectory, in Test Case Resources, or in Test Suite Resources. Running our Feature
test now will currently fail at the first step with a No Matching Step Definition and the following steps will be skipped.
Recording Step Implementation
In order to record the Scenario
, press the Record ( ) next to the respective Scenario
that is listed in the Scenarios tab in the Test Case Resources view.
This will cause Squish to run the AUT so that we can interact with it. Additionally, the Control Bar is displayed with a list of all steps that need to be recorded. Now all interaction with the AUT or any verification points added to the script will be recorded under the currently recording step, Given the address book application is running
(which is bolded in the Step list on the Control Bar). In order to verify that this precondition is met, we will add a Verification Point. To do this, click on Verify button in the Control Bar and then select Properties.
As a result the squishide
is put into Spy mode which displays all Application Objects and their Properties in dockable Views. In the Application Objects view, select the main window item (Address Book - Untitled_NSWindow
). Selecting it will update the Properties view on the right side. Next click on the checkbox in front of the property isVisible in the Properties View. Finally, click on the button Save and Insert Verifications. The squishide
disappears and the Control Bar is shown again.
When we are done with each Step, we can move to the next undefined step (playing back the ones that were previously defined) by clicking on the Finish Recording Step ( ) arrow button in the Control Bar that is located to the left of the current step.
Next, for the Step When I create a new address book
, click on the New button ( ) in the toolbar of the addressbook application and click on the Finish Recording Step ( ) in the Squish control bar, to finish recording and go to the next step.
Finally, for the step the address book should be empty
, verify that the table containing the address entries is empty. To record this verification, click on Verify while recording and select Properties. Now, in Application Objects, navigate or use the Object Picker ( ) to select the table containing the address book entries (in our case this table is empty). You will notice that the Table
object does not provide a property for the number of rows. This information is not directly available. Therefore we need to modify the recorded code later on. To let Squish generate some test.compare()
code for our step, insert a verification for a random property, for example alignment. Finally, click on the last Finish Recording Step ( ) arrow button in the Control Bar.
As a result, Squish will generate the following step definitions in the steps.*
file (in the Steps tab of the Test Suite Resources view):
@Given("the address book application is running") def step(context): startApplication('"' + os.environ["SQUISH_PREFIX"] + '/examples/mac/addressbook/SquishAddressBook"') test.compare(waitForObjectExists(names.o_NSWindow).isVisible, 0) @When("I create a new address book") def step(context): mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem)) @Then("the address book should be empty") def step(context): #test.compare(waitForObjectExists(names.address_Book_Untitled_NSTableView).alignment, 0) test.compare(waitForObjectExists(names.address_Book_Untitled_NSTableView).numberOfRows(), 0)
Given("addressbook application is running", function(context) { startApplication('"' + OS.getenv("SQUISH_PREFIX") + '/examples/mac/addressbook/SquishAddressBook"'); test.compare(waitForObjectExists(names.addressBookUntitledNSWindow).isVisible, 1); }); When("I create a new addressbook", function(context) { mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem)); }); Then("addressbook should have zero entries", function(context) { //test.compare((waitForObject(names.addressBookUntitledNSTableView)).alignment, 0); test.compare((waitForObject(names.addressBookUntitledNSTableView)).numberOfRows(), 0); });
Given("the address book application is running", sub { my $context = shift; startApplication("\"$ENV{'SQUISH_PREFIX'}/examples/mac/addressbook/SquishAddressBook\""); test::compare(waitForObjectExists($Names::address_book_untitled_nswindow)->isVisible, 1); }); When("I create a new address book", sub { my $context = shift; mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem)); }); Then("the address book should be empty", sub { my $context = shift; #test::compare(waitForObjectExists($Names::address_book_untitled_nstableview)->alignment, 1); test::compare(waitForObjectExists($Names::address_book_untitled_nstableview)->numberOfRows, 0); });
Given("the address book application is running") do |context| startApplication("\"#{ENV['SQUISH_PREFIX']}/examples/mac/addressbook/SquishAddressBook\"") Test.compare(waitForObjectExists(Names::O_NSWindow).isVisible, 0) end When("I create a new address book") do |context| mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem)) end Then("the address book should be empty") do |context| #Test.compare(waitForObjectExists(Names::Address_Book_Untitled_NSTableView).alignment, 0) Test.compare(waitForObjectExists(Names::Address_Book_Untitled_NSTableView).numberOfRows(), 0) end
In order to get the number of rows, we have to call a method of the underlying NSTableView
object.
