草庐IT

Class文件解析

carry1899 2023-04-16 原文

思考

编写-编译-运行。java文件编译后生成class文件,jvm是如何加载class文件?

1 准备工作

获取class文件byte[]

 public static byte[] getFileBytes(File file) {
        try (FileInputStream fileInputStream = new FileInputStream(file)) {
            int available = fileInputStream.available();
            byte[] data=new byte[available];
            fileInputStream.read(data,0,available-1);
            return data;
        } catch (Exception e) {
            e.printStackTrace();
        }
        return null;
    }

这里使用jdk的ByteBuffer包装bytes

ByteBuffer data = ByteBuffer.wrap(getFileBytes(file));

因为ByteBuffer没有无符号的读取方法,所以自己实现一下,也可以直接用netty的Bytebuf,里面方法齐全

    // 图方便直接返回int
    private int readUnsignedByte(ByteBuffer data) {
        return data.get() & 0xff;
    }

    // 图方便直接返回int
    private int readUnsignedShort(ByteBuffer data) {
        return data.getShort() & 0xffff;
    }
    
    private long readUnsignedInt(ByteBuffer data) {
        return data.getInt() & 0xffffffffL;
    }

定义class文件结构

参考: https://docs.oracle.com/javase/specs/jvms/se8/html/jvms-4.html

    private static class ClassFileStructure {
        long magic;
        int minorVersion;
        int majorVersion;
        int constantPoolCount;
        ConstantPool[] constantPool;
        int accessFlags;
        int thisClass;
        int superClass;
        int interfacesCount;
        int[] interfaces;
        int fieldsCount;
        FieldInfo[] fields;
        int methodsCount;
        MethodInfo[] methods;
        int attributesCount;
        AttributeInfo[] attributes;
    }

2 开始解析

2.1 magic

    private void magic(ClassFileStructure structure, ByteBuffer data) {
        structure.magic = readUnsignedInt(data);
    }

2.2 minorVersion

    private void minorVersion(ClassFileStructure structure, ByteBuffer data) {
        structure.minorVersion = readUnsignedShort(data);
    }

2.3 majorVersion

    private void majorVersion(ClassFileStructure structure, ByteBuffer data) {
        structure.majorVersion = readUnsignedShort(data);
    }

2.4 constantPoolCount

   private void constantPoolCount(ClassFileStructure structure, ByteBuffer data) {
        structure.constantPoolCount = readUnsignedShort(data);
   }

2.5 ConstantPool[]

ConstantPool不同tag解析方式不同,定义抽象类ConstantPool,子类按规则解析

    private abstract static class ConstantPool {
    int tag;

    public ConstantPool(int tag) {
        this.tag = tag;
    }

    abstract void parse(ByteBuffer data);
}

子类实现ConstantPool
ConstantUtf8:

 private class ConstantUtf8 extends ConstantPool {
    int length;
    byte[] bytes;

    public ConstantUtf8(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.length = readUnsignedShort(data);
        bytes = new byte[this.length];
        for (int i = 0; i < this.length; i++) {
            bytes[i] = (byte) readUnsignedByte(data);
        }
    }
}

ConstantMethodHandle:

 private class ConstantMethodHandle extends ConstantPool {
    short referenceKind;
    int referenceIndex;

    public ConstantMethodHandle(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.referenceKind = (short) readUnsignedByte(data);
        this.referenceIndex = readUnsignedShort(data);
    }
}

ConstantMethodType:

 private class ConstantMethodType extends ConstantPool {
    int descriptorIndex;

    public ConstantMethodType(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.descriptorIndex = readUnsignedShort(data);
    }
}

ConstantClass:

private class ConstantClass extends ConstantPool {
    int nameIndex;

    public ConstantClass(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.nameIndex = readUnsignedShort(data);
    }
}

ConstantClass:

   private class ConstantClass extends ConstantPool {
    int nameIndex;

    public ConstantClass(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.nameIndex = readUnsignedShort(data);
    }
}

