Unity+IKVM / NullReferenceException when calling Java method in dll - java

I'm developing an Unity(Unity3D 5.3.5f1) project with using Java library converted into dll with IKVM 8.1 RC0 (Additionally using VS for c# development)
I converted several related jars into one dll file. There was no warning or error in converting sequence. And I put them in Assets folder in Unity project.
I tried to run my java code in C#.
This is my C# code where executes java logics (Actually shorten a lot)
using UnityEngine;
using com.mypackage.hierarchy1;
using com.mypackage.hierarchy2; // Same as the package name from Java
public class Test : MonoBehaviour {
...... // several variables
private Class1 var1;
private Class2 var2;
// note that, Class1 and Class2 are the same name used in Java
void Start() {
var1 = new Class1();
var2 = new Class2();
}
void Update() {
method1(certain_param_in_int);
}
......
void method1(int param) {
method2(param, param2, param3,...);
}
void method2(int param, string param2, int param3, int param4) {
var1.method_v1(param, param2,...); // Works well with no problem
var2.initialize("str", var1.getResult(), "anotherstr");
var2.run(); // This method occurs the problem
}
}
The problem is that when var2.run() is executed. method from var1 has no problem. I checked the result was correct. Anyway, var2.run() shows NullReferenceException as below
NullReferenceException: Object reference not set to an instance of an object
com.mypackage.hierarchy2.Class2.run ()
Test.method2 (Int32 param, System.String param2, Int32 param3, Int32 param4) (at Assets/Scripts/Test.cs:93)
Test.method1 (Int32 param) (at Assets/Scripts/Test.cs:66)
Test.Update () (at Assets/Scripts/Test.cs:33)
So, I tested whether it is null or not by checking var2 == null and also var2.Equals(null) with Debug.Log() just before calling val2.run(), but all of them showed 'False'. They are not null.
What is the problem? Should I change Java side code and re-generate dll?
Thanks.
p.s. My java side code uses java.util.logging.Logger and related with other java project. (which is also referenced when converting jar into dll, as I said in second sentence of the question)
p.s.2. I'm sorry but I can't open java codes here because it is confidential code

I found that it was an error of our own code, but not java code itself. configurations.

Related

Call a java method in R using rJava

I use rJava to call a java code from R, trying to call an algorithm from SPMF tool. I tried to use a wrapper function as in this question, but this did not work with the SPMF code.
this is my R code:
library(rJava)
.jinit()
.jaddClassPath ( "C:/mydrive/eclipse-workspace/myfile/src")
print(.jclassPath())
obj <- .jnew("AlgoFPGrowth_Strings")
s <- .jcall(obj, returnSig= "V", method="runAlgorithm",
"input.csv","output.txt") , 0.4 )
it gives me error ,method runAlgorithm with signature (D)V not found
this is the main in java:
public static void main(String[] args) throws Exception {
AlgoFPGrowth_Strings fpwindow=new AlgoFPGrowth_Strings();
String input="input.csv";
String output="output.txt";
double minsupp = 0.4;
fpwindow.runAlgorithm( input, output, minsupp);
fpwindow.printStats();
}
I tried to change returnSig value into S and Ljava/lang/String; but I got the same error, could not find the method
when I apply the code on different java code with simple method it works, is there any idea how can I change my code?
Try the below methods,
Change your jclassPath, where you directly specify the complete pathname of your jar file including the jar name, say /home/user/mypath/myclass_name.jar
Or, you can unzip your jar file in a folder and refer to that path in your jclassPath.
If, the above does not work,
Try to write the 'runAlgorithm' method in the same class where you are calling. I have faced issues with calling external libraries/classes.

