Develop Reference

Android annotation processing - generate different code for different build flavor

I'm building a library that requires some annotation processing to generate code. I now run into an issue that the release build doesn't need to have as much code as the debug build does (since this is a library for modifying configuration variants - primarily used for testing purposes). The following code illustrates the situations. Let's say I want to create a class ConfigManager from some annotated classes and properties. In debug builds, I need this much:
public class ConfigManager {
public Class getConfigClass() {
return abc.class;
}
public void method1() {
doSomething1();
}
public void method2() {
doSomething2();
}
public void method3() {
doSomething3();
}
}
While in release builds, I only need this much:
public class ConfigManager {
public Class getConfigClass() {
return abc.class;
}
}
I have a feeling it may be possible by writing a Gradle plugin to check for build flavor at compile time and invoke a different processor/or somehow pass a parameter to a processor to generate different code. However this topic is pretty new to me so I'm not sure how to achieve this. A couple hours of googling also didnt help. So I'm wondering if anyone could give me a direction or example? Thanks

Pass an option (release=true/false) to your processor.
From javac https://docs.oracle.com/javase/8/docs/technotes/tools/windows/javac.html
-Akey[=value]
Specifies options to pass to annotation processors. These options are not interpreted by javac directly, but are made available for use by individual processors. The key value should be one or more identifiers separated by a dot (.).
In combination with Processor.html#getSupportedOptions https://docs.oracle.com/javase/8/docs/api/javax/annotation/processing/Processor.html#getSupportedOptions
Returns the options recognized by this processor. An implementation of the processing tool must provide a way to pass processor-specific options distinctly from options passed to the tool itself, see getOptions.
Implementation outline:
public Set<String> getSupportedOptions() {
Set<String> set = new HashSet<>();
set.add("release");
return set;
}
// -Arelease=true
boolean isRelease(ProcessingEnvironment env) {
return Boolean.parseBoolean(env.getOptions().get("release"));
}
See Pass options to JPAAnnotationProcessor from Gradle for how to pass options in a gradle build.

How can I find all paths to access an java API?

I'm analyzing an Android App, looking for security flaws. I've decompiled the APK with JEB and I found a vulnerable method in it.
My problem is: The App logic is too complex and it is very difficult to find a way to trigger this vulnerable method.
I would like to know if there exists a tool to find all the "paths" in the code to access some method.
For example, for the code below:
private void methodX() {
// This is the method I want to call
}
private void methodA() {
methodX();
}
private void methodB() {
methodA();
}
private void methodC() {
methodX();
}
The paths to access methodX are:
methodA( ) -> methodX( )
methodC( ) -> methodX( )
methodB( ) -> methodA( ) -> methodX( )
By the way, I'm using eclipse in the analysis, maybe there is some command on it to do this, but I haven't found yet.

In Eclipse, Ctrl+Alt+H will open the call hierarchy for a method, showing a tree view you can expand for finding "indirect" references to that method.
Here is an example tracing a method from Spring MVC's DispatcherServlet:

Ctrl+Shift+G on methodX() will show you all references

log all the attribute changes of a Class

For debugging purposes I need to keep track of a Class attributes changes.
For example consider the following class:
class Test {
int myInt;
String myString;
...
public void setMyInt(int a) ...
public void setMyString(String s) ...
public printDebugLog();
}
void main() {
Test t = new Test();
t.setMyInt(5);
t.setMyString("Hello");
t.printDebugLog();
}
I want to output to be something like:
myInt => 5
myString => Hello
The easy solution is to create logs instantly. i.e. adding a Log function as follow:
void Log(String s) {
System.out.println(s);
}
and then code the set functions like below:
void setMyString(String s) {
myString = s;
Log("myString => " + s);
}
this requires all the set functions to be written variously and I wonder if there are any better solution for such matter. For example it might be easier (if possible) to create a SetValue function which accepts two variables and set the first attribute to the value of the second object. or something like this.
Any idea?

To do this you should wrap your class with orthogonal code that performs logging.
Since your class does not implement interface you cannot use dynamic proxy, so you have to use one of solutions that use byte code engineering.
The strongest solution I know is AspectJ. But probably you even do not need it. You can use Javassist or CGLIb - the byte code engineering libraries that allow creating proxies that wrap classes, so you can add code that performs logging.

You can use AOP to intercept the setter methods and log when they are called. A quick google should give you a few examples.

If you debug via JPDA,
you can create a Breakpoint
on a field you like to watch.

Is there a way to export java functions to other packages within a library, but not to users of the library?

