If I pass an object to Rhino in Java, is there an annotation I can use in that object's class to hide methods\fields from Rhino JavaScripts to make them inaccessible to the JavaScripts? Or any other method of doing this?
I did some research, and it looks like I could wrap the Objects with Scriptable, but that seems like a messier solution than what might be potentially really easy to do as it seems like it would be a fairly standard feature.
Thanks.
I wasn't able to find any support for this in the Rhino engine (if anyone knows of such a thing, please say so.)
That said, it is easy to implement. You need to implement your own WrapFactory, and your own NativeJavaObject. You also need to invoke create your own ContextFactory to assure your WrapFactory is being used on all Context objects used throughout the script engine. This sounds like a lot of work... but really it is actually just a lot of wrapper code. Here is a snippet of my implementation. WHAT IS MISSING: is the actual call to ContextFactory.initGlobal to set the global ContextFactory to your implementation of ContextFactory.
Obviously this code is not thread safe
class ProtectedContextFactory extends ContextFactory
{
private static final ProtectedWrapFactory wrapper = new ProtectedWrapFactory();
#Override
protected Context makeContext()
{
Context c = super.makeContext();
c.setWrapFactory(wrapper);
return c;
}
}
class ProtectedWrapFactory extends WrapFactory
{
#Override
public Scriptable wrapAsJavaObject(Context cx, Scriptable scope, Object javaObject, Class<?> staticType)
{
return new ProtectedNativeJavaObject(scope, javaObject, staticType);
}
}
class ProtectedNativeJavaObject extends NativeJavaObject
{
private static final HashMap<Class<?>, ArrayList<String>> CLASS_PROTECTION_CACHE = new HashMap<Class<?>, ArrayList<String>>();
private ArrayList<String> m_protectedMembers;
public ProtectedNativeJavaObject(Scriptable scope, Object javaObject, Class<?> staticType)
{
super(scope, javaObject, staticType);
Class<?> clazz = javaObject != null ? javaObject.getClass() : staticType;
m_protectedMembers = CLASS_PROTECTION_CACHE.get(clazz);
if(m_protectedMembers == null)
m_protectedMembers = processClass(clazz);
}
private static ArrayList<String> processClass(Class<?> clazz)
{
ArrayList<String> protectedMethods = new ArrayList<String>();
CLASS_PROTECTION_CACHE.put(clazz, protectedMethods);
for(Method m : clazz.getMethods())
{
if(m.getAnnotation(ScriptHiddenMember.class) != null)
protectedMethods.add(m.getName());
}
for(Field f : clazz.getFields())
{
if(f.getAnnotation(ScriptHiddenMember.class) != null)
protectedMethods.add(f.getName());
}
return protectedMethods;
}
#Override
public boolean has(String name, Scriptable start)
{
if(m_protectedMembers.contains(name))
return false;
else
return super.has(name, start);
}
#Override
public Object get(String name, Scriptable start)
{
if(m_protectedMembers.contains(name))
return NOT_FOUND;
else
return super.get(name, start);
}
}
#Target({ElementType.METHOD, ElementType.FIELD})
#Retention(RetentionPolicy.RUNTIME)
public #interface ScriptHiddenMember {}
Using an interface and hiding methods by not including them in the interface will not work, the script will still be able to access methods without signatures in the interface, even when you add the object to the engine while casting it to it's interface. The way I've found to hide methods is to declare them private, protected, or (no modifier); this keeps the script from having access. If you use protected or (no modifier), then you can still use them in your code in other classes (assuming you are calling them from a valid place) while having them hidden from JS.
For variables, as long as you declare them private, protected, or (no modifier), and do not include any getter methods, the variable will be hidden. Public variables are accessible even without getter methods.