The application is automatically started at the beginning of the first step due to the recorded startApplication()
call. At the end of each Scenario, the OnScenarioEnd
hook is called, causing detach()
to be called on the application context. Because the AUT was started with startApplication()
, this causes it to terminate. This hook function is found in the file bdd_hooks.(py|js|pl|rb|tcl)
, which is located in the Scripts tab of the Test Suite Resources view. You can define additional hook functions there. For a list of all available hooks, please refer to Performing Actions During Test Execution Via Hooks.
@OnScenarioEnd def OnScenarioEnd(): for ctx in applicationContextList(): ctx.detach()
OnScenarioEnd(function(context) { applicationContextList().forEach(function(ctx) { ctx.detach(); }); });
OnScenarioEnd(sub { foreach (applicationContextList()) { $_->detach(); } });
OnScenarioEnd do |context| applicationContextList().each { |ctx| ctx.detach() } end
Step Parametrization
So far, our steps did not use any parameters and all values were hardcoded. Squish has different types of parameters like any
, integer
or word
, allowing our step definitions to be more reusable. Let us add a new Scenario
to our Feature
file which will provide step parameters for both the Test Data and the expected results. Copy the below section into your Feature file.
Scenario: State after adding one entry Given the address book application is running When I create a new address book And I add a new person 'John','Doe','john@m.com','500600700' to the address book Then '1' entries should be present
After auto-saving the Feature
file, the squishide
provides a hint that only 2 steps need to be implemented: When I add a new person 'John', 'Doe','john@m.com','500600700' to the address book
and Then '1' entries should be present
. The remaining steps already have a matching implementation.
To record the missing Steps, hit the Record ( ) next to the test case name in the Test Suites view. The script will play until it gets to the missing step and then prompt you to implement it. If you select the Add button, then you can type in the information for a new entry. Click on Finish Recording Step ( ) to move to the next step. For the second missing step, we could record an object property verification like we did with the step, the address book should be empty
. Or we could copy that step's implementation in the steps.(py|js|pl|rb|tcl)
file and increment the number at the end of the test.compare()
line. Instead of testing for zero items, we are testing for one item.
Now we parametrize the generated step implementation by replacing the values with parameter types. Since we want to be able to add different names, replace John
with '|word|'
. Note that each parameter will be passed to the step implementation function in the order of appearance in the descriptive name of the step. Finish parametrizing by editing the typed values into keywords, to look like this example step, When I add a new person 'John', 'Doe', 'john@m.com', '500600700' to the address book
:
@When("I add a new person '|word|','|word|','|any|','|integer|' to the address book") def step(context, forename, surname, email, phone): mouseClick(waitForObject(names.address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(names.add_Address_NSTextField), 42.6172, 5.125, 0, 0) type(waitForObject(names.add_Address_NSTextField), forename) mouseClick(waitForObject(names.add_Address_NSTextField_2), 95.7852, 14.4336, 0, 0) type(waitForObject(names.add_Address_NSTextField_2), surname) mouseClick(waitForObject(names.add_Address_NSTextField_3), 93.7422, 9.19531, 0, 0) type(waitForObject(names.add_Address_NSTextField_3), email) mouseClick(waitForObject(names.add_Address_NSTextField_4), 98.6094, 15.3789, 0, 0) type(waitForObject(names.add_Address_NSTextField_4), phone) mouseClick(waitForObject(names.add_Address_Add_NSButton)) context.userData = {'forename': forename, 'surname': surname}