ConstantFieldref:


private class ConstantFieldref extends ConstantPool {
    int classIndex;
    int nameAndTypeIndex;

    public ConstantFieldref(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.classIndex = readUnsignedShort(data);
        this.nameAndTypeIndex = readUnsignedShort(data);
    }
}

ConstantMethodref:

private class ConstantMethodref extends ConstantPool {
    int classIndex;
    int nameAndTypeIndex;

    public ConstantMethodref(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.classIndex = readUnsignedShort(data);
        this.nameAndTypeIndex = readUnsignedShort(data);
    }
}

ConstantInterfaceMethodref:

 private class ConstantInterfaceMethodref extends ConstantPool {
    int classIndex;
    int nameAndTypeIndex;

    public ConstantInterfaceMethodref(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.classIndex = readUnsignedShort(data);
        this.nameAndTypeIndex = readUnsignedShort(data);
    }
}

ConstantString:

 private class ConstantString extends ConstantPool {
    int stringIndex;

    public ConstantString(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.stringIndex = readUnsignedShort(data);
    }
}

ConstantInteger:

private class ConstantInteger extends ConstantPool {
    long bytes;

    public ConstantInteger(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.bytes = readUnsignedInt(data);
    }
}

ConstantFloat:

private class ConstantFloat extends ConstantPool {
    long bytes;

    public ConstantFloat(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.bytes = readUnsignedInt(data);
    }
}


ConstantLong:

private class ConstantLong extends ConstantPool {
    long highBytes;
    long lowBytes;

    public ConstantLong(int tag) {
        super(tag);
    }

    @Override
    void parse(ByteBuffer data) {
        this.highBytes = readUnsignedInt(data);
        this.lowBytes = readUnsignedInt(data);
    }
}

ConstantDouble:

   private class ConstantDouble extends ConstantPool {
        long highBytes;
        long lowBytes;
        public ConstantDouble(int tag) {
            super(tag);
        }
        @Override
        void parse(ByteBuffer data) {
            this.highBytes = readUnsignedInt(data);
            this.lowBytes = readUnsignedInt(data);
        }
    }

ConstantNameAndType:

  private class ConstantNameAndType extends ConstantPool {
        int nameIndex;
        int descriptorIndex;
        public ConstantNameAndType(int tag) {
            super(tag);
        }
        @Override
        void parse(ByteBuffer data) {
            this.nameIndex = readUnsignedShort(data);
            this.descriptorIndex = readUnsignedShort(data);
        }
    }

ConstantInvokeDynamic:

  private class ConstantInvokeDynamic extends ConstantPool {
        int bootstrapMethodAttrIndex;
        int nameAndTypeIndex;
        public ConstantInvokeDynamic(int tag) {
            super(tag);
        }
        @Override
        void parse(ByteBuffer data) {
            this.bootstrapMethodAttrIndex = readUnsignedShort(data);
            this.nameAndTypeIndex = readUnsignedShort(data);
        }
    }

以上是所有 constantPool 子类型,现在开始解析

坑:“所有 8 字节常量都占用文件constant_pool表中的两个条目class。 如果
CONSTANT_Long_infoorCONSTANT_Double_info结构是constant_pool表中索引n处的项目,则池中的下一个可用项目位于索引n +2 处。
constant_pool 索引n +1 必须有效但被视为不可用 。 回想起来,让 8 字节常量占用两个常量池条目是一个糟糕的选择。”