Using this interface:
public interface NodeInterface {
public int getPosition();
}
And this implementing class:
public class Node implements NodeInterface{
private int x = 5;
public int y = 8;
private int position = 0;
public int getPosition(){return position;}
private String getString(){return "hello";}
public String getBing(){return "bing";}
}
Results in this code:
ScriptEngineManager factory = new ScriptEngineManager();
ScriptEngine engine = factory.getEngineByName("JavaScript");
NodeInterface node = (NodeInterface)new Node();
engine.put("node", node);
try {
engine.eval("println(node.x)");//undefined
engine.eval("println(node.y)");//8
engine.eval("println(node.position)");//0
engine.eval("println(node.getPosition())");//0
engine.eval("println(node.getBing());");//hello
engine.eval("println(node.getString())");//TypeError: Cannot find function getString.
} catch (ScriptException e1) {
e1.printStackTrace();
}
Why you don't try to Java interfaces. For each class that will be exposed to Rhino, define an interface which contains getters and setters for the fields you want to expose, and methods that you want to expose.
More about how to access Java interfaces you can find on Rhino: Access Java interface variables in Javascript implementation
Related
I have an interface similar to this:
public interface Getter {
Object get(Params params);
}
that I implement using a reflective call to a different method:
public class GetterImpl implements Getter {
private final Object target;
private final Method method; //doStuff method
public GetterImpl(Object target, Method method) {
this.target = target;
this.method = method;
}
#Override
public Object get(Params params) {
//both the target and arguments depend on Params
return method.invoke(chooseTarget(params), prepareArgs(params));
}
private Object chooseTarget(Params params) {
if (params.getTargetOverride() != null) {
return params.getTargetOverride();
}
return target;
}
private Object[] prepareArgs(Params params) {
...
}
}
Is it possible to instead generate a class implementing Getter with equivalent logic but without reflection? Effectively a class like this:
public class GeneratedGetterImpl implements Getter {
...
#Override
public Object get(Params params) {
//somehow call doStuff directly (different method for each generated impl)
return target.doStuff(prepareArgs(params));
}
}
I'm looking into using Byte Buddy to generate such a class on the fly, but all the examples provide some sort of statically known method interceptor, and never delegate to a dynamically chosen target and method.
It's clearly not a trivial task, but can this be done with Byte Buddy? Or a different library?
UPDATE:
Here's my best attempt so far:
Target target = new Target();
Method method = Target.class.getMethod("doStuff", Book.class);
//Helper class that computes the new arguments based on the original
Prepare prepare = new Prepare();
Method doPrep = Prepare.class.getMethod("doPrep", Params.class);
Getter getter = (Getter) new ByteBuddy()
.subclass(Object.class)
.implement(Getter.class)
.method(named("get")).intercept(
MethodCall.invoke(method).on(target)
.withMethodCall(
MethodCall.invoke(doPrep).on(prepare).withAllArguments()
))
.make()
.load(getClass().getClassLoader())
.getLoaded()
.newInstance();
public static class Prepare {
public Book doPrep(Params params) {
return new Book(params.getTitle());
}
}
This does what I want, but only if the targeted method takes 1 argument (Book in my case). I'm struggling to figure out how to have it return an array that I then spread when calling the target method.
E.g.
public static class Prepare {
//returns all the arguments
public Object[] doPrep(Params params) {
return new Object[] { new Book(params.getTitle()) };
}
}
Such a facility does already exists in the JRE, if we restrict it to binding an interface to a matching target method.
public static void main(String[] args) throws NoSuchMethodException {
Function<Double,Double> f1 = create(Math.class.getMethod("abs", double.class));
System.out.println(f1.apply(-42.0));
Map<Double,Double> m = new HashMap<>();
Function<Double,Double> f2 = create(Map.class.getMethod("get", Object.class), m);
m.put(1.0, 123.0);
System.out.println(f2.apply(1.0));
}
static Function<Double,Double> create(Method m) {
MethodHandles.Lookup l = MethodHandles.lookup();
MethodType t = MethodType.methodType(Double.class, Double.class);
try {
return (Function)LambdaMetafactory.metafactory(l, "apply",
MethodType.methodType(Function.class), t.erase(), l.unreflect(m), t)
.getTarget().invoke();
} catch(Throwable ex) {
throw new IllegalStateException(ex);
}
}
static Function<Double,Double> create(Method m, Object target) {
MethodHandles.Lookup l = MethodHandles.lookup();
MethodType t = MethodType.methodType(Double.class, Double.class);
try {
return (Function)LambdaMetafactory.metafactory(l, "apply",
MethodType.methodType(Function.class, m.getDeclaringClass()),
t.erase(), l.unreflect(m), t)
.getTarget().invoke(target);
} catch(Throwable ex) {
throw new IllegalStateException(ex);
}
}
42.0
123.0
This demonstrates that adaptations like auto-boxing and casting as required for generic functions are included, but any other adaptations of parameters or results are not possible and have to be performed by pre-existing decorating code. Most notably, varargs processing is not included.
