I successfully implemented several tests within TestNG framework, where parameters are being read from xml file.
Here is the example block that is executed as first:
#Parameters({ "country" })
#BeforeSuite(alwaysRun = true)
public void prepareRequest(String country, ITestContext cnt) {
LoginInfoRequestParm loginParms = new LoginInfoRequestParm(country);
Headers reqHeaders = new Headers();
reqHeaders.setHeaders(loginParms);
}
The problem/question is, why does it work only if the ITestContext is specified? Once it is removed, the overall suite is broken and it will never come to the specified method prepareRequest(). I was not able to debug it, because I cannt set breakpoint before the method to be able to see what is going on in TestNG itself.
Thank you for your explanation.
To get out of this situation, try something like this
String myPar = context.getCurrentXmlTest().getParameter("country");
if (myPar == null) {
myPar = "INDIA";
}
now myPar can be used, only thing here is if you run class for debug or any other purpose then we are using INDIA. if we run from testng.xml file then it will take values from that file.
Related
My requirement is as follows:
I have a couple of .feature files. I want to create test data that would be common to all of these feature files. Once the test data is created the scenarios will be run from the feature files.
I also want some info back after the test data is created. eg., ids of the data that i have created. So i can use this info to call the api's, add in payloads in my scenarios.
I think we could do this by:
1. Creating a junit java file. I define a static method with #BeforeClass in there and use Karate's runner() to run my create-test-data.feature file (I can use Karate to hit application api to create some data). I define a property in my java class of type Object and set it with the result of Runner.runFeature().
Then I create a separate feature file test-data-details.feature. I define my Java Interop code here. eg.,
def test_data =
"""
var JavaOutput = Java.type('com.mycompany.JavaFile');
var testData = JavaOutput.propertyName;
"""
Now that I have my test data object in my test-data-details.feature file. I call this .feature file (callonce) in the Background section of my feature files that have test scenarios in. So I can retries the test data details like id, name. etc that I can then use in the api request paths and payloads.
I am not sure if the above design is the correct way to go ahead. I tried but getting some issues in my Java file where getClass() below complains that it cannot be used in static method.
#RunWith(Karate.class)
public class AccountRunner {
public static Object job = null;
#BeforeClass
public static void create_job(){
Map<String, Object> result = Runner.runFeature(getClass(), "test-data.feature", null, true);
job = result.get("job");
}
}
Now all of the above can be totally wrong. Need help on how to tackle this scenario in Karate.
Thanks
From your question I understand you have a common test data feature file, which you want to run before all the test and hold that response in a variable that can be used in all of the test features.
You can also achieve this in karate-config.js using karate.callSingle()
In your karate-config.js
config["testdata"] = karate.callSingle("test-data.feature")
Your test-data.feature will be executed once before all the tests and store the response in testdata you can use this variable directly in your feature.
So i have implemented the following design:
I have created two methods one with BeforeClass and other with AfterClass annotation in my TestRunner.java file. In these methods I am able to call the specific data creation and clean-up feature files and pass args as Json object.
#RunWith(Karate.class)
#KarateOptions(tags = {"~#ignore"})
public class AccountRunner {
public static Map<String, Object> result = null;
#BeforeClass
public static void create_job() throws IOException, ParseException {
Class clazz = AccountRunner.class;
URL file_loc = clazz.getResource("create-test-data-1.json");
File file = new File(file_loc.getFile());
JSONParser parser = new JSONParser();
Object obj = parser.parse(new FileReader(file));
JSONObject jsonObject = (JSONObject) obj;
Map<String, Object> args = new HashMap();
args.put("payload", jsonObject);
result = Runner.runFeature(CommonFeatures.class, "create-data.feature", args, true);
}
#AfterClass
public static void delete_investigation() {
Map<String, Object> args = new HashMap();
args.put("payload", result);
Runner.runFeature(CommonFeatures.class, "delete-job.feature", args, true);
}
}
To run these tests via command line using "mvn test" command, i have done following changes in pom.xml.