 private void constantPool(ClassFileStructure structure, ByteBuffer data) {
        structure.constantPool = new ConstantPool[structure.constantPoolCount - 1];
        for (int i = 0; i < structure.constantPool.length; i++) {
            int type = readUnsignedByte(data);
            int index = i;
            switch (type) {
                case 1: //
                    structure.constantPool[i] = new ConstantUtf8(type);
                    break;
                case 3: //
                    structure.constantPool[i] = new ConstantInteger(type);
                    break;
                case 4:
                    structure.constantPool[i] = new ConstantFloat(type);
                    break;
                case 5:
                    structure.constantPool[i] = new ConstantLong(type);
                    i++;// 占2位
                    break;
                case 6:
                    structure.constantPool[i] = new ConstantDouble(type);
                    i++;// 占2位
                    break;
                case 7: //
                    structure.constantPool[i] = new ConstantClass(type);
                    break;
                case 8: //
                    structure.constantPool[i] = new ConstantString(type);
                    break;
                case 9: //
                    structure.constantPool[i] = new ConstantFieldref(type);
                    break;
                case 10: //
                    structure.constantPool[i] = new ConstantMethodref(type);
                    break;
                case 11: //
                    structure.constantPool[i] = new ConstantInterfaceMethodref(type);
                    break;
                case 12: //
                    structure.constantPool[i] = new ConstantNameAndType(type);
                    break;
                case 15: //
                    structure.constantPool[i] = new ConstantMethodHandle(type);
                    break;
                case 16: //
                    structure.constantPool[i] = new ConstantMethodType(type);
                    break;
                case 18: //
                    structure.constantPool[i] = new ConstantInvokeDynamic(type);
                    break;
                default:
                    throw new ParserException("class file parser exception");
            }
            structure.constantPool[index].parse(data);
        }
    }

2.6 accessFlags

    private void accessFlags(ClassFileStructure structure, ByteBuffer data) {
        structure.accessFlags = readUnsignedShort(data);
    }

2.7 thisClass

    private void thisClass(ClassFileStructure structure, ByteBuffer data) {
        structure.thisClass = readUnsignedShort(data);
    }

2.8 superClass

    private void superClass(ClassFileStructure structure, ByteBuffer data) {
        structure.superClass = readUnsignedShort(data);
    }

2.9 interfacesCount

    private void interfacesCount(ClassFileStructure structure, ByteBuffer data) {
        structure.interfacesCount = readUnsignedShort(data);
    }

2.10 int[] interfaces

    private void interfaces(ClassFileStructure structure, ByteBuffer data) {
        structure.interfaces = new int[structure.interfacesCount];
        for (int i = 0; i < structure.interfacesCount; i++) {
            structure.interfaces[i] = readUnsignedShort(data);
        }
    }

2.11 fieldsCount

    private ClassFile fieldsCount(ClassFileStructure structure, ByteBuffer data) {
        structure.fieldsCount = readUnsignedShort(data);
        return this;
    }

2.12 FieldInfo[] fields

FieldInfo:

 private class FieldInfo {
        int accessFlags;
        int nameIndex;
        int descriptorIndex;
        int attributesCount;
        AttributeInfo[] attributes;

        public FieldInfo parse(ByteBuffer data) {
            this.accessFlags = readUnsignedShort(data);
            this.nameIndex = readUnsignedShort(data);
            this.descriptorIndex = readUnsignedShort(data);
            this.attributesCount = readUnsignedShort(data);
            this.attributes = new AttributeInfo[attributesCount];
            for (int i = 0; i < this.attributesCount; i++) {
                this.attributes[i] = new AttributeInfo().parse(data);
            }
            return this;
        }
    }

AttributeInfo:

  private class AttributeInfo {

        int attributeNameIndex;
        long attributeLength;
        short[] info;

        public AttributeInfo parse(ByteBuffer data) {
            this.attributeNameIndex = readUnsignedShort(data);
            this.attributeLength = readUnsignedInt(data);
            this.info = new short[(int) attributeLength];
            for (int i = 0; i < this.attributeLength; i++) {
                this.info[i] = (short) readUnsignedByte(data);
            }
            return this;
        }
    }


    private void fields(ClassFileStructure structure, ByteBuffer data) {
        structure.fields = new FieldInfo[structure.fieldsCount];
        for (int i = 0; i < structure.fieldsCount; i++) {
            structure.fields[i] = new FieldInfo().parse(data);
        }
    }