So, here's an example. I have a library in the package HTTP. I define sub-sections of the library in e.g. the package HTTP.TCPProtocol. Now I want to use TCPProtocol from the HTTP package, which means I have to make the TCPProtocol functionality public. At the same time, this functionality should not be exported to users of the library.
How do I do this? I don't want to shove my whole library into one package, as I feel the separate sub-packages really make the code more structured and navigation easier in eclipse. But browsing around, I couldn't find a method to expose functions within my project, but not export them outside my project.
EDIT: In light of me being able to come up with a better example, I'm updating the OP.

One simplistic approach is to whitelist your 'utility' methods so they take a caller instance of a certain type only.
package gameengine;
interface Whitelisted {} // marker
Then your method:
public void myMethod(Whitelisted caller, String arg)
And to invoke:
package gameengine.network;
class Foo implements Whitelisted {
...
Someclass.myMethod(this, "foo");

Check the caller's class to lock out all unwanted callers. The caller's class can be obtained from the stacktrace. In the example below, only instances of Bar will trigger the system.out.println, all all other will get an exception. You can even do package-level checks this way. Make sure that all allowed caller classes methods are not public, or they can call the doSomething method indirectly. You can even do deeper checks, by inspecting the stacktrace further.
Be aware though, that a skilled develper can circumvent anything you try do do in this matter. No solution is really "secure".
package one.two;
import one.Bar;
public class Foo {
public void doSomething() {
StackTraceElement[] stackTrace = Thread.currentThread().getStackTrace();
StackTraceElement stackTraceElement = stackTrace[2];
String className = stackTraceElement.getClassName();
if (Bar.class.getName().equals(className)) {
System.out.println("jay!");
} else {
throw new RuntimeException("not allowed");
}
}
}
package one;
import one.two.Foo;
public class Bar {
void makeCall() {
new Foo().doSomething();
}
public static void main(String[] args) {
new Bar().makeCall();
}
}

Without seeing your dependencies, the only real advice the community can give you is to refactor your code. If something in your networking package needs to know about your game engine, it seems like you have a leaky abstraction. Hard to say without seeing your code.

How to write automated unit tests for java annotation processor?

I'm experimenting with java annotation processors. I'm able to write integration tests using the "JavaCompiler" (in fact I'm using "hickory" at the moment). I can run the compile process and analyse the output. The Problem: a single test runs for about half a second even without any code in my annotation processor. This is way too long to using it in TDD style.
Mocking away the dependencies seems very hard for me (I would have to mock out the entire "javax.lang.model.element" package). Have someone succeed to write unit tests for an annotation processor (Java 6)? If not ... what would be your approach?

This is an old question, but it seems that the state of annotation processor testing hadn't gotten any better, so we released Compile Testing today. The best docs are in package-info.java, but the general idea is that there is a fluent API for testing compilation output when run with an annotation processor. For example,
ASSERT.about(javaSource())
.that(JavaFileObjects.forResource("HelloWorld.java"))
.processedWith(new MyAnnotationProcessor())
.compilesWithoutError()
.and().generatesSources(JavaFileObjects.forResource("GeneratedHelloWorld.java"));
tests that the processor generates a file that matches GeneratedHelloWorld.java (golden file on the class path). You can also test that the processor produces error output:
JavaFileObject fileObject = JavaFileObjects.forResource("HelloWorld.java");
ASSERT.about(javaSource())
.that(fileObject)
.processedWith(new NoHelloWorld())
.failsToCompile()
.withErrorContaining("No types named HelloWorld!").in(fileObject).onLine(23).atColumn(5);
This is obviously a lot simpler than mocking and unlike typical integration tests, all of the output is stored in memory.

You're right mocking the annotation processing API (with a mock library like easymock) is painful. I tried this approach and it broke down pretty rapidly. You have to setup to many method call expectations. The tests become unmaintainable.
A state-based test approach worked for me reasonably well. I had to implement the parts of the javax.lang.model.* API I needed for my tests. (That were only < 350 lines of code.)
This is the part of a test to initiate the javax.lang.model objects. After the setup the model should be in the same state as the Java compiler implementation.
DeclaredType typeArgument = declaredType(classElement("returnTypeName"));
DeclaredType validReturnType = declaredType(interfaceElement(GENERATOR_TYPE_NAME), typeArgument);
TypeParameterElement typeParameter = typeParameterElement();
ExecutableElement methodExecutableElement = Model.methodExecutableElement(name, validReturnType, typeParameter);
The static factory methods are defined in the class Model implementing the javax.lang.model.* classes. For example declaredType. (All unsupported operations will throw exceptions.)
public static DeclaredType declaredType(final Element element, final TypeMirror... argumentTypes) {
return new DeclaredType(){
#Override public Element asElement() {
return element;
}
#Override public List<? extends TypeMirror> getTypeArguments() {
return Arrays.asList(argumentTypes);
}
#Override public String toString() {
return format("DeclareTypeModel[element=%s, argumentTypes=%s]",
element, Arrays.toString(argumentTypes));
}
#Override public <R, P> R accept(TypeVisitor<R, P> v, P p) {
return v.visitDeclared(this, p);
}
#Override public boolean equals(Object obj) { throw new UnsupportedOperationException(); }
#Override public int hashCode() { throw new UnsupportedOperationException(); }
#Override public TypeKind getKind() { throw new UnsupportedOperationException(); }
#Override public TypeMirror getEnclosingType() { throw new UnsupportedOperationException(); }
};
}
The rest of the test verifies the behavior of the class under test.
Method actual = new Method(environment(), methodExecutableElement);
Method expected = new Method(..);
assertEquals(expected, actual);
You can have a look at the source code of the Quickcheck #Samples and #Iterables source code generator tests. (The code is not optimal, yet. The Method class has to many parameters and the Parameter class is not tested in its own test but as part of the Method test. It should illustrate the approach nevertheless.)
Viel Glück!