When("I add a new person '|word|','|word|','|any|','|integer|' to address book", function(context, forename, surname, email, phone) { mouseClick(waitForObject(names.addressBookUntitledAddNSToolbarItem)); mouseClick(waitForObject(names.addAddressNSTextField), 54.3047, 9.10156, 0, 0); type(waitForObject(names.addAddressNSTextField), forename); mouseClick(waitForObject(names.addAddressNSTextField_2), 45.3555, 10.9453, 0, 0); type(waitForObject(names.addAddressNSTextField_2), surname); mouseClick(waitForObject(names.addAddressNSTextField_3), 49.4805, 8.01562, 0, 0); type(waitForObject(names.addAddressNSTextField_3), email); mouseClick(waitForObject(names.addAddressNSTextField_4), 46.3125, 12.1641, 0, 0); type(waitForObject(names.addAddressNSTextField_4), phone); mouseClick(waitForObject(names.addAddressAddNSButton)); context.userData["forename"] = forename; context.userData["surname"] = surname; });
When("I add a new person '|word|','|word|','|any|','|integer|' to the address book", sub { my $context = shift; my $forename = shift; my $surname = shift; my $email = shift; my $phone = shift; mouseClick( waitForObject($Names::address_book_untitled_add_nstoolbaritem) ); type( waitForObject($Names::add_address_nstextfield), $forename ); mouseClick( waitForObject($Names::add_address_nstextfield_2), 62.4609, 14.2227, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_2), $surname ); mouseClick( waitForObject($Names::add_address_nstextfield_3), 60.375, 15.5, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_3), $email ); mouseClick( waitForObject($Names::add_address_nstextfield_4), 108.117, 15.7734, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_4), $phone ); mouseClick( waitForObject($Names::add_address_add_nsbutton) ); $context->{userData}{'forename'} = $forename; $context->{userData}{'surname'} = $surname; });
When("I add a new person '|word|','|word|','|any|','|integer|' to the address book") do |context, forename, surname, email, phone| mouseClick(waitForObject(Names::Address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(Names::Add_Address_NSTextField), 57.1602, 4.91016, 0, 0) type(waitForObject(Names::Add_Address_NSTextField), forename) mouseClick(waitForObject(Names::Add_Address_NSTextField_2), 71.3672, 17.5703, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_2), surname) mouseClick(waitForObject(Names::Add_Address_NSTextField_3), 71.3672, 16.5, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_3), email) mouseClick(waitForObject(Names::Add_Address_NSTextField_4), 79.707, 16.7734, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_4), phone) mouseClick(waitForObject(Names::Add_Address_Add_NSButton)) context.userData = Hash.new context.userData[:forename] = forename context.userData[:surname] = surname end
Provide parameters for Step in table
The next Scenario
will test adding multiple entries to the address book. We could use step When I add a new person John','Doe','john@m.com','500600700' to the address book
multiple times just with different data. But lets instead define a new step called And I add new persons to the address book
which will handle data from a table.