2.13 methodsCount

    private ClassFile methodsCount(ClassFileStructure structure, ByteBuffer data) {
        structure.methodsCount = readUnsignedShort(data);
        return this;
    }


2.14 MethodInfo[]
MethodInfo:

 private class MethodInfo {
        int accessFlags;
        int nameIndex;
        int descriptorIndex;
        int attributesCount;
        AttributeInfo[] attributes;

        public MethodInfo parse(ByteBuffer data) {
            this.accessFlags = readUnsignedShort(data);
            this.nameIndex = readUnsignedShort(data);
            this.descriptorIndex = readUnsignedShort(data);
            this.attributesCount = readUnsignedShort(data);
            this.attributes = new AttributeInfo[attributesCount];
            for (int i = 0; i < this.attributesCount; i++) {
                this.attributes[i] = new AttributeInfo().parse(data);
            }
            return this;
        }
    }


    private void methods(ClassFileStructure structure, ByteBuffer data) {
        structure.methods = new MethodInfo[structure.methodsCount];
        for (int i = 0; i < structure.methodsCount; i++) {
            structure.methods[i] = new MethodInfo().parse(data);
        }
    }

2.15 attributesCount

    private void attributesCount(ClassFileStructure structure, ByteBuffer data) {
        structure.attributesCount = readUnsignedShort(data);
    }

2.16 AttributeInfo[]

    private void attributes(ClassFileStructure structure, ByteBuffer data) {
        structure.attributes = new AttributeInfo[structure.attributesCount];
        for (int i = 0; i < structure.attributesCount; i++) {
            structure.attributes[i] = new AttributeInfo().parse(data);
        }
    }

有关Class文件解析的更多相关文章

  1. Ruby 解析字符串 - 2

    我有一个字符串input="maybe(thisis|thatwas)some((nice|ugly)(day|night)|(strange(weather|time)))"Ruby中解析该字符串的最佳方法是什么?我的意思是脚本应该能够像这样构建句子:maybethisissomeuglynightmaybethatwassomenicenightmaybethiswassomestrangetime等等,你明白了......我应该一个字符一个字符地读取字符串并构建一个带有堆栈的状态机来存储括号值以供以后计算,还是有更好的方法?也许为此目的准备了一个开箱即用的库?

  2. ruby - 使用 RubyZip 生成 ZIP 文件时设置压缩级别 - 2

    我有一个Ruby程序,它使用rubyzip压缩XML文件的目录树。gem。我的问题是文件开始变得很重,我想提高压缩级别,因为压缩时间不是问题。我在rubyzipdocumentation中找不到一种为创建的ZIP文件指定压缩级别的方法。有人知道如何更改此设置吗?是否有另一个允许指定压缩级别的Ruby库? 最佳答案 这是我通过查看ruby​​zip内部创建的代码。level=Zlib::BEST_COMPRESSIONZip::ZipOutputStream.open(zip_file)do|zip|Dir.glob("**/*")d

  3. ruby - 其他文件中的 Rake 任务 - 2

    我试图在一个项目中使用rake,如果我把所有东西都放到Rakefile中,它会很大并且很难读取/找到东西,所以我试着将每个命名空间放在lib/rake中它自己的文件中,我添加了这个到我的rake文件的顶部:Dir['#{File.dirname(__FILE__)}/lib/rake/*.rake'].map{|f|requiref}它加载文件没问题,但没有任务。我现在只有一个.rake文件作为测试,名为“servers.rake”,它看起来像这样:namespace:serverdotask:testdoputs"test"endend所以当我运行rakeserver:testid时

