Is that possible to give custom warning message in my own API like below? Is Resource leak:'ois' is never closed message related with Java API or JVM?
It's possible using a compiler API. You have to extend an AbstractProcessor and then make sure compiler knows about it.
Lets say we don't like programmers to swear in the source code. So, when someone defines a field with name "shit", we want to show a warning. Here is a simple implementation:
import java.util.List;
import java.util.Set;
import javax.annotation.processing.AbstractProcessor;
import javax.annotation.processing.RoundEnvironment;
import javax.annotation.processing.SupportedAnnotationTypes;
import javax.annotation.processing.SupportedSourceVersion;
import javax.lang.model.SourceVersion;
import javax.lang.model.element.Element;
import javax.lang.model.element.ElementKind;
import javax.lang.model.element.TypeElement;
import javax.tools.Diagnostic.Kind;
#SupportedSourceVersion(SourceVersion.RELEASE_7)
#SupportedAnnotationTypes("*")
public class Test extends AbstractProcessor {
public int shit;
public int foo;
#Override
public boolean process(Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
Set<? extends Element> rootElements = roundEnv.getRootElements();
for (Element element : rootElements) {
if (element.getKind() == ElementKind.CLASS) {
List<? extends Element> classElements = element.getEnclosedElements();
for (Element classElement : classElements) {
if (classElement.getKind() == ElementKind.FIELD) {
if (classElement.getSimpleName().contentEquals("shit")) {
processingEnv.getMessager().printMessage(
Kind.WARNING,
"How dare you to swear in the source code?!",
classElement
);
}
}
}
}
}
return false;
}
public static void main(String[] args) {
//
}
}
Now, we want to apply such a processor just for this very class, because there is an ugly bad-called field too.
Using a command line:
javac Test.java
javac -processor Test Test.java
We need to firstly build a processor and then apply it while compiling (in this case to the same file).
And this is the output we get:
Test.java:17: warning: How dare you to swear in the source code?!
public int shit;
^
1 warning
To have the same warning in Eclipse or any other IDE, it's necessary to change compiler settings so it uses this custom processor.
Update: In the comments, kapep sent a link on how to set a custom processor in Eclipse: http://kerebus.com/2011/02/using-java-6-processors-in-eclipse/
Just for the record: Exactly the same warning may be achieved by implementing interface Closeable:
import java.io.Closeable;
import java.io.IOException;
public class Test implements Closeable {
#Override
public void close() throws IOException {
// TODO Auto-generated method stub
}
public static void main(String[] args) {
new Test();
}
}
And you see the same warning:
You can create warnings, notes, errors and other diagnostic messages like this using an annotation processor. It's a compiler plugin api integrated in the JDK. It lets you analyse the outline structure of source code. Despite the name you don't really need to handle any annotation when processing code. Messages are created using the Messager class. If you provide an element, it will be marked and the message will be shown next to it in the source code editor.
You won't be able to show message on elements inside methods or expressions though, only on declarations like types, properties, methods or parameters. It's possible to additionally parse the method body and generate messages based on the content using other tools, but as far as I know you can't show the message on the actual local element then. You could still show the message on the enclosing method or don't specify any element at all and show it in the IDE's log.
The IDE also needs to support this. I know that Eclipse and NetBeans do support messages generated by annotation processors, but there are probably other modern IDE that do so as well. If you need more features like messages on elements inside method bodies or the quick fix feature as shown in the example, I guess you need to create a plugin for the IDE.
I would believe that it is related to the eclipse ide, you could possibly write a plugin which displays warnings like that.
For example, when you use a method which has the annotation '#Deprecated' the ide automatically tells the programmer that the method is deprecated.
Related
Given the following java source:
import org.apache.commons.lang3.StringUtils;
public class Example {
public static void main(String[] args) {
StringUtils.capitalize("hello world!");
}
}
Is there a way to get the full-qualified-name of "StringUtils" without methodbinding/typebinding, which is not available when having only the source code available without proper environment settings including the commons-lang.jar.
The information could be obtained by incorporating the import statements manually, I guess, but I'm wondering if there's something like this builtin into the JDT Parser/AST.
Here is my class:
package pepelu;
import pepelu.ImportTest.InnerClass.InnerEnum;
import javax.annotation.Resource;
public class ImportTest {
#Resource
public static class InnerClass {
public enum InnerEnum {
A
}
}
public static void main(String[] args) {
System.out.println(InnerEnum.A);
}
}
When I use maven to build, it will give a compilation error:
mvn clean compile
[ERROR] /Users/finup/Desktop/a/importtest/src/main/java/pepelu/ImportTest.java:[8,6] cannot find symbol
After changing the import order to:
import javax.annotation.Resource;
import pepelu.ImportTest.InnerClass.InnerEnum;
I got a successful maven build.
I searched for documents, but cannot find an explain for this.
Could anyone please explain how import works in this case?