The documentation is exhaustive. It’s strongly recommended to read it in all details before using the class. But the things you can do wrong here, are similar to the things you can do wrong when implementing your own bytecode generator.
Using Byte Buddy, you can create a MethodCall instance that represents your proxied method and use it as an implementation. I assume that you looked into delegation which requires a more static model:
MethodCall.invoke(SomeClass.class.getMethod("foo")).with(...)
You can also provide other method call instances as arguments to methods to achieve what you have in your example code.
As for your updated question, I'd recommend you a hybrid approach. Implement some container:
class Builder<T> {
Builder with<T>(T value);
T[] toArray();
}
and then you can use Byte Buddy to invoke it for creating your result value:
MethodCall builder = MethodCall.construct(Builder.class.getConstructor());
for (SomeInfoObject info : ...) {
builder = MethodCall.invoke(Builder.class.getMethod("with", Object.class))
.on(builder)
.with(toMethodCall(info));
}
builder = MethodCall.invoke(Builder.class.getMethod("toArray")).on(builder);
Byte Buddy's goal is to make weaving code easy, not to replace writing static code which is the much better option if you have the opportunity.
I was wondering lately, which one of the three methods of passing parameters to the method - presented below - are the best for you, your CPU, memory and why. I am considering methods which allow me to pass more arguments in future, without changing the method signature.
If you know something better, I am here to listen and learn.
Pass by methods
Params.java
public interface Params {
int getParamOne();
int getParamTwo();
}
Calling
obj.foo(new Params() {
#Override
public int getParamOne() {
return 1;
}
#Override
public int getParamOne() {
return 2;
}
});
Receiving
public void foo(Params p) {
int p1 = p.getParamOne();
int p2 = p.getParamTwo();
}
Pass by class fields
Params.java
public class Params {
private int paramOne;
private int paramTwo;
// Getters and setters here
}
Calling and receiving
No magic here, just create a new Params object, use setters, pass it to the method and use getters.
Pass by Properties class
Calling
properties.put("paramOne", 1);
properties.put("paramTwo", 2);
obj.foo(properties);
Receiving
public void foo(Properties properties) {
int a = (int) properties.get("paramOne");
int b = (int) properties.get("paramTwo");
}
I was pleased to show an real-life example of code, which actually needs passing varying types and number of properties. I'm using the third method - passing by the properties:
public interface DataProvider {
public String getContent(Properties properties);
}
public class HttpProvider implements DataProvider {
#Override
public String getContent(Properties properties) {
InputStream in = new URL(properties.get("URL")).openStream();
String content = IOUtils.toString(in);
IOUtils.closeQuietly(in);
return content;
}
public class FtpProvider implements DataProvider {
#Override
public String getContent(Properties properties) {
FTPClient ftpClient = new FTPClient();
ftpClient.connect(properties.get("server"), properties.get("port"));
ftpClient.login(properties.get("user"), properties.get("pass"));
// Get file stream and save the content to a variable here
return content;
}
}
One interface for a different methods of obtaining a file. I am not persisting that this is good or not, it's just an example of code I've stumbled upon in my current project in work and I was wondering if could it be done better.
The usage of a "Params" class is better than properties, in performance. The java compiler can handle such short lived classes quite well.
One sees properties on some constructors / factory methods, like for XML and such.
One sees a parameter containing class in larger systems, to keep the API restricted to one parameter, and not use overloaded methods.