  4. ruby-on-rails - 在 Rails 中将文件大小字符串转换为等效千字节 - 2

    我的目标是转换表单输入,例如“100兆字节”或“1GB”,并将其转换为我可以存储在数据库中的文件大小(以千字节为单位)。目前,我有这个:defquota_convert@regex=/([0-9]+)(.*)s/@sizes=%w{kilobytemegabytegigabyte}m=self.quota.match(@regex)if@sizes.include?m[2]eval("self.quota=#{m[1]}.#{m[2]}")endend这有效,但前提是输入是倍数(“gigabytes”,而不是“gigabyte”)并且由于使用了eval看起来疯狂不安全。所以,功能正常,

  5. ruby-on-rails - Rails 3 中的多个路由文件 - 2

    Rails2.3可以选择随时使用RouteSet#add_configuration_file添加更多路由。是否可以在Rails3项目中做同样的事情? 最佳答案 在config/application.rb中:config.paths.config.routes在Rails3.2(也可能是Rails3.1)中,使用:config.paths["config/routes"] 关于ruby-on-rails-Rails3中的多个路由文件,我们在StackOverflow上找到一个类似的问题

  6. ruby - 将差异补丁应用于字符串/文件 - 2

    对于具有离线功能的智能手机应用程序,我正在为Xml文件创建单向文本同步。我希望我的服务器将增量/差异(例如GNU差异补丁)发送到目标设备。这是计划:Time=0Server:hasversion_1ofXmlfile(~800kiB)Client:hasversion_1ofXmlfile(~800kiB)Time=1Server:hasversion_1andversion_2ofXmlfile(each~800kiB)computesdeltaoftheseversions(=patch)(~10kiB)sendspatchtoClient(~10kiBtransferred)Cl

  7. ruby - 如何将脚本文件的末尾读取为数据文件(Perl 或任何其他语言) - 2

    我正在寻找执行以下操作的正确语法(在Perl、Shell或Ruby中):#variabletoaccessthedatalinesappendedasafileEND_OF_SCRIPT_MARKERrawdatastartshereanditcontinues. 最佳答案 Perl用__DATA__做这个:#!/usr/bin/perlusestrict;usewarnings;while(){print;}__DATA__Texttoprintgoeshere 关于ruby-如何将脚

  8. ruby - 解析 RDFa、微数据等的最佳方式是什么,使用统一的模式/词汇(例如 schema.org)存储和显示信息 - 2

    我主要使用Ruby来执行此操作,但到目前为止我的攻击计划如下:使用gemsrdf、rdf-rdfa和rdf-microdata或mida来解析给定任何URI的数据。我认为最好映射到像schema.org这样的统一模式,例如使用这个yaml文件,它试图描述数据词汇表和opengraph到schema.org之间的转换:#SchemaXtoschema.orgconversion#data-vocabularyDV:name:namestreet-address:streetAddressregion:addressRegionlocality:addressLocalityphoto:i

  9. ruby - 使用 Vim Rails,您可以创建一个新的迁移文件并一次性打开它吗? - 2

    使用带有Rails插件的vim,您可以创建一个迁移文件,然后一次性打开该文件吗?textmate也可以这样吗? 最佳答案 你可以使用rails.vim然后做类似的事情::Rgeneratemigratonadd_foo_to_bar插件将打开迁移生成的文件,这正是您想要的。我不能代表textmate。 关于ruby-使用VimRails,您可以创建一个新的迁移文件并一次性打开它吗?,我们在StackOverflow上找到一个类似的问题: https://sta

  10. ruby - 用逗号、双引号和编码解析 csv - 2

    我正在使用ruby​​1.9解析以下带有MacRoman字符的csv文件#encoding:ISO-8859-1#csv_parse.csvName,main-dialogue"Marceu","Giveittohimóhe,hiswife."我做了以下解析。require'csv'input_string=File.read("../csv_parse.rb").force_encoding("ISO-8859-1").encode("UTF-8")#=>"Name,main-dialogue\r\n\"Marceu\",\"Giveittohim\x97he,hiswife.\"\

随机推荐