I'm creating a node extension for KNIME with the KNIME SDK (which is just an adapted Eclipse). I'm done with my code, I tested it in Eclipse, it worked perfectly. So I exported it, copied the .jar to the subfolder dropins in my KNIME installation, and started KNIME. Made a new workflow, added my node, tried to execute it and then the following error appeared:
Execute failed: Support code location could not be determined. Could not convert from URL to URI location.
URL Location: file:/C:/Program Files/KNIME_2.11.2/configuration/org.eclipse.osgi/bundles/528/1/.cp/matlabcontrol-4.1.0.jar
Code Source: (file:/C:/Program Files/KNIME_2.11.2/configuration/org.eclipse.osgi/bundles/528/1/.cp/matlabcontrol-4.1.0.jar <no signer certificates>)
Protection Domain: ProtectionDomain (file:/C:/Program Files/KNIME_2.11.2/configuration/org.eclipse.osgi/bundles/528/1/.cp/matlabcontrol-4.1.0.jar <no signer certificates>)
null
<no principals>
java.security.AllPermissionCollection#6ac1c077 (
("java.security.AllPermission" "<all permissions>" "<all actions>")
)
Class Loader: org.eclipse.osgi.internal.baseadaptor.DefaultClassLoader#b836456
Class Loader Class: class org.eclipse.osgi.internal.baseadaptor.DefaultClassLoader
According to a thread on the KNIME forums, a similar problem also occurs with the community extension "Matlab Snippet", which builds on the same matlabcontrol framework I use (which is also the one mentioned in the error).
I tried evading the problem by including the matlabcontrol as a package in my project (it's open source, so I could just import the sources), then the same error appeared and mentioned the package .jar created by the OSGI system (in my case, matlab.jar, which is the name I chose in my build.properties), so I assume it was just the first one alphabetically and all of the files in the OSGI bundle folder are affected by this problem.
What confuses me is that the problem only occurs once I try to execute the node and connect to MATLAB. Unfortunately, the problem doesn't seem to have been resolved in the KNIME forum thread linked above, apparently an older version of that extension didn't have the issue and users just installed the old version.
So, can I somehow make the URL correctly convert to URI? Is the Protection Domain the cause and can I change it? Does it have anything to do with the default Class Loader?
I have now installed the plugin via a feature in a folder (see comments). The error persists, but the path changed:
Execute failed: Support code location could not be determined. Could not convert from URL to URI location.
URL Location: file:/C:/Program Files/KNIME_2.11.2/plugins/org.knime.ext.matlab_1.0.0/matlab.jar
Code Source: (file:/C:/Program Files/KNIME_2.11.2/plugins/org.knime.ext.matlab_1.0.0/matlab.jar <no signer certificates>)
Protection Domain: ProtectionDomain (file:/C:/Program Files/KNIME_2.11.2/plugins/org.knime.ext.matlab_1.0.0/matlab.jar <no signer certificates>)
null
<no principals>
java.security.AllPermissionCollection#2da869b5 (
("java.security.AllPermission" "<all permissions>" "<all actions>")
)
Class Loader: org.eclipse.osgi.internal.baseadaptor.DefaultClassLoader#372bf03c
Class Loader Class: class org.eclipse.osgi.internal.baseadaptor.DefaultClassLoader
Related
A related problem: Get the class path from the context class loader (of WebLogic for instance)
This is already solved in How to set up the context class loader's classpath for runtime compilation?.
Current problem: Get the same (proper) class path used to run a web app
Reflectively calling the "getClassPath()" method works but it returns a dirty classpath containing unanted modules from $ORACLE_HOME/oracle_common/modules directory.
Problematic scenario:
Deploy a web app "Parent" in WebLogic // <- works
Get the context class path (it's a String object) by reflexively calling "getClassPath()" // <- works
Fork a process out of the main "Parent" process and run it using the context class path // <- fails
04:36:45,238 [Thread-41] ERROR ChildProcess - Exception in thread "main" java.lang.NoSuchMethodError: com.fasterxml.jackson.databind.ObjectMapp
er.configOverride(Ljava/lang/Class;)Lcom/fasterxml/jackson/databind/cfg/MutableConfigOverride;
Explanation
While the context class path contains the necessary dependencies for the child process to run, they are overshadowed by WebLogic's own dependencies. The result is runtime failures such as the one shown above.
Workaround
a) Use a new version of WebLogic server that hopefully would use newer versions of the artifacts needed by the child process // risky endeavour
b) Manually process the context class path and remove any artifact that would shadow their more-recent counterparts
Solution b) looks more practical, but I don't like it for meany reasons:
The reflexive call to "getClassPath" returns a String, and looking for artifact's names in Strings feels frail and weak
I wouldn't know what shadows what. WebLogic prepends its weird artifacts at the start of the string before listing the web app's own dependencies.
Only weblogic.xml has info on the web app's package preferences. I wish I could mimic how WebLogic processes this file to run the web app (Parent) and use that to properly run the child process
It seems to me that forking a process out from a web app running in WebLogic does not enjoy the same package preferences expressed in "weblogic.xml" that the web app (Parent) enjoyed when it was deployed and started running.