Getting java.lang.UnsatisfiedLinkError at runtime

I'm trying to use JNI to access C++ methods from a Java class. I'm able to compile (both in Eclipse or on command line) my Java class fine, but on executing the class at runtime, I'm getting:
Exception in thread "main" java.lang.UnsatisfiedLinkError: com.domain.services.CallServiceAPIS.createSession()I
at com.domain.services.CallServiceAPIS.createSession(Native Method)
at com.domain.services.CallServiceAPIS.main(CallServiceAPIS.java:18)
Java code is as follows:
package com.domain.services;
public class CallServiceAPIS {
static {
System.loadLibrary("service.client");
}
public native int createSession();
public static void main(String[] args) {
System.out.println(System.getProperty("java.library.path"));
new CallServiceAPIS().createSession();
}
}
I included the printout of the java.library.path just to make sure it's pointing to the correct location of the C++ library - and it is. I also tried setting the LD_LIBRARY_PATH in my Eclipse environment. But neither worked.
Note that the System.loadLibrary call IS working since 1) the code compiles and 2) the error occurs on line 18, which is the new CallServiceAPIs call.
C++ code:
int createSession(const PosServiceInfo info, const SessionArgs& args, Domain::UUID& uuidSession)
{
return int::undefined;
}
Any ideas?
Never mind. I realized that I was using the JNI interface incorrectly. I was thinking you could load an EXISTING C++ library using EXISTING C++ source. But you basically have to rewrite the existing code to make use of the JNI interface.

Does JNI and Java naming Convention matter when passing strings

Situation: I have a library of JNI files, the library is comprised of several functions that are called by the main header file in that JNI library (i.e., code1.h). I have a Java file (i.e., code2.java) that I want to pass to and from JNI header file (code1.h). I created a source code for the (code1.h) called (code1.c).
My question is: Does (code1.h), (code1.c), and (code2.java) have to be the same name for the communication between the JNI and the java?
EDIT: So (code1.h), (code1.c), and (code1.java) all have to be the same name in order for the (code1.java) to pass strings to/from (code1.c)/(code1.h)? And it is not possible to have (code2.java) pass strings to/from (code1.c)/(code1.h) because they are not named the same, is this correct?
For instance,
public class code1 { /*this is code2.java, but should the name be changed to (code1.java) to match that of the JNI?*/
static {
System.loadLibrary("myjni");
}
to pass strings to code1.h/code1.c
This will be compiled for android using Linux Debian"Wheezy" and Eclipse with Android SDK and NDK
While Java requires a match between compilation unit name (SomeClass.java being the name and public class SomeClass{ being the declaration, C does not require this.
You may name the C source and header files as you see fit as long as the function names/exported symbol names match the name of the native method on the java side. For example:
//JavaClass.java
public class JavaClass{
public native String getAString(String in);
}
And header would be:
// any name
JNIEXPORT jstring JNICALL
Java_JavaClass_getAString(JNIEnv *, jobject, jstring);
with matching C files. You could name this header catsMakeTheWorldGoRound.h for all Java cares.
Here is an example of what your "JNI object" should look like.
//In my experience, it is better to put the JNI object into a separate package.
package org.example;
public class Code1
{
static
{
// On a Linux system, the actual name of the library
// is prefixed with "lib" and suffixed with ".so"
// -- e.g. "myjni-java.so"
// Windows looks for "myjni-java.dll"
//
// On a Windows system, we also need to load the prequisite
// libraries first. (Linux loaders do this automatically).
//
String osname = System.getProperty("os.name");
if (osname.indexOf("win") > -1 || osname.indexOf("Win") > -1)
{
System.loadLibrary("myjni");
}
System.loadLibrary("myjni-java");
}
// Now we declare the C functions which we will use in our Java code.
public static native void foo(int bar);
public static native int bar(String foo);
//...
}
Given that you have compiled your JNI library correctly, you can then call the C functions from other Java classes like this:
//Again, in my experience, it is better to explicitly give the package name here.
org.example.Code1 Code1= new org.example.Code1();
Code1.foo(123);
int a= Code1.bar("Hello C function from Java function!");
Does this help you with your question? (I am not an expert in JNI, so I might not be able to help further.)

how to get ExePath/MainClassPath/MainJarPath of java App at runtime?