Scenario: State after adding two entries Given the address book application is running When I create a new address book And I add new persons to the address book | forename | surname | email | phone | | John | Smith | john@m.com | 123123 | | Alice | Thomson | alice@m.com | 234234 | Then '2' entries should be present
The step implementation to handle such tables looks like this:
@When("I add new persons to the address book") def step(context): table = context.table # Drop initial row with column headers for row in table[1:]: forename = row[0] surname = row[1] email = row[2] phone = row[3] mouseClick(waitForObject(names.address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(names.add_Address_NSTextField), 42.6172, 5.125, 0, 0) type(waitForObject(names.add_Address_NSTextField), forename) mouseClick(waitForObject(names.add_Address_NSTextField_2), 95.7852, 14.4336, 0, 0) type(waitForObject(names.add_Address_NSTextField_2), surname) mouseClick(waitForObject(names.add_Address_NSTextField_3), 93.7422, 9.19531, 0, 0) type(waitForObject(names.add_Address_NSTextField_3), email) mouseClick(waitForObject(names.add_Address_NSTextField_4), 98.6094, 15.3789, 0, 0) type(waitForObject(names.add_Address_NSTextField_4), phone) mouseClick(waitForObject(names.add_Address_Add_NSButton))
When("I add new persons to address book", function(context) { var table = context.table; // Drop initial row with column headers for (var i = 1; i < table.length; ++i) { var forename = table[i][0]; var surname = table[i][1]; var email = table[i][2]; var phone = table[i][3]; mouseClick(waitForObject(names.addressBookUntitledAddNSToolbarItem)); mouseClick(waitForObject(names.addAddressNSTextField), 54.3047, 9.10156, 0, 0); type(waitForObject(names.addAddressNSTextField), forename); mouseClick(waitForObject(names.addAddressNSTextField_2), 45.3555, 10.9453, 0, 0); type(waitForObject(names.addAddressNSTextField_2), surname); mouseClick(waitForObject(names.addAddressNSTextField_3), 49.4805, 8.01562, 0, 0); type(waitForObject(names.addAddressNSTextField_3), email); mouseClick(waitForObject(names.addAddressNSTextField_4), 46.3125, 12.1641, 0, 0); type(waitForObject(names.addAddressNSTextField_4), phone); mouseClick(waitForObject(names.addAddressAddNSButton)); } });
When("I add new persons to the address book", sub { my %context = %{shift()}; my @table = @{$context{'table'}}; # Drop initial row with column headers shift(@table); for my $row (@table) { my ($forename, $surname, $email, $phone) = @{$row}; mouseClick( waitForObject($Names::address_book_untitled_add_nstoolbaritem) ); type( waitForObject($Names::add_address_nstextfield), $forename ); mouseClick( waitForObject($Names::add_address_nstextfield_2), 62.4609, 14.2227, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_2), $surname ); mouseClick( waitForObject($Names::add_address_nstextfield_3), 60.375, 15.5, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_3), $email ); mouseClick( waitForObject($Names::add_address_nstextfield_4), 108.117, 15.7734, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_4), $phone ); mouseClick( waitForObject($Names::add_address_add_nsbutton) ); } });
When("I add new persons to the address book") do |context| table = context.table # Drop initial row with column headers table.shift for forename,surname,email,phone in table do mouseClick(waitForObject(Names::Address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(Names::Add_Address_NSTextField), 57.1602, 4.91016, 0, 0) type(waitForObject(Names::Add_Address_NSTextField), forename) mouseClick(waitForObject(Names::Add_Address_NSTextField_2), 71.3672, 17.5703, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_2), surname) mouseClick(waitForObject(Names::Add_Address_NSTextField_3), 71.3672, 16.5, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_3), email) mouseClick(waitForObject(Names::Add_Address_NSTextField_4), 79.707, 16.7734, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_4), phone) mouseClick(waitForObject(Names::Add_Address_Add_NSButton)) end end
Sharing Data Between Steps and Scenarios
Lets add a new Scenario
to the Feature
file. This time we would like to check not the number of entries in the address book list, but if this list contains proper data. Because we enter data into the address book in one step and verify them in another, we must somehow share information about entered data between those steps in order to perform a verification.
Scenario: Forename and surname is added to table Given the address book application is running When I create a new address book And I add a new person 'Bob','Doe','Bob@m.com','123321231' to the address book Then previously entered forename and surname shall be at the top
To share this data, context.userData can be used.