Other than the above suggestions, I am welcoming of any stronger solutions
I changed the ivy.jar.url to the above mentioned url and now it came up with a new error as follows:
[Console output redirected to
file:C:\Rahul13615\liferay.metadata.plugins\com.liferay.ide.sdk.core\sdk.log]
Buildfile: C:\Rahul13615\liferay\plugins\portlets\build.xml
[get] Getting: https://repository.liferay.com/nexus/content/repositories/liferay-public-snapshots/com/liferay/org.apache.ivy/2.4.0.LIFERAY-PATCHED-1-SNAPSHOT/org.apache.ivy-2.4.0.LIFERAY-PATCHED-1-SNAPSHOT.jar
[get] To: C:\Rahul13615\liferay\plugins.ivy\org.apache.ivy-2.4.0.LIFERAY-PATCHED-1-SNAPSHOT.jar
[taskdef] Could not load definitions from resource org/apache/ivy/ant/antlib.xml. It could not be found.
BUILD FAILED
C:\Rahul13615\liferay\plugins\portlets\build.xml:5: The following error occurred while executing this line:
C:\Rahul13615\liferay\plugins\build-common-plugins.xml:5: The following error occurred while executing this line:
C:\Rahul13615\liferay\plugins\build-common.xml:47: The following error occurred while executing this line:
C:\Rahul13615\liferay\plugins\build-common-ivy.xml:102: The following error occurred while executing this line:
C:\Rahul13615\liferay\plugins\build-common-ivy.xml:38: The following error occurred while executing this line:
C:\Rahul13615\liferay\plugins\build-common-ivy.xml:48: Problem: failed to create task or type antlib:org.apache.ivy.ant:settings
Cause: The name is undefined.
Action: Check the spelling.
Action: Check that any custom tasks/types have been declared.
Action: Check that any / declarations have taken place.
No types or tasks have been defined in this namespace yet
This appears to be an antlib declaration.
Action: Check that the implementing library exists in one of:
-ANT_HOME\lib
-the IDE Ant configuration dialogs
Total time: 11 seconds
Below please find my steps how I created a new Liferay Plugin Project.
I hope this will be useful to you:
I have downloaded Liferay Developer Studio 3.0.1. GA2 from:
htt ps://web.liferay.com/downloads/liferay-projects/liferay-ide or
ht tps://sourceforge.net/projects/lportal/files/Liferay%20IDE/
Note: there are some known issue with the 3.0.0 version
I extracted the IDE to C:\dstudio3 on my computer
I downloaded the Liferay SDK from:
http s://www.liferay.com/downloads?559250categoryIds=559255
(bottom of the page, additional files section)
I copied the SDK to the IDE's folder and created a personalized build.peter.properties file, based on:
htt ps://web.liferay.com/community/forums/-/message_boards/message/76813485
I started Eclipse and set up a new Liferay Server environment
In Eclipse, I created a new "Liferay Plugin Project", and in the wizard, I defined the location of the Plugins SDK.
I created a new Liferay Portlet project inside the Plugin project.
Best Regards,
Peter
Scanning through the WSO2 documentation and several blogs, I found three different type of registry mechanisms to add new claims to a Java PIP module:
By use of setup file calledentitlement.properties and by adding of Java static properties for each variable as described in the
sample case from the WSO2 manual, see: Writing a Custom Policy Info Point;
By use of supportedAttributesIds.add as described in link How to write a PIP point for WSO2 IS;
By use of a setter method: public Set getSupportedIds() { Set<String> ids = new HashSet<String>(); ids.add("http://kmarket.com/id/role"); return ids; } as described in the Stackoverflow post How To Add User Defined Attribute in PIP Attribute User Store
Please instruct which of the three options is the updated and recommended solution for adding new fields produced by the lookup of external data sources executed by the Balana PIP as an extension of the "AttributeFinderModule" class.
I've also tested the code posted in this example and compiled it with Java 6 JDK, see source reference
How To Add User Defined Attribute in PIP Attribute User Store
After the restart, the WSO2 server continues to display the following error:
Error while loading entitlement policies. Exception occurred while trying to invoke service method getAllPolicies
The following error details are available. Please refer logs for more details.
org.apache.axis2.AxisFault: Exception occurred while trying to invoke service method getAllPolicies
at org.wso2.carbon.identity.entitlement.ui.client.EntitlementPolicyAdminServiceClient.handleException(EntitlementPolicyAdminServiceClient.java:478)
at org.wso2.carbon.identity.entitlement.ui.client.EntitlementPolicyAdminServiceClient.getAllPolicies(EntitlementPolicyAdminServiceClient.java:81)
at org.apache.jsp.entitlement.index_jsp._jspService(org.apache.jsp.entitlement.index_jsp:183)
when adding the extended AttributeFinder Java extension. Is this a known bug in the WSO2 server?