`<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-surefire-plugin</artifactId>
<version>3.0.0-M3</version>
<configuration>
<includes>
<include>**/*Runner.java</include>
</includes>
</configuration>
</plugin>`
With this solution I able to run my tests in IDE by executing the runner directly or by command line. However, I have not found a way to run all of my tests by following the karate suggested approach where I have a *Test.java file at test suite level and I use default maven configuration with "mvn test". The features fails to run as the .feature file is called before the Runner file is executed which has method to create test data for tests.
Maybe someone can suggest what else, I could do to use Karate approach of running *Test.java file instead of each *Runner.java file.
I'm trying to create a IntelliJ plugin (mostly for learning purposes). My aim is that by pressing a keyboard shortcut the plugin will generate a corresponding PHP unit test method stub in the test file.
So let's say Db.php is open, the upon pressing Ctrl+Shift+U the plugin will create a unit test stub in DbTest.php.
So far I've figured out how to get the method name at cursor and how to locate the corresponding Unit test file (i.e. Db => DbTest) as PsiFile.
PsiFile[] search = FilenameIndex.getFilesByName(project, testFileName, scope); //scope is the test directory
PsiFile testFile = search[0];
What I cannot figure out is how to insert the generated new method stub this in testFile and then save the changes?
P.S. I see there exists a createMethodFromText function but how do I get the PsiClass from PsiFile? Also how do I save the changes?
There're just a few simple steps.
Find PhpClass you want to insert a new method in. As you already have PsiFile you can either traverse a tree manually or use PhpElementVisitor.
1.1. To travers a tree manually you can use PsiTreeUtil#findChildOfType method. In your case you'll need to find GroupStatement first, then the class you need.
1.2. Invoke PsiElement#accept method (PsiFile is an instance of PsiElement) provided with PhpElementVisitor with overridden #visitPhpGroupStatement and #visitPhpClass methods.
Use PhpPsiElementFactory#createMethod to create the new method from text. Note that this class isn't a part of the public API, so theoretically it can be easily changed/moved/removed/whatever in the future.
Use PsiElement#add (PhpClass is also an instance of PsiElement) to insert the method into the class.
That's all. You don't need to explicitly save the changes.
Here is what worked for me in the end. Thanks everyone for the help
for (int i = 0; i < found.getTextLength(); i++) {
PsiElement ele = found.findElementAt(i);
PhpClass phpClass = PsiTreeUtil.getParentOfType(ele, PhpClass.class);
if (phpClass != null) {
Method methodExists = findMethod(phpClass, methodName);
if (methodExists == null) {
new WriteCommandAction.Simple(phpClass.getProject(), phpClass.getContainingFile()) {
#Override
protected void run() throws Throwable {
PsiElement brace = phpClass.getLastChild();
if (brace != null) {
Method method = PhpPsiElementFactory.createMethod(phpClass.getProject(), "public function " + methodName + "() {\n\n}");
CodeStyleManager styleManager = CodeStyleManager.getInstance(getProject());
styleManager.reformat(method);
PsiElement newMethod = phpClass.addBefore(method, brace);
PsiNavigateUtil.navigate(newMethod);
}
}
}.execute();
} else {
PsiNavigateUtil.navigate(methodExists);
}
break;
}
}
I make use of a message bundle in one of my services in a Grails 2.0 project for internationalized text. The use case is an email subject that is sent via the mail plugin in an asynchronous way, so it really doesn't make sense to have this in a controller or TagLib (given the usual argument of not accessing your text or views in a service). This code works fine in my running Grails app, but I'm not sure how to test it.
I tried a PluginAwareResourceBundleMessageSource in my defineBeans as that is what my running application injects, but it led to nullpointers as it appears it needs a bunch of setup around plugin managers and such that my test environment is not giving (even integration).
I then tried a ReloadableResourceBundleMessageSource as it was pure Spring, but it can't seem to see my .properties files, and fails with a No message found under code 'my.email.subject' for locale 'en'.
I feel like I'm going down a wormhole a bit as accessing Grails i18n in a service is not documented in the grails docs, so if there is a preferred way to do this, let me know.