@When("I add a new person '|word|','|word|','|any|','|integer|' to the address book") def step(context, forename, surname, email, phone): mouseClick(waitForObject(names.address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(names.add_Address_NSTextField), 42.6172, 5.125, 0, 0) type(waitForObject(names.add_Address_NSTextField), forename) mouseClick(waitForObject(names.add_Address_NSTextField_2), 95.7852, 14.4336, 0, 0) type(waitForObject(names.add_Address_NSTextField_2), surname) mouseClick(waitForObject(names.add_Address_NSTextField_3), 93.7422, 9.19531, 0, 0) type(waitForObject(names.add_Address_NSTextField_3), email) mouseClick(waitForObject(names.add_Address_NSTextField_4), 98.6094, 15.3789, 0, 0) type(waitForObject(names.add_Address_NSTextField_4), phone) mouseClick(waitForObject(names.add_Address_Add_NSButton)) context.userData = {'forename': forename, 'surname': surname}
When("I add a new person '|word|','|word|','|any|','|integer|' to address book", function(context, forename, surname, email, phone) { mouseClick(waitForObject(names.addressBookUntitledAddNSToolbarItem)); mouseClick(waitForObject(names.addAddressNSTextField), 54.3047, 9.10156, 0, 0); type(waitForObject(names.addAddressNSTextField), forename); mouseClick(waitForObject(names.addAddressNSTextField_2), 45.3555, 10.9453, 0, 0); type(waitForObject(names.addAddressNSTextField_2), surname); mouseClick(waitForObject(names.addAddressNSTextField_3), 49.4805, 8.01562, 0, 0); type(waitForObject(names.addAddressNSTextField_3), email); mouseClick(waitForObject(names.addAddressNSTextField_4), 46.3125, 12.1641, 0, 0); type(waitForObject(names.addAddressNSTextField_4), phone); mouseClick(waitForObject(names.addAddressAddNSButton)); context.userData["forename"] = forename; context.userData["surname"] = surname; });
When("I add a new person '|word|','|word|','|any|','|integer|' to the address book", sub { my $context = shift; my $forename = shift; my $surname = shift; my $email = shift; my $phone = shift; mouseClick( waitForObject($Names::address_book_untitled_add_nstoolbaritem) ); type( waitForObject($Names::add_address_nstextfield), $forename ); mouseClick( waitForObject($Names::add_address_nstextfield_2), 62.4609, 14.2227, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_2), $surname ); mouseClick( waitForObject($Names::add_address_nstextfield_3), 60.375, 15.5, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_3), $email ); mouseClick( waitForObject($Names::add_address_nstextfield_4), 108.117, 15.7734, 0, 0 ); type( waitForObject($Names::add_address_nstextfield_4), $phone ); mouseClick( waitForObject($Names::add_address_add_nsbutton) ); $context->{userData}{'forename'} = $forename; $context->{userData}{'surname'} = $surname; });
When("I add a new person '|word|','|word|','|any|','|integer|' to the address book") do |context, forename, surname, email, phone| mouseClick(waitForObject(Names::Address_Book_Untitled_Add_NSToolbarItem)) mouseClick(waitForObject(Names::Add_Address_NSTextField), 57.1602, 4.91016, 0, 0) type(waitForObject(Names::Add_Address_NSTextField), forename) mouseClick(waitForObject(Names::Add_Address_NSTextField_2), 71.3672, 17.5703, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_2), surname) mouseClick(waitForObject(Names::Add_Address_NSTextField_3), 71.3672, 16.5, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_3), email) mouseClick(waitForObject(Names::Add_Address_NSTextField_4), 79.707, 16.7734, 0, 0) type(waitForObject(Names::Add_Address_NSTextField_4), phone) mouseClick(waitForObject(Names::Add_Address_Add_NSButton)) context.userData = Hash.new context.userData[:forename] = forename context.userData[:surname] = surname end
All data stored in context.userData can be accessed from steps implementations and Hooks
in all Scenarios
of the given Feature
. Finally, we need to implement the step, Then previously entered forename and surname shall be at the top
.