I would do:
public class Params {
public final int a;
public final int b;
public Params(int a, int b) {
this.a = a;
this.b = b;
}
}
And in the class immediately use params.a.
For the rest there is also the Builder Pattern, but that would be more a substitute for a complex constructor.
Signatures in interfaces should not ever change!!! If you contemplate to change APIs in the future (i.e. change, add or remove a parameter), an acceptable way may be by incapsulating your parameters in objects in order to do not break signatures.
I have a public abstract class java.nio.ByteBuffer instance which is actually an instance of private class java.nio.HeapByteBuffer and I need to make a proxy object which would call some invocation method handler to check access permissions and then call the invoked method on the actual instance.
The problem is that the java.nio.ByteBuffer class has only private constructors and also has some final methods, thus I can not create proxy instances with javassist.util.proxy.ProxyFactory class.
So, how can I make a proxy object to control the invocation of a java.nio.ByteBuffer instance including those final methods invocation?
Please be aware that I am presenting a solution based on my own (FOSS) framework Byte Buddy which is however already mentioned as a potential solution in one of the comments.
Here is a simple proxy approach which creates a subclass. First, we introduce a type for creating proxies for ByteBuffers:
interface ByteBufferProxy {
ByteBuffer getOriginal();
void setOriginal(ByteBuffer byteBuffer);
}
Furthermore, we need to introduce an interceptor to use with a MethodDelegation:
class Interceptor {
#RuntimeType
public static Object intercept(#Origin(cacheMethod = true) Method method,
#This ByteBufferProxy proxy,
#AllArguments Object[] arguments)
throws Exception {
// Do stuff here such as:
System.out.println("Calling " + method + " on " + proxy.getOriginal());
return method.invoke(proxy.getOriginal(), arguments);
}
}
This interceptor is capable of intercepting any method as the #RuntimeType casts the return type in case that it does not fit the Object signature. As you are merely delegating, you are safe. Plase read the documentation for details. As you can see from the annotations, this interceptor is only applicable for instances of ByteBufferProxy. Bases on this assumption, we want to:
Create a subclass of ByteBuffer.
Add a field to store the original (proxied) instance.
Implement ByteBufferProxy and implement the interface methods to access the field for the stored instance.
Override all other methods to call the interceptor that we defined above.
This we can do as follows:
#Test
public void testProxyExample() throws Exception {
// Create proxy type.
Class<? extends ByteBuffer> proxyType = new ByteBuddy()
.subclass(ByteBuffer.class)
.method(any()).intercept(MethodDelegation.to(Interceptor.class))
.defineField("original", ByteBuffer.class, Visibility.PRIVATE)
.implement(ByteBufferProxy.class).intercept(FieldAccessor.ofBeanProperty())
.make()
.load(getClass().getClassLoader(), ClassLoadingStrategy.Default.WRAPPER)
.getLoaded();
// Create fake constructor, works only on HotSpot. (Use Objenesis!)
Constructor<? extends ByteBufferProxy> constructor = ReflectionFactory
.getReflectionFactory()
.newConstructorForSerialization(proxyType,
Object.class.getDeclaredConstructor());
// Create a random instance which we want to proxy.
ByteBuffer byteBuffer = ByteBuffer.allocate(42);
// Create a proxy and set its proxied instance.
ByteBufferProxy proxy = constructor.newInstance();
proxy.setOriginal(byteBuffer);
// Example: demonstrates interception.
((ByteBuffer) proxy).get();
}
final methods are obviously not intercepted. However as the final methods in ByteBuffer only serve as convenience methods (e.g. put(byte[]) calls put(byte[],int,int) with the additional arguments 0 and the array length), you are still able to intercept any method invocation eventually as these "most general" methods are still overridable. You could even trace the original invocation via Thread.currentCallStack().
Byte Buddy normally copies all constructors of its super class if you do not specify another ConstructorStrategy. With no accessible constructor, it simply creates a class without constructors what is perfectly legal in the Java class file format. You cannot define a constructor because, by definition, this constructor would need to call another constructor what is impossible. If you defined a constructor without this property, you would get a VerifierError as long as you do not disable the verifier altogether (what is a terrible solution as it makes Java intrinsically unsafe to run).