Note my .properties file is in the standard grails-app/i18n location.
The test
#TestFor(EmailHelperService)
class EmailHelperServiceTests {
void testSubjectsDefaultLocale() {
defineBeans {
//messageSource(PluginAwareResourceBundleMessageSource); Leads to nullpointers
messageSource(ReloadableResourceBundleMessageSource);
}
String expected = "My Expected subject Passed1 Passed2";
String actual = service.getEmailSubjectForStandardMustGiveGiftFromBusiness(Locale.ENGLISH, Passed1 Passed2);
assertEquals("email subject", expected, actual);
}
Service:
class EmailHelperService {
def messageSource;
public String getEmailSubject(Locale locale, String param1, String param2) {
Object[] params = [param1, param2].toArray();
return messageSource.getMessage("my.email.subject", params, locale );
}
There is already a messageSource in unit tests in Grails, it is a StaticMessageSource (see http://static.springsource.org/spring/docs/2.5.4/api/org/springframework/context/support/StaticMessageSource.html), you can add mock messages with the addMessage method:
messageSource.addMessage("foo.bar", request.locale, "My Message")
In unit tests and the local side of functional tests, sometimes you want the real properties that are in the 18n directory.
This works for me:
MessageSource getI18n() {
// assuming the test cwd is the project dir (where application.properties is)
URL url = new File('grails-app/i18n').toURI().toURL()
def messageSource = new ResourceBundleMessageSource()
messageSource.bundleClassLoader = new URLClassLoader(url)
messageSource.basename = 'messages'
messageSource
}
i18n.getMessage(key, params, locale)
In a unit test you could ensure that you're wired up correctly by doing something like this:
void testSubjectsDefaultLocale() {
def messageSource = new Object()
messageSource.metaClass.getMessage = {subject, params, locale ->
assert "my.email.subject" == subject
assert ["Passed1", "Passed2"] == params
assert Locale.ENGLISH == locale
"It Worked!!!"
}
service.messageSource = messageSource
String actual = service.getEmailSubjectForStandardMustGiveGiftFromBusiness(Locale.ENGLISH, Passed1 Passed2)
assert "It Worked!!!" == actual
}
This will help ensure that you're wired up correctly but it will not ensure that what you're doing actually works. If you're comfortable with that then this would work for you. If you're trying to test that when you give "XYZ" to your .properties file it returns "Hello" then this will not work for you.
I just started a new job and one of the first things I've been asked to do is build unit tests for the code base (the company I now work for is committed to automated testing but they do mostly integration tests and the build takes forever to complete).
So everything started nicely, I started to break dependencies here and there and started writing isolated unit tests but now I'm having an issue with rhino mocks not being able to handle the following situation:
//authenticationSessionManager is injected through the constructor.
var authSession = authenticationSessionManager.GetSession(new Guid(authentication.SessionId));
((IExpirableSessionContext)authSession).InvalidateEnabled = false;
The type that the GetSession method returns is SessionContext and as you can see it gets casted into the IExpirableSessionContext interface.
There is also an ExpirableSessionContext object that inherits from SessionContext and implements the IExpirableSessionContext interface.