@Then("previously entered forename and surname shall be at the top") def step(context): table = waitForObject(names.address_Book_Untitled_NSTableView) children = object.children(table) forename = children[0] surname = children[1] test.compare(forename.stringValue, context.userData["forename"]) test.compare(surname.stringValue, context.userData["surname"])
Then("previously entered forename and surname shall be at the top", function(context) { test.compare(object.children(waitForObjectExists(names.addressBookUntitledNSTableView))[0], "Bob"); test.compare(object.children(waitForObjectExists(names.addressBookUntitledNSTableView))[1], "Doe"); });
Then("previously entered forename and surname shall be at the top", sub { my $context = shift; my $table = waitForObject($Names::address_book_untitled_nstableview); my @children = object::children($table); my $forename = $children[0]; my $surname = $children[1]; test::compare( $forename->stringValue, $context->{userData}{'forename'} ); test::compare( $surname->stringValue, $context->{userData}{'surname'} ); });
Then("previously entered forename and surname shall be at the top") do |context| table = waitForObject(Names::Address_Book_Untitled_NSTableView) children = Squish::Object.children(table) forename = children[0] surname = children[1] Test.compare(forename.stringValue, context.userData[:forename]) Test.compare(surname.stringValue, context.userData[:surname]) end
If you copied the step implementation from here or wrote it yourself instead of recording, you might encounter an error during replay: Object ':Bob_NSString' not found. Object name not found in Object Map.
This happens because we introduced new GUI objects without accessing them through Squish first which would add them to the object map. To resolve the issue, press the Pick New Object
button in the error dialog and then click on the missing object. Now select Retry
to continue replaying.
Scenario Outline
Assume our Feature
contains the following two Scenarios
:
Scenario: State after adding one entry Given the address book application is running When I create a new address book And I add a new person 'John','Doe','john@m.com','500600700' to the address book Then "1" entries should be present Scenario: State after adding one entry Given the address book application is running When I create a new address book And I add a new person 'Bob','Koo','bob@m.com','500600800' to the address book Then "1" entries should be present
As we can see, those Scenarios
perform the same actions using different test data. The same can be achieved by using a Scenario Outline
(a Scenario
template with placeholders) and Examples
(a table with parameters).
Scenario Outline: Adding single entries multiple times Given the address book application is running When I create a new address book And I add a new person '<forename>','<surname>','<email>','<phone>' to the address book Then '1' entries should be present Examples: | forename | surname | email | phone | | John | Doe | john@m.com | 500600700 | | Bob | Koo | bob@m.com | 500600800 |
Please note that the OnScenarioEnd
hook will be executed at the end of each loop iteration in a Scenario Outline
.
Test Execution
In the squishide
, users can execute all Scenarios
in a Feature
, or execute only one selected Scenario
. In order to execute all Scenarios
, the proper test case has to be executed by clicking on the Play button in the Test Suites view.
In order to execute only one Scenario
, you need to open the Feature
file, right-click on the given Scenario
and choose Run Scenario. An alternative approach is to click on the Play button next to the respective Scenario
in the Scenarios tab in Test Case Resources.
After a Scenario
is executed, the Feature
file is colored according to the execution results. More detailed information (like logs) can be found in the Test Results view.
Test Debugging
Squish offers the possibility to pause an execution of a Test Case at any point in order to check script variables, spy application objects or run custom code in the Squish Script Console. To do this, a breakpoint has to be placed before starting the execution, either in the Feature
file at any line containing a step or at any line of executed code (i.e., in the middle of step definition code).
After the breakpoint is reached, you can inspect all application objects and their properties. If a breakpoint is placed at a step definition or a hook is reached, then you can additionally add Verification Points or record code snippets.
Re-using Step Definitions
BDD test maintainability can be increased by reusing step definitions in test cases located in another directory. For more information, see collectStepDefinitions().
Tutorial: Migration of Existing Tests to BDD
This chapter is aimed at users that have existing Squish tests and who would like to introduce Behavior Driven Testing. The first section describes how to keep the existing tests and add new tests with the BDD approach. The second section describes how to convert existing tests to BDD.
Extending Existing Tests to BDD
The first option is to keep any existing Squish script-based tests and extend them by adding new BDD tests. It's possible to have a Test Suite
containing both script-based test cases and BDD tests. Simply open an existing Test Suite
with test cases and choose New BDD Test Case from the drop down list.
