HTTPConnector

[Jim Fox]

I have some more info on this double execution.  Apologize for the longish post.

Spoiler alert: If I run the script from a ScriptFile it works correctly.

Note that I am using Java 8 and nashorn.

This is the script.
=============================================================
   <Script>
    <![CDATA[
     var Excp = Java.type("java.lang.Exception");
     logger = Java.type("org.slf4j.LoggerFactory").getLogger("net.shibboleth.idp.attribute");
     logger.info("=== gws groups - in");
     new Excp().printStackTrace();
     var HashSet = Java.type("java.util.HashSet");
     var HttpClientSupport = Java.type("net.shibboleth.utilities.java.support.httpclient.HttpClientSupport");
     var IdPAttribute = Java.type("net.shibboleth.idp.attribute.IdPAttribute");
     var StringAttributeValue = Java.type("net.shibboleth.idp.attribute.StringAttributeValue");

     logger.info("=== gws groups - 2");

     var body = HttpClientSupport.toString(response.getEntity(), "UTF-8", 65536);
     var result = JSON.parse(body);

     var attr = new IdPAttribute("memberOf");
     var values = new HashSet();
     if (result.data != null) {
         for (var i=0; i<result.data.length; i++) {
             values.add(new StringAttributeValue(result.data[i].id));
         }
     }
     attr.setValues(values);
     connectorResults.add(attr);
     logger.info("=== gws groups - out");
    ]]>
   </Script>
=============================================================

Note the printStackTrace at line 4.  When I run this I get two stack traces.
The first starts:
java.lang.Exception
    at jdk.nashorn.internal.scripts.Script$26$\^eval\_.:program(<eval>:4)
    at jdk.nashorn.internal.runtime.ScriptFunctionData.invoke(ScriptFunctionData.java:640)
    ...
The trace comes from line 4.

The second trace starts:
java.lang.Exception
    at jdk.nashorn.internal.scripts.Script$26$\^eval\_.:program(<eval>:31)
    at jdk.nashorn.internal.runtime.ScriptFunctionData.invoke(ScriptFunctionData.java:640)
    ...
The trace comes from line 31.  That's past the end of the script.  One gets the impression that the script was not run twice, but compiled twice.


As I said above, when I move the script to a ScriptFile it works correctly.


Jim