Instead, for instantiation, we call a popular trick that is used by many mocking frameworks but which requires an internal call into the JVM. Note that you should probably use a library such as Objenesis instead of directly using the ReflectionFactory because Objenesis is more robust when code is run on a different JVM than HotSpot. Also, rather use this in non-prduction code. Do however not worry about performance. When using a reflective Method that can be cached by Byte Buddy for you (via cacheMethod = true), the just-in-time compiler takes care of the rest and there is basically no performance overhead (see the benchmark on bytebuddy.net for details.) While reflective lookup is expensive, reflective invocation is not.
I just released Byte Buddy version 0.3 and I am currently working on documentation. In Byte Buddy 0.4, I plan to introduce an agent builder which allows you to redefine classes during load-time without knowing a thing about agents or byte code.
I can suggest you 2 solutions.
First, simple, not universal, but probably useful for you.
As far as I can see ByteBuffer has several package-private constructors that allow its subclassing and the following final methods:
public final ByteBuffer put(byte[] src) {
public final boolean hasArray() {
public final byte[] array() {
public final int arrayOffset() {
public final ByteOrder order() {
ByteBuffer extends Buffer that declares some of these methods:
public final boolean hasArray() {
public final Object array() {
public final int arrayOffset() {
As you can see, put() and order() are absent here, return type of array() is a little bit confusing, but still can be used.
So, if you use only these 3 methods you can subclass Buffer and create universal wrapper that wraps any other Buffer including ByteBuffers. If you want you can use javaassist's proxy although IMHO it is not necessarily here.
Second, more universal but more tricky solution. You can create agent that removes final modifiers from speicific class (ByteBuffer in your case) during class loading. Then you can create javassist proxy.
Variation of second solution is following. Copy ByteBuffer soruce code to separate project. Remove final modifiers and compile it. Then push it into bootstrap classpath. This solutions is probably easier than second.
Good luck anyway.
Thanks to #raphw I have managed to make a proxy object construction class which makes a proxy for java.nio.ByteBuffer but that class has final methods which I can not overcome and they are extensively used in the required code, those final methods are Buffer.remaining() and Buffer.hasRemaining(), thus they just can not be proxy mapped.
But I would like to share the classes I have made, just as a report.
public final class CacheReusableCheckerUtils {
private static ByteBuddy buddy = new ByteBuddy();
private static Objenesis objenesis = new ObjenesisStd();
public static <T> T createChecker(T object) {
return createChecker(new CacheReusableCheckerInterceptor<>(object));
}
public static <T> T createChecker(CacheReusableCheckerInterceptor<T> interceptor) {
return objenesis.getInstantiatorOf(createCheckerClass(interceptor)).newInstance();
}
private static <T> Class<? extends T> createCheckerClass(CacheReusableCheckerInterceptor<T> interceptor) {
Class<T> objectClass = interceptor.getObjectClass();
Builder<? extends T> builder = buddy.subclass(objectClass);
builder = builder.implement(CacheReusableChecker.class).intercept(StubMethod.INSTANCE);
builder = builder.method(MethodMatchers.any()).intercept(MethodDelegation.to(interceptor));
return builder.make().load(getClassLoader(objectClass, interceptor), Default.WRAPPER).getLoaded();
}
private static <T> ClassLoader getClassLoader(Class<T> objectClass, CacheReusableCheckerInterceptor<T> interceptor) {
ClassLoader classLoader = objectClass.getClassLoader();
if (classLoader == null) {
return interceptor.getClass().getClassLoader();
} else {
return classLoader;
}
}
}
public class CacheReusableCheckerInterceptor<T> {
private T object;
private boolean allowAccess;
private Throwable denyThrowable;
public CacheReusableCheckerInterceptor(#NotNull T object) {
this.object = object;
}
#SuppressWarnings("unchecked")
public Class<T> getObjectClass() {
return (Class<T>) object.getClass();
}
#RuntimeType
public final Object intercept(#Origin(cacheMethod = true) Method method, #This T proxy, #AllArguments Object[] arguments) {
try {