The first way is configurable, you don't need to rebuild and deploy the PIP when you want to add new attibute. You just have to add them to entitlement.properties file. and restart the IS.
In other two ways, you have to rebuild the PIP module for each attribute changes.
But, If you are going to add and attribute, you have to implement the logic for that too, so in that case going for last two ways also make sense.
I've replaced the KmarketPIPAttributeFinder program module by the latest version from the svn source repository (see link below) and compiled it. This solved the problem under WSO2 Identity Server Version 5 and the routine started to register the PIP attributes, now visible at the PDP extensions.
Download the latest code from: https://svn.wso2.org/repos/wso2/carbon/platform/trunk/components/identity/org.wso2.carbon.identity.samples.entitlement.pip/src/main/java/org/wso2/carbon/identity/samples/entitlement/pip/KmarketPIPAttributeFinder.java
I am new to Jython and Python, trying to build a prototype that makes use of Python code to be called from within Java. The code I am developing works in Jetty and in standalone mode (running java -jar from the command line), but not when deployed to weblogic.
How can I make weblogic(10.3.5) server/Jython recognize the Lib folder within jython-standalone-2.5.4-rc1.jar?
My Java code uses the JythonObjectFactory to invoke python modules as outlined in the Jython book:
http://www.jython.org/jythonbook/en/1.0/JythonAndJavaIntegration.html
The Python modules are using external libraries like csv, logging etc. that are not packaged with jython.jar, hence I am using jython-standalone jar.
The java code includes an interface that would define the class type of the first invoked py module from within java. The interface and the input and output (to python modules) type classes are in a package structure as com.abc.xpackage. and the py modules exist at the root of this package. A controller layer calls the objectfactory and in turn executes the python code thus:
JythonObjectFactory calFactory = new JythonObjectFactory(CalcType.class, "Calculate", "Calculate");
CalcType engine = (CalcType)calFactory.createObject();
output = engine.execute(input);
The entire code is bundled as a jar file which would become part of a web application deployed on weblogic. The code was compiled with maven (with jython dependencies included in the repository) and runs fine on the included Jetty runtime within eclipse.
When deployed on weblogic, however, I get a "ImportError: no module named csv" error.
To analyze what is happening, I tried printing the Jython system state path on weblogic and the standalone environment/Jetty. What I found is,
on Jetty, the system path consists of the following:
C:\.m2\repo\org\python\jython\jython-standalone-2.5.3-rc1.jar\Lib, ____classpath__, ____pyclasspath__
on Weblogic, printing the system path by default shows the following:
____classpath__, ____pyclasspath__
I tried forcing the inclusion of the missing path using the code as follows:
public JythonObjectFactory(PySystemState state, Class interfaceType, String moduleName, String className) {
String pathToAppend = new File(state.getClass().getProtectionDomain().getCodeSource().getLocation().toURI().getPath()).getAbsolutePath()+"\\Lib";
state.path.insert(0, new PyString(pathToAppend));
state.path.append(new PyString(pathToAppend));
System.out.println("Jython sys path: "+state.path);
Please note, I prepended as well as appended the path in different trials. The sys path on weblogic now displays the following:
Jython sys path: ['C:\\wldomain\\wls135\\servers\\cgServer\\tmp\\app-1\\war\\WEB-INF\\lib\\jython-standalone-2.5.4-rc1.jar\\Lib', '__classpath__', '__pyclasspath__/', 'C:\\wldomain\\wls135\\servers\\cgServer\\tmp\\app-1\\war\\WEB-INF\\lib\\jython-standalone-2.5.4-rc1.jar\\Lib']
I am still getting ImportError despite this forcing of sys path. Please help why this works in a local environment, and not on weblogic, and if there is any configuration I am missing. Apologize for the rambling long post, I did not know how to explain the problem better. I will try and include any code/artifacts as needed.
Based on a comment(by Lassi) on the blog post below:
http://www.petervannes.nl/files/e1c3c56d15d25dcfd4adb5397a9ef71e-53.php
The jython issue was resolved after explicitly adding the Lib folder python.path to the weblogic startup script as a JAVA_OPTION.
In my case I added the exploded Lib folder to the domain server lib, but based on my test this works also from within the jython jar. Both the following JAVA_OPTIONS worked:
-Dpython.path=C:\wldomain\wls135\lib\Lib
-Dpython.path=C:\wldomain\wls135\lib\jython-standalone-2.5.4-rc1.jar\Lib
The programmatic way of sys.path.append worked for the local environment(jetty) but did not seem to work for weblogic.
I am trying to setup some boiler plate code in my local workspace which uses weblogic10.3. I have a documentation that asks me to import Content.jar from /wlportal_10.3/content-mgmt/lib into window->preferences-->Weblogic-->Shared Library which I did. However when I try to deploy portlet this is the error am getting
Error Deployer BEA-149107 An attempt was made to deploy library 'content' with specified specification version '10.3.2' and implementation version '10.3.2 '. However, the library manifest has specification version '10.3.2' and implementation version '10.3.2'