Assuming your existing test cases make use of a library and you are calling shared functions to interact with the AUT, those functions can also be used from BDD test cases. In the example below, a function is used in multiple script-based test cases:
def createNewAddressBook():
mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem))
function createNewAddressBook(){
mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem));
}
sub createNewAddressBook{
mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem));
}
def createNewAddressBook
mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem))
end
New BDD test cases can easily use the same function:
@When("I create a new address book")
def step(context):
createNewAddressBook()
When("I create a new address book", function(context){ createNewAddressBook() });
When("I create a new address book", sub { createNewAddressBook(); });
When("I create a new address book") do |context| createNewAddressBook end
Converting Existing Tests to BDD
The second option is to convert an existing Test Suite
that contains script-based tests into behavior driven tests. Since a Test Suite
can contain both, migration can be done gradually. A Test Suite
containing a mix of both test case types can be executed and results analyzed without any extra effort required.
The first step is to review all test cases of the existing Test Suite
and group them by the Feature
they test. Each script-based test case will be transformed into a Scenario
, which is a part of a Feature
. For example, assume we have 5 script-based tests. After review, we realize that they examine two Features
. Therefore, when migration is completed, our Test Suite will contain two BDD feature files, each of them containing one Feature
. Each Feature
will contain multiple Scenarios
. In our example the first Feature
contains three Scenarios
and the second Feature
contains two Scenarios
.
At the beginning, open a Test Suite
in the squishide
that contains script-based tests to be migrated to BDD. Next, create a new test case by choosing New BDD Test Case from the context menu. Each BDD test corresponds to a test.feature
file containing one Feature
. Next, open the test.feature
file to describe the Feature
using the Gherkin language. Following the syntax from the template, edit the Feature
name and optionally provide a short description. Next, analyze which actions and verifications in the test case need to be migrated. This is how an example test case for the address book application may look:
def main(): startApplication("SquishAddressBook") mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem)) test.compare(waitForObjectExists(names.address_Book_Untitled_NSTableView).numberOfRows(), 0)
function main() { startApplication("SquishAddressBook"); mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem)); test.compare((waitForObjectExists(names.addressBookUntitledNSTableView)).numberOfRows(), 0); }
sub main { startApplication("SquishAddressBook"); mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem)); test.compare(waitForObjectExists($Names::address_book_untitled_nstableview).numberOfRows(), 0) }
def main startApplication("SquishAddressBook") mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem)) Test.compare(waitForObjectExists(Names::Address_Book_Untitled_NSTableView).numberOfRows(), 0)
After analyzing the above script-based test we can create the following Scenario
and add it to test.feature
:
Scenario: Initial state of the created address book
Given the address book application is running
When I create a new address book
Then the address book should be empty
Next, right-click on the Scenario
and choose the option Create Missing Step Implementations from the context menu. This will create a skeleton of Steps
definitions:
@Given("the address book application is running") def step(context): test.warning("TODO implement the address book application is running") @When("I create a new address book") def step(context): test.warning("TODO implement I create a new address book") @Then("the address book should be empty") def step(context): test.warning("TODO implement the address book should be empty")
Given("the address book application is running", function(context) { test.warning("TODO implement the address book application is running"); }); When("I create a new address book", function(context) { test.warning("TODO implement I create a new address book"); }); Then("the address book should be empty", function(context) { test.warning("TODO implement the address book should be empty"); });
Given("the address book application is running", sub { my $context = shift; test::warning("TODO implement the address book application is running"); }); When("I create a new address book", sub { my $context = shift; test::warning("TODO implement I create a new address book"); }); Then("the address book should be empty", sub { my $context = shift; test::warning("TODO implement the address book should be empty"); });