switch (method.getName()) {
case "allowAccess":
allowAccess();
return null;
case "denyAccess":
denyAccess();
return null;
default:
return invokeMethod(method, arguments);
}
} catch (Exception e) {
throw new CacheReusableCheckerException(method, object, proxy, e);
}
}
private Object invokeMethod(Method method, Object[] arguments) throws IllegalAccessException, InvocationTargetException {
checkMethodAccess(method.getName());
return method.invoke(object, arguments);
}
private void allowAccess() {
if (allowAccess) {
error("double use");
}
allowAccess = true;
onAccessAllowedAfter(object);
}
private void denyAccess() {
if (!allowAccess) {
error("double free");
}
onAccessDeniedBefore(object);
allowAccess = false;
denyThrowable = new Throwable();
}
private void checkMethodAccess(String name) {
if (!allowAccess) {
switch (name) {
case "hash":
case "equals":
case "toString":
case "finalize":
break;
default:
error("use after free");
}
}
}
private void error(String message) {
throw new CacheReusableCheckerException(message, denyThrowable);
}
protected void onAccessAllowedAfter(T object) {
}
protected void onAccessDeniedBefore(T object) {
}
}
public interface CacheReusableChecker {
void allowAccess();
void denyAccess();
}
Hope you can help me with this:
I have ...
a string list of class names called classNameList
a generic class Geography<T>
a static generic method <T> void read(Class<T> cl, Geography<T> geo)
I want to loop through the string class name list and call the generic method for each of these classes.
What I tried but obviously did not work:
for (int i = 0; i < classNameList.length; i++) {
Class<?> myClass = Class.forName(classNameList[i].getName());
Geography<myClass.newInstance()> geo;
read(myClass, geo);
}
Error: myClass.newInstance cannot be resolved to a type
My code runs perfectly for a single call of the generic function:
Geography<ExampleClass> ExampleGeo;
read(ExampleClass.class, ExampleGeo);
Any ideas how I could do this?
UPDATE:
Thanks for the helpful input, still it's hard for me to adopt it to my real code.
So this is the non simplyfied problem:
I do ready in shapefile-Data with a shapefileLoader, for each feature of the Shapefile a class (GuadAgent) is initialized with a predifined class (PlantWind). I have shapefiles in my input-directory with the names of the Classes their features do represent. I want Java to read in the shapefiles and create the respective agent class. (the agents are also placed in a context and a geography..)
Used classes are: ShapefileLoader, Geography, the other classes can be find at the same website
This part is in the main-method:
Geography<GuadAgent> guadGeography = GeographyFactoryFinder.createGeographyFactory(null).createGeography("guadGeography", context, new GeographyParameters<GuadAgent>());
Context<GuadAgent> context = new DefaultContext<GuadAgent>();
FileFilter filter = new FileFilter() {
#Override
public boolean accept(File file) {
return file.getName().endsWith(".shp"); // return .shp files
}
};
String shapefileDir = System.getProperty("user.dir")+"\\input\\shp\\";
File folder = new File(shapefileDir);
File[] listOfFiles = folder.listFiles(filter);
for (File classFile : listOfFiles) {
try {
readForName(classFile,context,guadGeography);
} catch (ClassNotFoundException | MalformedURLException
| FileNotFoundException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
The static Method that reads in the names:
static <T> void readForName(File classFile, Context<GuadAgent> context,Geography<GuadAgent> guadGeography) throws ClassNotFoundException, MalformedURLException, FileNotFoundException {
String shapefileDir = System.getProperty("user.dir")+"\\input\\shp\\";
String className = classFile.getName().split("\\.(?=[^\\.]+$)")[0];
File shapefile = null;
shapefile = new File(shapefileDir+classFile.getName());
if (!shapefile.exists()) {
throw new FileNotFoundException("Could not find the given shapefile: " + shapefile.getAbsolutePath());
}
switch (className) {
case "PlantWind":
ShapefileLoader<PlantWind> PlantWindLoader = new ShapefileLoader<PlantWind>(PlantWind.class,shapefile.toURI().toURL() , guadGeography, context);
PlantWindLoader.load();
PlantWindLoader.close();
System.out.println(context.getObjects(PlantWind.class).size());
break;
// Todo Add other Agent types
default:
break;
}
How can I get rid of the switch? Although their number is finit, there are very many different agents...
