Test passes but an exception recorded in logs

[andyredhead]

Hi,

This is another newbie question...

I've put together a very simple app which I'm using to get started with jsfunit (based on Seam 2.2.0.GA, RichFaces 3.3.2.SR1 and Facelets).

I'm using the snapshot version of jsfunit which Stan posted on this forum a few days back and I've got RichFaces set up to use the Tidy filter.

The test case is very simple - it basically uses a setup method to create a JSFSession based on "/app/hello.html" and then create a JSFClient. There is a single test method which just checks that the JSFClient is not null (the test page and test class are attached).

The test case passes but there is an exception recorded in the log:

2010-02-23 21:50:12,464 ERROR [com.gargoylesoftware.htmlunit.javascript.StrictErrorReporter] [JavaScript Job Thread 1] [] runtimeError: message=[Exception: java.lang.IllegalStateException: No active event context] sourceName=[http://localhost:8080/wk/nl/tracker/app/a4j/g/3_3_2.SR1/org/ajax4jsf/framework.pack.js] line=[5746] lineSource=[null] lineOffset=[0]
2010-02-23 21:50:12,465 ERROR [com.gargoylesoftware.htmlunit.javascript.background.JavaScriptJobManagerImpl] [JavaScript Job Thread 1] [] Job run failed with unexpected RuntimeException: Exception: java.lang.IllegalStateException: No active event context (http://localhost:8080/wk/nl/tracker/app/a4j/g/3_3_2.SR1/org/ajax4jsf/framework.pack.js#5746)
======= EXCEPTION START ========
Exception class=[net.sourceforge.htmlunit.corejs.javascript.EvaluatorException]
com.gargoylesoftware.htmlunit.ScriptException: Exception: java.lang.IllegalStateException: No active event context (http://localhost:8080/wk/nl/tracker/app/a4j/g/3_3_2.SR1/org/ajax4jsf/framework.pack.js#5746)
        at com.gargoylesoftware.htmlunit.javascript.JavaScriptEngine$HtmlUnitContextAction.run(JavaScriptEngine.java:527)

The full log is attached.

What are the implications of this error? Can I ignore it (after all, no exception was actually thrown by the setup method) or does this indicate that something needs to be addressed (and if so, where do I start! )

Thanks, Andy

[andyredhead]

Doh, I should have said that my test app is using HtmlUnit 2.7 and Neko 1.9.14.

[ssilvert]

Hi Andrew,

You have:

_jsfSession = new JSFSession("/app/hello.html");

The argument to the JSFSession constructor must be a URL that maps to the FacesServlet.  Instead, it should be something like:

_jsfSession = new JSFSession("/app/hello.jsf");

Stan

[andyredhead]

Hi,

I map .html to the facelets servlet in web.xml

    <servlet>
        <servlet-name>Faces Servlet</servlet-name>
        <servlet-class>javax.faces.webapp.FacesServlet</servlet-class>
        <load-on-startup>1</load-on-startup>
    </servlet>

    <servlet-mapping>
        <servlet-name>Faces Servlet</servlet-name>
        <!--url-pattern>*.seam</url-pattern-->
        <url-pattern>*.html</url-pattern>
    </servlet-mapping>

(perhaps I should have attached web.xml as well?)

So I don't think thats the cause...

Cheers,

Andy

[ssilvert]

OK, I'm feeling dumb.  I should have recognized the error from Upgrade HtmlUnit to 2.7?.  I don't think it will cause any problems but I do catch it and ignore in the latest JSFUnit code.  The JSFUnit 1.2 release should be out in a few days but if you want to get the fix right now you can use the latest from the snapshot repo:

http://snapshots.jboss.org/maven2/org/jboss/jsfunit/jboss-jsfunit-core/1.2.0.GA-SNAPSHOT/

Stan

[andyredhead]

Hi,

I've just had a quick go with the snapshot releases:

1) jboss-jsfunit-core-1.2.0.GA-20100217.183530-5.jar

2) jboss-jsfunit-richfaces-1.2.0.GA-20100210.172716-4.jar

I'm still seeing (what looks to me like) the same error (log attached).

If you are interested in the test, I've attached the app (it should be self contained). The test is configured to be on:

http://localhost:8080/wk/nl/tracker/ServletTestRunner?suite=com.innodata.apps.proctracker.jsfunit.doctask.TestDocsInRoleWk&xsl=cactus-report.xsl

Cheers, Andy

[andyredhead]

Now I can see I'm being dim (the benefit of a good nights sleep) - the error is still being recorded but its being treated as having no impact.