学习Java编译器API，为什么trees.getElement（treepath）返回null？

Question

I'm trying to parse a java file with Java Compiler APIs.

The documents are very poor. After hours of digging I still cannot get the Trees#getElement work for me. Here's my code:

import com.sun.source.tree.*;
import com.sun.source.util.*;

import javax.tools.JavaCompiler;
import javax.tools.JavaFileObject;
import javax.tools.StandardJavaFileManager;
import javax.tools.ToolProvider;
import java.io.IOException;
import java.nio.CharBuffer;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;

class CodeAnalyzerTreeVisitor extends TreePathScanner<Object, Trees> {
    @Override
    public Object visitClass(ClassTree classTree, Trees trees) {
        System.out.println("className " + classTree.getSimpleName());
        //prints name of class
        TreePath path = getCurrentPath();

        printLocationAndSource(trees, path, classTree);
        //prints the original source code

  while (path != null) { System.out.println("treepath"); System.out.println(trees.getElement(path)); path = path.getParentPath(); } //it prints several nulls here //why? 

        return super.visitClass(classTree, trees);
    }

    public static void printLocationAndSource(Trees trees,
                                              TreePath path, Tree tree) {
        SourcePositions sourcePosition = trees.getSourcePositions();

        long startPosition = sourcePosition.
                getStartPosition(path.getCompilationUnit(), tree);
        long endPosition = sourcePosition.
                getEndPosition(path.getCompilationUnit(), tree);

        JavaFileObject file = path.getCompilationUnit().getSourceFile();
        CharBuffer sourceContent = null;
        try {
            sourceContent = CharBuffer.wrap(file.getCharContent(true).toString().toCharArray());
        } catch (IOException e) {
            e.printStackTrace();
        }
        CharBuffer relatedSource = null;
        if (sourceContent != null) {
            relatedSource = sourceContent.subSequence((int) startPosition, (int) endPosition);
        }

        System.out.println("start: " + startPosition + " end: " + endPosition);
//        System.out.println("source: "+relatedSource);
        System.out.println();
    }
}

public class JavaParser {
    private static final JavaCompiler javac
            = ToolProvider.getSystemJavaCompiler();
    private static final String filePath = "/home/pinyin/Source/hadoop-common/hadoop-yarn-project/hadoop-ya" +
            "rn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/ya" +
            "rn/server/resourcemanager/ResourceManager.java";

    public static void main(String[] args) throws IOException {
        StandardJavaFileManager jfm = javac.getStandardFileManager(null, null, null);
        Iterable<? extends javax.tools.JavaFileObject> javaFileObjects = jfm.getJavaFileObjects(filePath);
        String[] sourcePathParam = {
                "-sourcepath",
                "/home/pinyin/Source/hadoop-common/hadoop-yarn-project/hadoop-yarn/" +
                        "hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/"
        };
        List<String> params = new ArrayList<String>();
        params.addAll(Arrays.asList(sourcePathParam));

        JavacTask task = (JavacTask) javac.getTask(null, jfm, null, params, null, javaFileObjects);

        Iterable<? extends CompilationUnitTree> asts = task.parse();
        Trees trees = Trees.instance(task);

        for (CompilationUnitTree ast : asts) {
            new CodeAnalyzerTreeVisitor().scan(ast, trees);
        }
    }
}

The lines about params and -sourcepath are added because I thought the compiler is trying to find the source file in the wrong places. They didn't work.

I'm still trying to understand how the Trees , javac and related JSRs work together, are there any recommended documents for beginners? Thanks for your help.

edit: The java file I'm trying to analyze is: https://github.com/apache/hadoop-common/blob/trunk/hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-resourcemanager/src/main/java/org/apache/hadoop/yarn/server/resourcemanager/ResourceManager.java

The file can be compiled without errors in its maven project, but its dependencies are not passed to javac in my situation. I'm not sure if this is the problem.

The trees.getElement returns null in the middle part of the code above, while the other parts seems to work well.

Answer 1

According to this answer , it seems that the Element s' information is not usable until the compilation is completed.

So calling task.analyze() solved my problem. Although javac is complaining about missing dependencies.

Please correct me if I'm wrong, thanks.