Given("the address book application is running") do |context| Test.warning "TODO implement the address book application is running" end When("I create a new address book") do |context| Test.warning "TODO implement I create a new address book" end Then("the address book should be empty") do |context| Test.warning "TODO implement the address book should be empty" end
Now we put code snippets from the script-based test case into respective step definitions and remove the lines containing test.warning
. If your script-based tests make use of shared scripts, you can call those functions inside the step definition as well. For example, the final result could look like this:
@Given("the address book application is running") def step(context): startApplication('"' + os.environ["SQUISH_PREFIX"] + '/examples/mac/addressbook/SquishAddressBook"') test.compare(waitForObjectExists(names.o_NSWindow).isVisible, 0) @When("I create a new address book") def step(context): mouseClick(waitForObject(names.address_Book_Untitled_New_NSToolbarItem)) @Then("the address book should be empty") def step(context): #test.compare(waitForObjectExists(names.address_Book_Untitled_NSTableView).alignment, 0) test.compare(waitForObjectExists(names.address_Book_Untitled_NSTableView).numberOfRows(), 0)
Given("addressbook application is running", function(context) { startApplication('"' + OS.getenv("SQUISH_PREFIX") + '/examples/mac/addressbook/SquishAddressBook"'); test.compare(waitForObjectExists(names.addressBookUntitledNSWindow).isVisible, 1); }); When("I create a new addressbook", function(context) { mouseClick(waitForObject(names.addressBookUntitledNewNSToolbarItem)); }); Then("addressbook should have zero entries", function(context) { //test.compare((waitForObject(names.addressBookUntitledNSTableView)).alignment, 0); test.compare((waitForObject(names.addressBookUntitledNSTableView)).numberOfRows(), 0); });
Given("the address book application is running", sub { my $context = shift; startApplication("\"$ENV{'SQUISH_PREFIX'}/examples/mac/addressbook/SquishAddressBook\""); test::compare(waitForObjectExists($Names::address_book_untitled_nswindow)->isVisible, 1); }); When("I create a new address book", sub { my $context = shift; mouseClick(waitForObject($Names::address_book_untitled_new_nstoolbaritem)); }); Then("the address book should be empty", sub { my $context = shift; #test::compare(waitForObjectExists($Names::address_book_untitled_nstableview)->alignment, 1); test::compare(waitForObjectExists($Names::address_book_untitled_nstableview)->numberOfRows, 0); });
Given("the address book application is running") do |context| startApplication("\"#{ENV['SQUISH_PREFIX']}/examples/mac/addressbook/SquishAddressBook\"") Test.compare(waitForObjectExists(Names::O_NSWindow).isVisible, 0) end When("I create a new address book") do |context| mouseClick(waitForObject(Names::Address_Book_Untitled_New_NSToolbarItem)) end Then("the address book should be empty") do |context| #Test.compare(waitForObjectExists(Names::Address_Book_Untitled_NSTableView).alignment, 0) Test.compare(waitForObjectExists(Names::Address_Book_Untitled_NSTableView).numberOfRows(), 0) end
Note that the test.log("Create a new address book")
got removed while migrating this script-based test to BDD. When the step I create a new address book
is executed, the step name will be logged into Test Results, so the test.log
call would have been redundant.
When the test case execution ends, Squish terminates the AUT. Squish also ensures that the AUT is terminated at the end of each Scenario
. This is done from the auto-generated OnScenarioEnd
hook.
@OnScenarioEnd def hook(context): for ctx in applicationContextList(): ctx.detach()
OnScenarioEnd(function(context) { applicationContextList().forEach(function(ctx) { ctx.detach(); }); });
# Detach (i.e., potentially terminate) all AUTs at the end of a scenario OnScenarioEnd(sub { foreach (applicationContextList()) { $_->detach(); } });
OnScenarioEnd do |context| applicationContextList().each { |ctx| ctx.detach() } end
The above example was simplified for this tutorial. In order to take full advantage of Behavior Driven Testing in Squish, please familiarize yourself with the section Behavior Driven Testing in API Reference.
© 2024 The Qt Company Ltd.
Documentation contributions included herein are the copyrights of
their respective owners.
The documentation provided herein is licensed under the terms of the GNU Free Documentation License version 1.3 as published by the Free Software Foundation.
Qt and respective logos are trademarks of The Qt Company Ltd. in Finland and/or other countries worldwide. All other trademarks are property
of their respective owners.