jOOR is a small Java reflection library that also provides simplified access to the in-memory Java compilation API in javax.tool.JavaCompiler. We added support for this to unit test jOOQ's annotation processors. You can easily write unit tests like this:
#Test
public void testCompileWithAnnotationProcessors() {
AProcessor p = new AProcessor();
try {
Reflect.compile(
"org.joor.test.FailAnnotationProcessing",
"package org.joor.test; " +
"#A " +
"public class FailAnnotationProcessing { " +
"}",
new CompileOptions().processors(p)
).create().get();
Assert.fail();
}
catch (ReflectException expected) {
assertFalse(p.processed);
}
}
The above example has been taken from this blog post

I was in a similar situation, so I created the Avatar library. It won't give you the performance of a pure unit test with no compilation, but if used correctly you shouldn't see much of a performance hit.
Avatar lets you write a source file, annotate it, and convert it to elements in a unit test. This allows you to unit test methods and classes which consume Element objects, without manually invoking javac.

I ran into the same problem awhile ago and found this question. Although the other answers provided are decent, I felt that that there was still room for improvement. Based on the other answers for this question, I created Elementary, a suite of JUnit 5 extensions that provide a real annotation processing environment for unit tests.
Most libraries test annotation processors by running them. However, most annotation processors are pretty complex and broken into more fine-grained components. It is not feasible to test individual components by running the annotation processor. Instead, we make the annotation processing environment available to these tests.
The following code snippet illustrates how to test a Lint component:
import com.karuslabs.elementary.junit.Cases;
import com.karuslabs.elementary.junit.Tools;
import com.karuslabs.elementary.junit.ToolsExtension;
import com.karuslabs.elementary.junit.annotations.Case;
import com.karuslabs.elementary.junit.annotations.Introspect;
import com.karuslabs.utilitary.type.TypeMirrors;
#ExtendWith(ToolsExtension.class)
#Introspect
class ToolsExtensionExampleTest {
Lint lint = new Lint(Tools.typeMirrors());
#Test
void lint_string_variable(Cases cases) {
var first = cases.one("first");
assertTrue(lint.lint(first));
}
#Test
void lint_method_that_returns_string(Cases cases) {
var second = cases.get(1);
assertFalse(lint.lint(second));
}
#Case("first") String first;
#Case String second() { return "";}
}
class Lint {
final TypeMirrors types;
final TypeMirror expectedType;
Lint(TypeMirrors types) {
this.types = types;
this.expectedType = types.type(String.class);
}
public boolean lint(Element element) {
if (!(element instanceof VariableElement)) {
return false;
}
var variable = (VariableElement) element;
return types.isSameType(expectedType, variable.asType());
}
}
By annotating the test class with #Introspect and test cases with #Case, we can declare test cases in the same file as the tests. The corresponding Element representation of the test cases can be retrieved by a test using Cases.
If anyone is interested, I wrote an article, The Problem with Annotation Processors that details the problems with unit testing annotation processors.

I have used http://hg.netbeans.org/core-main/raw-file/default/openide.util.lookup/test/unit/src/org/openide/util/test/AnnotationProcessorTestUtils.java though this is based on java.io.File for simplicity and so has the performance overhead you complain about.
Thomas's suggestion of mocking the whole JSR 269 environment would lead to a pure unit test. You might instead want to write more of an integration test which checks how your processor actually runs inside javac, giving more assurance it is correct, but merely want to avoid disk files. Doing this would require you to write a mock JavaFileManager, which is unfortunately not as easy as it seems and I have no examples handy, but you should not need to mock other things like Element interfaces.

An option is to bundle all tests in one class. Half a second for compiling etc. is then a constant for a given set of tests, the real test time for a test is negligible, I assume.