The "Path of the running EXE" is well defined for Windows Applications written in C#, C++, VB, etc.
-
Java Applications on Windows are not "EXE-Applications" but a class-file or a jar-file is started rather than an EXE-file.
So for java Applications the term "ExePath" should be translated to "MainClassPath" or to "JarPath" resp.
-
In some cases a programmer needs to know the physical Path of the Application's jar or MainClass.
(e.g. when you develop a large project both in java and in c# with identical classes and identical methods)
-
thanks to other stackoverflow users this statement does the job:
String exePath = URLDecoder.decode(this.getClass().getProtectionDomain().getCodeSource().getLocation().getPath(), "UTF-8")
and now comes my question:
If I put the same code into any helper/utils jar-library then it will return the path of the helperlib.jar, it will NOT return the path of my MainClass/AppJar !
-
So the final getExePath() helper method should look something like:
return(URLDecoder.decode(Thread.currentThread().getStartingThread().getMainClass().getProtectionDomain().getCodeSource().getLocation().getPath(), "UTF-8"));
(if there only were methods like getStartingThread() and getMainClass() in java ...)
-
please, point me to the final solution, how can I implement these steps:
get the starting Thread
get the Main Class of the starting Thread
get the Path of the Main Class
if I get you right, just put the method with a parameter of the desired class into your helperlib-class ...
e.g.:
public static String getExePath(Object main) {
String path = "";
try {
path = URLDecoder.decode(main.getClass().getProtectionDomain().getCodeSource().getLocation().getPath(), "UTF-8");
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
}
return path;
}
then you can call it e.g. from inside your main-jar with the parameter 'this' ...
System.out.println("AppPath:\n" + helperlib.getExePath(this));
... and you got the path of the class specified in you parameter
hope it helps and sorry for my bad english ... ;)

JNI native call *prototypes* in their own class/namespace?

I have written JNI wrappers to export the API of a C application (G-WAN) which embeds a JVM. The native calls are implemented in the C application and exported with RegisterNatives().
Ideally I would have a 'gwan' class for the G-WAN API:
import gwan // G-WAN API
public class hello {
public static int jmain(long env, String[] args) {
gwan.xbuf_cat(gwan.get_reply(env), "Hello World");
return 200; // HTTP status (200:'OK')
}
}
I would like to do something like the "#import gwan" above to import the native call prototypes, but currently I only have the following (which works):
public class hello {
public static int jmain(long env, String[] args) {
gwan_xbuf_cat(gwan_get_reply(env), "Hello World");
return 200; // HTTP status (200:'OK')
}
public static native long gwan_get_reply(long env);
public static native void gwan_xbuf_cat(long ctx, String str);
}
Again, the implementation of the native calls in made in the G-WAN executable (not in a Java class stored on disk).
Because the G-WAN API is quite large, I would like to have the native call prototypes in their own 'gwan' class (or namespace) if possible (like in the first hello example above).
Any suggestion about how to do that?
(please post Java or JNI code because I am not a Java expert)
Disclamer: I am involved in the development of this project.
I would suggest that you read following paper on JNI from Sun now Oracle
http://java.sun.com/docs/books/jni/html/jniTOC.html
And after that it should be understandable but some pseudocode and its not tested would be to move the two gwanapi calls into its own file named gwanapi.java
public class gwanapi {
public static native long get_reply(long answer);
public static native void xbuf_cat(long ctx,String str);
}
then you compile that file with javac gwanapi.java -> output: gwanapi.class
you type javah -jni for the c/c++ header implementation:
javah -jni gwanapi
the next you should in your hello java class is to call static{ System.loadLibrary("gwanapi");}
Pseudo code and NOT tested
public class hello{
static{
System.loadLibrary("gwanapi");
}
public static int jmain(long env,String args[]){
gwanapi.xbuf_cat(gwanapi.get_reply(env),"Hello World!");
return 200;
}
}
and you should be good to go.
But I might have missed a point or two but I think this is the smallest amount of work you should do.
Oh by the way http://en.wikipedia.org/wiki/Java_Native_Interface is also some form of source for JNI calls and how it works and lead you to more sites with more information.
Thanks
Being a C programmer, I have had to read the C source code of the JVM to find that Java remaps Class (path) names with dots instead of slashes.
Since the G-WAN directory hierarchy uses IP addresses to define listeners and virtual hosts (192.168.10.10_80/#domain.com/csp), those dots were confusing the FindClass() JNI call, making it fail to find the classes.
I also found that the classpath path separator is a ":" for Unix and a ";" for Windows. That was not the cause of my problem, but it might cause the same issue.
Finally, I stopped using the GCJ compiler's JVM because it does not support formating doubles (since at least 2006). Using either OpenJDK or the SUN/ORACLE JVM works as expected.
All works fine now. I post all this here just in case it may help others.

Categories

Resources