Class文件详解 (1)

作者：asklxf 文章来源：http://www.j2medev.com/Article/ShowArticle.asp?ArticleID=105

我们都知道，Java编译器负责将.java文件编译成.class文件，class文件存储的是java字节码，与.java文件无关（只要你愿意写一个编译器，也可以将别的语言写的源代码编译成.class文件），本文准备详细解剖class文件的内部结构，并且把class文件结构读取并显示出来。

Class文件的格式由JVM规范规定，一共有以下部分：
1. magic number，必须是0xCAFEBABE，用于快速识别是否是一个class文件。
2. version，包括major和minor，如果版本号超过了JVM的识别范围，JVM将拒绝执行。
3. constant pool，常量池，存放所有用到的常量。
4. access flag，定义类的访问权限。
5. this class和super class，指示如何找到this class和super class。
6. interfaces，存放所有interfaces。
7. fields，存放所有fields。
8. methods，存放所有methods。
9. attributes，存放所有attributes。

先写一个Test.java：

package example.test;

public final class TestClass {
public int id = 1234567;
public void test() {}
}

然后编译，放在C:\example\test\Test.class。

我们用Java来读取和分析class，ClassAnalyzer的功能便是读取Test.class，分析结构，然后显示出来：

package classfile.format;

import java.io.*;

public class ClassAnalyzer {

public static void main(String[] args) {
DataInputStream input = null;
try {
input = new DataInputStream(new BufferedInputStream(new FileInputStream(
"C:\\example\\test\\TestClass.class"
)));
analyze(input);
}
catch(Exception e) {
System.out.println("Analyze failed!");
}
finally {
try { input.close(); } catch(Exception e) {}
}
}

public static void analyze(DataInputStream input) throws IOException {
// read magic number:
int magic = input.readInt();
if(magic==0xCAFEBABE)
System.out.println("magic number = 0xCAFEBABE");
else
throw new RuntimeException("Invalid magic number!");
// read minor version and major version:
short minor_ver = input.readShort();
short major_ver = input.readShort();
System.out.println("Version = " + major_ver + "." + minor_ver);
// read constant pool:
short const_pool_count = input.readShort();
System.out.println("constant pool size = " + const_pool_count);
// read each constant:
for(int i=1; i<const_pool_count; i++) {
analyzeConstant(input, i);
}
}

public static void analyzeConstant(DataInputStream input, int index) throws IOException {
byte flag = input.readByte();
// for read:
byte n8;
short n16;
int n32;
long n64;
float f;
double d;
byte[] buffer;
System.out.println("\nconst index = " + index + ", flag = " + (int)flag);
switch(flag) {
case 1: // utf-8 string
System.out.println(" const type = Utf8");
n16 = input.readShort();
System.out.println(" length = " + n16);
buffer = new byte[n16];
input.readFully(buffer);
System.out.println(" value = " + new String(buffer));
break;
case 3: // integer
System.out.println(" const type = Integer");
n32 = input.readInt();
System.out.println(" value = " + n32);
break;
case 4: // float
System.out.println(" const type = Float");
f = input.readFloat();
System.out.println(" value = " + f);
break;
case 5: // long
System.out.println(" const type = Long");
n64 = input.readLong();
System.out.println(" value = " + n64);
break;
case 6: // double
System.out.println(" const type = Double");
d = input.readDouble();
System.out.println(" value = " + d);
break;
case 7: // class or interface reference
System.out.println(" const type = Class");
n16 = input.readShort();
System.out.println(" index = " + n16 + " (where to find the class name)");
break;
case 8: // string
System.out.println(" const type = String");
n16 = input.readShort();
System.out.println(" index = " + n16);
break;
case 9: // field reference
System.out.println(" const type = Fieldref");
n16 = input.readShort();
System.out.println("class index = " + n16 + " (where to find the class)");
n16 = input.readShort();
System.out.println("nameAndType = " + n16 + " (where to find the NameAndType)");
break;
case 10: // method reference
System.out.println(" const type = Methodref");
n16 = input.readShort();
System.out.println("class index = " + n16 + " (where to find the class)");
n16 = input.readShort();
System.out.println("nameAndType = " + n16 + " (where to find the NameAndType)");
break;
case 11: // interface method reference
System.out.println(" const type = InterfaceMethodref");
n16 = input.readShort();
System.out.println("class index = " + n16 + " (where to find the interface)");
n16 = input.readShort();
System.out.println("nameAndType = " + n16 + " (where to find the NameAndType)");
break;
case 12: // name and type reference
System.out.println(" const type = NameAndType");
n16 = input.readShort();
System.out.println(" name index = " + n16 + " (where to find the name)");
n16 = input.readShort();
System.out.println(" descripter = " + n16 + " (where to find the descriptor)");
break;
default:
throw new RuntimeException("Invalid constant pool flag: " + flag);
}
}
}

输出结果为：

magic number = 0xCAFEBABE
Version = 48.0
constant pool size = 22

const index = 1, flag = 1
const type = Utf8
length = 22
value = example/test/TestClass

const index = 2, flag = 7
const type = Class
index = 1 (where to find the class name)

const index = 3, flag = 1
const type = Utf8
length = 16
value = java/lang/Object

const index = 4, flag = 7
const type = Class
index = 3 (where to find the class name)

const index = 5, flag = 1
const type = Utf8
length = 2
value = id

const index = 6, flag = 1
const type = Utf8
length = 1
value = I

const index = 7, flag = 1
const type = Utf8
length = 6
value = <init>

const index = 8, flag = 1
const type = Utf8
length = 3
value = ()V

const index = 9, flag = 1
const type = Utf8
length = 4
value = Code

const index = 10, flag = 12
const type = NameAndType
name index = 7 (where to find the name)
descripter = 8 (where to find the descriptor)

const index = 11, flag = 10
const type = Methodref
class index = 4 (where to find the class)
nameAndType = 10 (where to find the NameAndType)

const index = 12, flag = 3
const type = Integer
value = 1234567

const index = 13, flag = 12
const type = NameAndType
name index = 5 (where to find the name)
descripter = 6 (where to find the descriptor)

const index = 14, flag = 9
const type = Fieldref
class index = 2 (where to find the class)
nameAndType = 13 (where to find the NameAndType)

const index = 15, flag = 1
const type = Utf8
length = 15
value = LineNumberTable

const index = 16, flag = 1
const type = Utf8
length = 18
value = LocalVariableTable

const index = 17, flag = 1
const type = Utf8
length = 4
value = this

const index = 18, flag = 1
const type = Utf8
length = 24
value = Lexample/test/TestClass;

const index = 19, flag = 1
const type = Utf8
length = 4
value = test

const index = 20, flag = 1
const type = Utf8
length = 10
value = SourceFile

const index = 21, flag = 1
const type = Utf8
length = 14
value = TestClass.java

我们暂时只读取到constant pool，后面的信息下次再继续 :)

参考：《Inside the Java Virture Mechine》