Unfortunately, there's no syntax close to your intention (nice idea though).
The basic problem is that Class.forName() returns an unknown Class<?>, so you need a cast somewhere. It's just a mater of where you put it.
I suggest this approach (which compiles) that bundles up doing a read() based on a class name:
static <T> void readForName(String className) throws ClassNotFoundException {
Class<T> myClass = (Class<T>) Class.forName(className);
Geography<T> geo = new Geography<T>(); // No code shown. Adjust as required
read(myClass, geo);
}
May I also suggest using the foreach loop syntax, for tidier code:
for (String className : classNameList) {
readForName(className.getName());
}
Creating instances from Generic Types at Runtime
I am not entirely clear on what you are trying to accomplish, but at first look it looks like the simplest solution is the best solution.
It could be solved with using a scripting environment ( Groovy, JavaScript, JRuby, Jython ) that could dynamically evaluate and execute arbitrary code to create the objects, but that got extremely convoluted and overly complex, just to create an object.
But unfortunately I think it has a very pedestrian solution.
As long as there is a predefined set of supported types, you can use a Factory pattern. Here I just leverage the Provider<>T interface from the javax.inject/com.google.inject package.
Q26289147_ProviderPattern.java
public class Q26289147_ProviderPattern
{
private static final List<String> CLASS_NAMES = ImmutableList.of("String", "Integer", "Boolean");
private static final Map<String, Provider<StrawManParameterizedClass>> PROVIDERS;
static
{
final ImmutableMap.Builder<String, Provider<StrawManParameterizedClass>> imb = ImmutableMap.builder();
for (final String cn : CLASS_NAMES)
{
switch (cn)
{
case "String":
imb.put(cn, new Provider<StrawManParameterizedClass>()
{
#Override
public StrawManParameterizedClass<String> get() { return new StrawManParameterizedClass<String>() {}; }
});
break;
case "Integer":
imb.put(cn, new Provider<StrawManParameterizedClass>()
{
#Override
public StrawManParameterizedClass<Integer> get() { return new StrawManParameterizedClass<Integer>() {}; }
});
break;
case "Boolean":
imb.put(cn, new Provider<StrawManParameterizedClass>()
{
#Override
public StrawManParameterizedClass<Integer> get() { return new StrawManParameterizedClass<Integer>() {}; }
});
break;
default:
throw new IllegalArgumentException(String.format("%s is not a supported type %s", cn, Joiner.on(",").join(CLASS_NAMES)));
}
}
PROVIDERS = imb.build();
}
static <T> void read(#Nonnull final StrawManParameterizedClass<T> smpc) { System.out.println(smpc.type.toString()); }
static abstract class StrawManParameterizedClass<T>
{
final TypeToken<T> type = new TypeToken<T>(getClass()) {};
#Override
public String toString() { return type.getRawType().getCanonicalName(); }
}
public static void main(final String[] args)
{
for (final String cn : CLASS_NAMES)
{
read(PROVIDERS.get(cn).get());
}
}
}
Disclaimer:
This is just a proof of concept example, I would never use a switch
statement like that in production code I would use a Strategy
Pattern or Chain of Responsibility Pattern to encapsulate the logic
of what type to create based on the ClassName key.
This initially looked like a generics problem, it isn't, it is a creation problem.
That said, you don't need to pass around instances of Class<?> you can get Generic Type information off of Parameterized classes at runtime with TypeToken from Guava.
You can even create instances of any generic type at runtime with TypeToken from the Guava library.
The main problem is this syntax isn't supported: Geography<myClass.newInstance()> geo; and I can't think of anyway to fake it other than the Provider implementation above.