I guess the reason is a "circular" dependency: you have some element X that you import within the same file/class where you are defining it.
Meaning:
import pepelu.ImportTest.InnerClass.InnerEnum;
actually refers to code following in the very same file:
public static class InnerClass {
public enum InnerEnum {
This means: for the compiler, in order to process that import, it has to look into the body of the class in the same file.
It seems that javac does that "immediately". Meaning: it starts reading import statements, and importing from the same class makes it "suspend" looking at imports, but checking out the following class definition.
And guess what: that class definition makes use of another import. In order to "process" the definition of that enum, the compiler needs to understand where/what that #Resource annotation is about. But it doesn't know the annotation yet (because the compiler didn't see the import, yet).
When you change the order, the compiler understands that #Resource usage in the class definition.
And of course: the real answer is not to reorder imports. The real answer is to not import something from the class that is following after the import statements. There is absolutely no point in doing so.
Edit, given the comment by the OP about how this can work in Redisson: honestly, I don't know. It might depend on how exactly that class is compiled. Maybe such code works with newer (or older) versions of javac, maybe this works with specific versions of the eclipse or intellij or xyz compiler.
Meaning: I gave you an explanation why you are running into this problem. That doesn't mean that any compiler must necessarily run into the same problem.
I want to create an api using my own custom annotation that the hidden code should be triggered.
I have created my annotation and have created the processor as well.
But now the problem is, I don't know how to build it.
Let me explain in better way:
Its a console applicatioyn, I have to print a text once a method is called.
So, I have created an annotation #PrintText and also created a PrintTextProcessor.
But when I try to compile it, it doesn't show the required output.
I am annotating a method.
But it looks annotation doesn't work.
Am I missing anything.
Following is my code
Annotation Class:
#Target(ElementType.METHOD)
#Retention(RetentionPolicy.RUNTIME)
public #interface PrintText{
}
Annotation Processor Class:
#SupportedAnnotationTypes("com.example.PrintText")
public class PrintTextProcessor extends AbstractProcessor {
#Override
public boolean process(Set<? extends TypeElement> annotations,
RoundEnvironment roundEnv) {
Set<? extends Element> elements = roundEnv.getElementsAnnotatedWith(PrintText.class);
for(Element e : elements){
if(!e.getClass().equals(ParticularType.class)){
processingEnv.getMessager().printMessage(Kind.ERROR,
"#PrintText annotated fields must be of type ParticularType");
}
}
return true;
}
}
Now my main class comes:
public class Main{
#PrintMe
public void testMethod(){
System.out.println("In test method");
}
public static void main(String s[]){
new Main().testMethod();
}
}
Now when I try to compile this program and run it, it only prints the following text:
In test method
I used following commands
javac Main.java
java Main
Did I miss something?
I have been gone through several posts on the internet and found that there is apt tool.
But I don't know how to build and run it via command line.
I am using java6.
Thanks in advance.
Here is very good example https://github.com/provegard/aptdemo
but basically you have to create package META-INF/services
and put javax.annotation.processing.Processor file with classpath to your processor (in other words register your processor), then build your app let's say with mvn to get jar file (mvn package, see example) and then compile with javac (javac -cp /path/to/aptdemo-1.0-SNAPSHOT.jar SomeTestClass.java)
I thought I would use the new ResourceBundleControlProvider framework in Java 8 to fix something which Oracle themselves will never fix - the default encoding used when reading resource bundles.
So I made a control:
package com.acme.resources;
import java.io.IOException;
import java.util.Locale;
import java.util.ResourceBundle;
public class AcmeResourceBundleControl extends ResourceBundle.Control
{
#Override
public ResourceBundle newBundle(String baseName, Locale locale, String format,
ClassLoader loader, boolean reload)
throws IllegalAccessException, InstantiationException, IOException
{
throw new UnsupportedOperationException("TODO");
}
}
Then I made a provider:
package com.acme.resources;
import java.util.ResourceBundle;
import java.util.spi.ResourceBundleControlProvider;
public class AcmeResourceBundleControlProvider implements ResourceBundleControlProvider
{
private static final ResourceBundle.Control CONTROL = new AcmeResourceBundleControl();
#Override
public ResourceBundle.Control getControl(String baseName)
{
if (baseName.startsWith("com.acme."))
{
return CONTROL;
}
else
{
return null;
}
}
}
Then in META-INF/services/java.util.spi.ResourceBundleControlProvider:
com.acme.resources.AcmeResourceBundleControlProvider
Then I just tried to run our application from IDEA and I find that it never loads my provider (otherwise the exception would be raised.)
I have checked the names and they all seem to match up. I have checked the compiler output directory IDEA is using and it does contain the service file. I wrote a simple test program which just tries to look up the service:
public static void main(String[] args)
{
for (ResourceBundleControlProvider provider :
ServiceLoader.load(ResourceBundleControlProvider.class))
{
System.out.println(provider.getClass());
}
}
This does print out one entry which is the name of my implementation class. So the issue is not in the service file.
