[Resolved] running testCase from testSuite teardown

I wouldn't do this in a tearDown, but use a Groovy script and the TestMonitor class. I think this would work well for you:

import com.eviware.soapui.monitor.TestMonitor

def maxSecondsToWait = 10 // or whatever value of your choosing
while (TestMonitor.hasRunningTests(testSuite.project) && retries <= maxSecondsToWait) {
try { Thread.sleep(1000) } catch (InterruptedException e) {}
}
// verify!

Nope

Looking at the API here: http://www.soapui.org/apidocs/index.htm ... nitor.html (would be really kool to have some documentation there )

import com.eviware.soapui.monitor.TestMonitor
TestMonitor.hasRunningTests()

gives me

Wed Oct 16 08:16:40 PDT 2013:ERROR:groovy.lang.MissingMethodException: No signature of method: static com.eviware.soapui.monitor.TestMonitor.hasRunningTests() is applicable for argument types: () values: [] Possible solutions: hasRunningTests(), hasRunningTests(com.eviware.soapui.impl.wsdl.WsdlProject), hasRunningTest(com.eviware.soapui.model.testsuite.TestCase), hasRunningTestCase(com.eviware.soapui.model.testsuite.TestCase)

Since a Groovy Step does not have the property testSuite, I tried:

import com.eviware.soapui.monitor.TestMonitor
TestMonitor.hasRunningTests(testRunner.testCase.testSuite.project)

which gives me:

Wed Oct 16 08:18:33 PDT 2013:ERROR:groovy.lang.MissingMethodException: No signature of method: static com.eviware.soapui.monitor.TestMonitor.hasRunningTests() is applicable for argument types: (com.eviware.soapui.impl.wsdl.WsdlProjectPro) values: [com.eviware.soapui.impl.wsdl.WsdlProjectPro@45ea368b] Possible solutions: hasRunningTests(), hasRunningTests(com.eviware.soapui.impl.wsdl.WsdlProject), hasRunningTest(com.eviware.soapui.model.testsuite.TestCase), hasRunningTestCase(com.eviware.soapui.model.testsuite.TestCase)

Does this mean I have to downgrade to the non-Pro version?

The secret make it work switch is:

def tc = new com.eviware.soapui.monitor.TestMonitor
tc.hasRunningTests(...)

I do not think this works.

1. Am I allowed to run this code from the testCase Setup Script? I tried it both from the Setup Script as well as a Groovy testStep.

2. In the while loop I inserted a log.info("I wuz here!"); this message never gets logged! Why not?

3. In the code sample you provided what is "retries"?

4. In the code sample you provided what is the point of the try..catch?

Hi!

Yes, you should be able to run this from the Setup Script as well as a Groovy test step.
Sorry for the code being a bit incomplete. The following code probably works better:

import com.eviware.soapui.monitor.TestMonitor

def monitor = new com.eviware.soapui.monitor.TestMonitor()
def maxSecondsToWait = 10 // or whatever value of your choosing
def retry = 0
while ( monitor.hasRunningTests() && retry < maxSecondsToWait) {
 Thread.sleep(1000)
 retry++
}

To answer your other questions,
2. Probably because "retries" was never initialized and not incremented in the loop
3. It's just a counter that is incremented for each loop
4. try..catch is needed in java because Thread.sleep() throws a checked exception, but it is not needed in groovy.

Hope this helps,
Anders
SmartBear Sweden

Now it starts to get complicated. I have two test cases in a test suite, mode=PARALLEL.

TestCase 1.
1. SOAP call to a service
2. Transfer step to retrieve a value from #1 to a TestSuite property "startingValue". This property has value "empty" at the start of test.

TestCase 2.
1. Groovy Script

import com.eviware.soapui.monitor.TestMonitor

def monitor = new com.eviware.soapui.monitor.TestMonitor()
while(monitor.hasRunningTest("TestCase 1.")) {
       log.info "I wuz here!"
       Thread.sleep(100)
}

2. SOAP call to a service that uses ${#TestSuite#startingValue}

In my last step above, I always see the value "empty" when I look at the Raw Request of the SOAP call, however I see the value updated in the testSuite property when the test is finished running. Also, the info message never gets logged.

I also tried hasRunningTestCase() instead of hasRunningTest(); is there a difference?

I did not use hasRunningTests(), because in my situation as described here that seems like it is going to be asking: am I running while I am running?

What am I doing wrong?

Hi,

It seems as if the TestMonitor solution didn't work as we hoped. We're trying to figure out another solution, and we'll get back to you as soon as possible.

Regards,
Arian
SmartBear Sweden

Hello again,

I've come up with a solution. I need to tell you that this is really outside the scope of the support offering - we generally don't support Groovy scripts. However, I decided to make an exception for you because my initial suggestion was messed up - sorry about that - and also because your problem piqued my interest so I looked at it in my own time. :-)

Solution that seems to work for me:

1. Add a test case first in your test suite, with a single Groovy test step containing the following code:
def monitor = new com.eviware.soapui.monitor.TestMonitor()
def thisTestCase = context.testCase
thisTestCase.testSuite.testCases.each {
testCase = it.value
if (testCase != thisTestCase && testCase.label != "Wait for completion") {
log.info("Monitoring $testCase.label")
monitor.monitorTestCase(testCase)
}
}
context["#TestSuiteRunner#"].runContext.testMonitor = monitor

2. Create another test case at the end of the suite, with the label "Wait for completion" and the following code:
def monitor = context["#TestSuiteRunner#"].runContext.testMonitor
if (monitor == null) {
log.warn("Monitor not found in completion check")
}
else {
def maxWaitTimeInSeconds = 10
def retries = 0
while (monitor.hasRunningTests() && retries < maxWaitTimeInSeconds) {
log.info("Waiting for tests to complete ...")
try {
Thread.sleep(1000)
}
catch (InterruptedException ignored) { }
retries++
}
log.info("Tests completed: " + !monitor.hasRunningTests())
}


... and then obviously replace the final log statements with the checks you want to perform. By doing this you should be able to fail the whole suite properly.

Hope this helps,
Manne

SmartBear Support wrote:
It seems as if the TestMonitor solution didn't work as we hoped.

So does this mean there is a defect, or user error?

I tried your solution - thank you for taking the time.

Apparently there is no runContext; I get the following for both scripts.

Tue Oct 22 09:30:36 PDT 2013:ERROR:java.lang.NullPointerException: Cannot get property 'runContext' on null object

Also, I am still unclear on how all this is suppose to be chained together. I need to end up with running one test case (retrieve and set up some data), followed by several test cases in parallel, followed by one test case (verification).