The way the session object is stored and retrieved is shown in the following snippet:
private readonly Dictionary<Guid, SessionContext<TContent>> Sessions= new Dictionary<Guid, SessionContext<TContent>>();
public override SessionContext<TContent> GetSession(Guid sessionId)
{
var session = base.GetSession(sessionId);
if (session != null)
{
((IExpirableSessionContext)session).ResetTimeout();
}
return session;
}
public override SessionContext<TContent> CreateSession(TContent content)
{
var session = new ExpirableSessionContext<TContent>(content, SessionTimeoutMilliseconds, new TimerCallback(InvalidateSession));
Sessions.Add(session.Id, session);
return session;
}
Now my problem is when I mock the call to GetSession, even though I'm telling rhino mocks to return an ExpirableSessionContext<...> object, the test throws an exception on the line where it's being casted into the IExpirableSession interface, here is the code in my test (I know I'm using the old syntax, please bear with me on this one):
Mocks = new MockRepository();
IAuthenticationSessionManager AuthenticationSessionMock;
AuthenticationSessionMock = Mocks.DynamicMock<IAuthenticationSessionManager>();
var stationAgentManager = new StationAgentManager(AuthenticationSessionMock);
var authenticationSession = new ExpirableSessionContext<AuthenticationSessionContent>(new AuthenticationSessionContent(AnyUserName, AnyPassword), 1, null);
using (Mocks.Record())
{
Expect.Call(AuthenticationSessionMock.GetSession(Guid.NewGuid())).IgnoreArguments().Return(authenticationSession);
}
using (Mocks.Playback())
{
var result = stationAgentManager.StartDeploymentSession(anyAuthenticationCookie);
Assert.IsFalse(((IExpirableSessionContext)authenticationSession).InvalidateEnabled);
}
I think it makes sense the cast fails since the method returns a different kind of object and the production code works since the session is being created as the correct type and stored in a dictionary which is code the test will never run since it is being mocked.
How can I set this test up to run correctly?
Thank you for any help you can provide.
Turns out everything is working fine, the problem was that on the setup for each test there is an expectation on that method call:
Expect.Call(AuthenticationSessionMock.GetSession(anySession.Id)).Return(anySession).Repeat.Any();
So this expectation was overriding the one I set on my own test. I had to take this expectation out of the setup method, include it on a helper method and have all the other tests use this one instead.
Once out of the way, my test started working.
In my app, I use Groovy as a scripting language. To make things easier for my customers, I have a global scope where I define helper classes and constants.
Currently, I need to run the script (which builds the global scope) every time a user script is executed:
context = setupGroovy();
runScript( context, "global.groovy" ); // Can I avoid doing this step every time?
runScript( context, "user.groovy" );
Is there a way to setup this global scope once and just tell the embedded script interpreter: "Look here if you can't find a variable"? That way, I could run the global script once.
Note: Security is not an issue here but if you know a way to make sure the user can't modify the global scope, that's an additional plus.
Shamelessly stolen from groovy.codehaus :
The most complete solution for people
who want to embed groovy scripts into
their servers and have them reloaded
on modification is the
GroovyScriptEngine. You initialize the
GroovyScriptEngine with a set of
CLASSPATH like roots that can be URLs
or directory names. You can then
execute any Groovy script within those
roots. The GSE will also track
dependencies between scripts so that
if any dependent script is modified
the whole tree will be recompiled and
reloaded.
Additionally, each time you run a
script you can pass in a Binding that
contains properties that the script
can access. Any properties set in the
script will also be available in that
binding after the script has run. Here
is a simple example:
/my/groovy/script/path/hello.groovy:
output = "Hello, ${input}!"
import groovy.lang.Binding;
import groovy.util.GroovyScriptEngine;
String[] roots = new String[] { "/my/groovy/script/path" };
GroovyScriptEngine gse = new GroovyScriptEngine(roots);
Binding binding = new Binding();
binding.setVariable("input", "world");
gse.run("hello.groovy", binding);
System.out.println(binding.getVariable("output"));
This will print "Hello, world!".
Found: here
Would something like that work for you?
A simple solution is to use the code from groovy.lang.GroovyShell: You can precompile the script like so:
GroovyCodeSource gcs = AccessController.doPrivileged( new PrivilegedAction<GroovyCodeSource>() {
public GroovyCodeSource run() {
return new GroovyCodeSource( scriptCode, fileName, GroovyShell.DEFAULT_CODE_BASE );
}
} );
GroovyClassLoader loader = AccessController.doPrivileged( new PrivilegedAction<GroovyClassLoader>() {
public GroovyClassLoader run() {
return new GroovyClassLoader( parentLoader, CompilerConfiguration.DEFAULT );
}
} );
Class<?> scriptClass = loader.parseClass( gcs, false );
That's was the expensive part. Now use InvokeHelper to bind the compiled code to a context (with global variables) and run it:
Binding context = new javax.script.Binding();
Script script = InvokerHelper.createScript(scriptClass, context);
script.run();