If I breakpoint inside ResourceBundle, I seem to be able to access the custom provider class. Initial forays into the debugger show that ServiceLoader isn't finding any implementations, but I can't figure out why. I'm sure there is some dodgy class loader magic going on which results in not loading my class. :(
Some scary documentation on the Javadoc makes it sound like it might have to be installed as a global extension. If that really is the case, it's a bit of a shame, because it seemed like a useful way to override the default (and in my opinion broken) behaviour. But I also read the tutorial on the matter and it didn't seem to be describing anything like that (unless the good behaviour was pulled out of Java 8 at the very last minute and the docs are out of date!)
The tutorial does state that the JAR containing the ResourceBundleControlProvider must be in the JVM's system extension directory. Section 6 of the tutorial describes the requirement:
java -Djava.ext.dirs=lib -cp build RBCPTest
When you install a Java extension, you typically put the JAR file of the extension in the lib/ext directory of your JRE. However, this command specifies the directory that contains Java extensions with the system property java.ext.dirs.
The JavaDoc for ServiceLoader.loadInstalled() also states that providers on the application's class path are ignored.
Your problem is that the java.util.ResourceBundle that comes with the JVM does a ServiceLoader.loadInstalled(ResourceBundleControlProvider.class) to obtain a list of providers in the static initializer, and uses the thus obtained list ever after.
I have an annotation that can be added on METHOD and TYPE and is used in thousands of places in our project.
#Retention(java.lang.annotation.RetentionPolicy.RUNTIME)
#Target({METHOD, TYPE})
#Inherited
public #interface RequiredStore{
Store value();
}
Is it possible to make the annotation deprecated only on methods while keeping it non-deprecated on types? I want other developers to be notified by IDE that it should not be used on methods any more, until we'll refactor all existing usages and finally remove the METHOD part.
If it's not possible, is there any Way to handle such case beside creating new annotation only for types and deprecating the old one?
You could use an annotation Processor.
For example, the annotation and its processor would be placed in its own .jar file and added as a dependency of the sources that use the annotation.
The custom .jar would have the following structure:
src/main/
java/com/company/annotations/
RequiredStore.java
RequiredStoreProcessor.java
resources/META-INF/services
javax.annotation.processing.Processor
RequiredStore.java stays as you have it above.
RequiredStoreProcessor.java could look something like this:
package com.company.annotations;
import java.util.Set;
import javax.annotation.processing.AbstractProcessor;
import javax.annotation.processing.RoundEnvironment;
import javax.annotation.processing.SupportedAnnotationTypes;
import javax.lang.model.SourceVersion;
import javax.lang.model.element.Element;
import javax.lang.model.element.ElementKind;
import javax.lang.model.element.TypeElement;
import javax.tools.Diagnostic;
#SupportedAnnotationTypes("com.company.annotations.RequiredStore")
public class RequiredStoreProcessor extends AbstractProcessor {
#Override
public boolean process(
Set<? extends TypeElement> annotations,
RoundEnvironment roundEnv) {
for (Element element
: roundEnv.getElementsAnnotatedWith(RequiredStore.class)) {
if (element.getKind().equals(ElementKind.METHOD)) {
processingEnv.getMessager().printMessage(
Diagnostic.Kind.WARNING,
"Using #RequiredStore on methods has been deprecated\n"
+ "Class: " + element.getEnclosingElement() + "\n"
+ "Method: " + element.getSimpleName() + "\n");
}
}
// Other processing...
return false;
}
#Override
public SourceVersion getSupportedSourceVersion() {
return SourceVersion.latest();
}
}
The javax.annotation.processing.Processor file allows javac to pickup the Processor via SPI and simply contains
com.company.annotations.RequiredStoreProcessor
Finally, compile this into a .jar and add it to the classpath where the annotations are being used. Any methods that have the #RequiredStore will produce a compiler warning. For example, for this class,
package com.company.business;
import com.company.annotations.RequiredStore;
#RequiredStore
public interface Business {
#RequiredStore
public void someMethod();
}
The compiler warning would be this:
warning: Using #RequiredStore on methods has been deprecated
Class: com.company.business.Business
Method: someMethod
As for an indication in the IDE, you might have to write a custom inspection and unfortunately this depends on the IDE used.
Notes:
Decent custom annotations reference: Code Generation using Annotation Processors in the Java language
If you are okay about using native aspectj, another option is to use AspectJ's code enforcement policy this way:
public aspect RequiredStoreAnnotationCheck {
declare warning: execution(#RequiredStore * *.*(..)) : "Required store annotation not appropriate for methods..";
}
If the IDE is integrated with AspectJ, this would be flagged as a compile time check.
AspectJ in action book has a good amount of detail on this too.
Here is one of my blog articles for more context: http://www.java-allandsundry.com/2012/03/code-policy-enforcement-using-aspectj.html