Here is a straw man example of how to use TypeToken so that your
parameterized classes will always know their types!
Q26289147.java
import com.google.common.reflect.TypeToken;
public class Q26289147
{
public static void main(final String[] args) throws IllegalAccessException, InstantiationException
{
final StrawManParameterizedClass<String> smpc = new StrawManParameterizedClass<String>() {};
final String string = (String) smpc.type.getRawType().newInstance();
System.out.format("string = \"%s\"",string);
}
static abstract class StrawManParameterizedClass<T>
{
final TypeToken<T> type = new TypeToken<T>(getClass()) {};
}
}
Notes:
Works great for classes that have a default no arg constructor.
Works better than using straight reflection if there are no default no arg constructors.
Should play well with Guice allowing you to use the ".getRawType()generatedClassto pass togetInstance()` of an Injector. have not tried this yet, I just thought of it!
You can use Class<T>.cast() to do casting that doesn't need #SuppressWarning("unchecked") all over the place.`
You can create a static factory method in Geography (or in any other class):
public static <T> Geography<T> newInstance(Class<T> cls)
throws ReflectiveOperationException {
return new Geography<T>(cls.newInstance());
}
I made a guess at the Geography class's constructor. If I guessed wrong, edit your question to include the constructor(s) in Geography.
You can create a static factory method in Geography (or in any other class):
public static <T> Geography<T> newInstance(Class<T> cls)
throws ReflectiveOperationException {
return new Geography<T>(cls.newInstance());
}
I made a guess at the Geography class's constructor. If I guessed wrong, edit your question to include the constructor(s) in Geography.
Update: I'm not sure what the Geography class is meant to do. If it needs a generically typed object, it might look like this:
public class Geography<T> {
private final T data;
public Geography(T data) {
this.data = Objects.requireNonNull(data);
}
}
If it needs a class, the constructor might look like this:
public class Geography<T> {
private final Class<T> dataClass;
public Geography(Class<T> cls) {
this.dataClass = Objects.requireNonNull(cls);
}
}
I'm working with the Java AWS API to monitor some EC2 instances and on every refresh I need to query the service which returns a bunch of Instance objects (freshly constructed). I want to extend the functionality of these objects, I figure I can maintain a map of MyInstance objects which can be refreshed with the new Instances on each poll.
Now I could easily do this with a simple wrapper class that holds the original Instance as a property the issue is I would like to keep access to the base Instance API as I already use these functions in my code. Would it be possible to replace only the super-class parts of an instantiated object? Contrived example of what I'm going for:
class Instance {
protected int prop;
public Instance(int prop) {
this.prop = prop;
}
}
class MyInstance extends Instance {
protected int prop2;
public MyInstance(int prop, int prop2) {
super(prop);
this.prop2 = prop2;
}
}
MyInstance foo = new MyInstance(1, 2);
Instance foster = new Instance(3);
//what i want to do
foo.adoptedBy(foster);
//with the result that foo.prop == 3
Obviously this example would be trivial to transform but in my actual case there are far more properties that need to be transferred. Can Reflection do this? What kind of performance impact am I looking at if I use Reflection for 10 of these every second? Thanks for reading!
The best solution is to combine both your ideas:
Wrap the original Instance in a class that extends the Instance class. (In the constructor of the child class, you can create a new Instance object and set it)
Delegate all methods to the wrapped instance (and add new properties)
In your foster method, you simply change the wrapped Instance.
class Instance {
private int prop;
public Instance(int prop) {
this.prop = prop;
}
public int getProp() {
return prop;
}
}
class MyInstance extends Instance {
private Instance delegate;
private int prop2;
public MyInstance(Instance delegate, int prop2) {
super(prop);
this.delegate = delegate;
this.prop2 = prop2;
}
#Override
public int getProp() {
return delegate.getProp();
}
public int getProp2() {
return prop2;
}
public void foster(Instance i) {
delegate = i;
}
}
MyInstance foo = new MyInstance(1, 2);
Instance foster = new Instance(3);
//what i want to do
foo.adoptedBy(foster);
//with the result that foo.getProp() == 3