Android项目总结复盘2

文章将会同步到个人微信公众号:Android部落格

2、系统数据检查

2.1 dex更新

我们开发的java代码通过编译生成.class文件,然后通过dx工具生成机器可以识别的dex文件。

Android中采用ClassLoader加载dex文件,加载完成之后可以通过反射调用其中的方法,适合那些不依赖文件等资源的业务,而打点恰好比较适合使用dex加载的方式。

Android中有PathClassLoader和DexClassLoader ,他们都继承自ClassLoader。他们的继承关系如下:

  • DexClassLoader:可以加载jar/apk/dex,可以从SD卡中加载未安装的apk;
  • PathClassLoader:要传入系统中apk的存放Path,所以只能加载已经安装的apk文件。
  • BaseDexClassLoader是DexClassLoader和PathClassLoader的父类,针对传入不同的参数做差异化处理。

看看三个类的源码:

2.1.1 PathClassLoader

dalvik/system/PathClassLoader.java

public class PathClassLoader extends BaseDexClassLoader {
    public PathClassLoader(String dexPath, ClassLoader parent) {
        super(dexPath, null, null, parent);
    }
    
    public PathClassLoader(String dexPath, String librarySearchPath, ClassLoader parent) {
        super(dexPath, null, librarySearchPath, parent);
    }
}

2.1.2 DexClassLoader

dalvik/system/DexClassLoader.java

public class DexClassLoader extends BaseDexClassLoader {
    public DexClassLoader(String dexPath, String optimizedDirectory, String librarySearchPath, ClassLoader parent) {
        super(dexPath, null, librarySearchPath, parent);
    }
}

2.1.3 BaseDexClassLoader

public class BaseDexClassLoader extends ClassLoader {
    private final DexPathList pathList;
    public BaseDexClassLoader(String dexPath, File optimizedDirectory,
            String librarySearchPath, ClassLoader parent) {
        this(dexPath, optimizedDirectory, librarySearchPath, parent, false);
    }
    
    public BaseDexClassLoader(String dexPath, File optimizedDirectory,
            String librarySearchPath, ClassLoader parent, boolean isTrusted) {
        super(parent);
        this.pathList = new DexPathList(this, dexPath, librarySearchPath, null, isTrusted);

        if (reporter != null) {
            reportClassLoaderChain();
        }
    }
    
    public BaseDexClassLoader(ByteBuffer[] dexFiles, ClassLoader parent) {
        // TODO We should support giving this a library search path maybe.
        super(parent);
        this.pathList = new DexPathList(this, dexFiles);
    }

在BaseDexClassLoader中通过DexPathList类具体的处理Dex,他的构造函数如下:

2.1.4 DexPathList

private Element[] dexElements;
DexPathList(ClassLoader definingContext, String dexPath, String librarySearchPath, File optimizedDirectory, boolean isTrusted) {
    this.dexElements = makeDexElements(splitDexPath(dexPath), optimizedDirectory, suppressedExceptions, definingContext, isTrusted);
}
            
private static Element[] makeDexElements(List<File> files, File optimizedDirectory, List<IOException> suppressedExceptions, ClassLoader loader, boolean isTrusted) {
    Element[] elements = new Element[files.size()];
    int elementsPos = 0;
    for (File file : files) {
        if (file.isDirectory()) {
            elements[elementsPos++] = new Element(file);
        } else if (file.isFile()) {
            String name = file.getName();
            
            DexFile dex = null;
            dex = loadDexFile(file, optimizedDirectory, loader, elements);
        }
    }
    retrun elements;
}

private static DexFile loadDexFile(File file, File optimizedDirectory, ClassLoader loader, Element[] elements) throws IOException {
    if (optimizedDirectory == null) {
        return new DexFile(file, loader, elements);
    } else {
        String optimizedPath = optimizedPathFor(file, optimizedDirectory);
        return DexFile.loadDex(file.getPath(), optimizedPath, 0, loader, elements);
    }
}

最终通过DexFile加载loadDex方法在native层实现对dex的加载和处理。

而分析PathClassLoaderDexClassLoader的构造函数可以看到Android 9.0中的DexClassLoader构造函数的optimizedDirectory参数默认是null。所以这里要针对版本的不同做差异化处理。

我们一般加载Dex的方式是:

classLoader = context.getClassLoader();
classLoader.loadClass("you class path");

这里做一下差异化:

public ClassLoader load(Context context,String dexName) {
    mLoaded = false;
    ClassLoader classLoader = null;
    File dexOutputDir = context.getDir("dex", 0);
    File dexFile = new File(dexOutputDir.getAbsolutePath(), dexName);
    Log.d(TAG, "load start");
    if (!dexFile.exists()) {
        return null;
    }
    String dexPath = dexFile.getAbsolutePath();
    Log.d(TAG, "dPath = " + dexPath);
    int version = android.os.Build.VERSION.SDK_INT;
    if (version >= 25) {
        BaseDexClassLoader parent = (BaseDexClassLoader) context.getClassLoader();
        Class<BaseDexClassLoader> c = BaseDexClassLoader.class;
        Method method;
        try {
            method = c.getMethod("addDexPath", String.class);
            method.invoke(parent, dexPath);
            mLoaded = true;
            classLoader = parent;
            return classLoader;
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    Log.d(TAG, "mLoaded1 = " + mLoaded);
    if (!mLoaded) {
        ArrayList<File> files = new ArrayList<File>();
        files.add(dexFile);

        classLoader = context.getClassLoader();
        classLoader.loadClass("you class path");
        try {
            Field pathListField = findField(classLoader, "pathList");
            Object dexPathList = pathListField.get(classLoader);
            ArrayList<IOException> suppressedExceptions = new ArrayList<IOException>();
            if (version < 19) {
                expandFieldArray(dexPathList, "dexElements",
                        makeDexElements(dexPathList, files, null));
            } else if (version < 23) {
                expandFieldArray(dexPathList, "dexElements",
                        makeDexElements(dexPathList, files, null, suppressedExceptions));
            } else {
                expandFieldArray(dexPathList, "dexElements",
                        makePathElements(dexPathList, files, null, suppressedExceptions));
            }

        } catch (Exception e1) {
            e1.printStackTrace();
        }
    }
    Log.d(TAG, "mLoaded2 = " + mLoaded);
    if (mLoaded) {
        return classLoader;
    }
    return null;
}
  • SDK_INI大于等于25

当sdk_int大于等于25时,通过反射BaseDexClassLoader的addDexPath方法直接添加dex文件到DexPathList的Element[]数组中,而后续findClass方法的逻辑就是遍历这个数据找到对应的dex文件。

  • SDK_INI小于25

这种情况下,先反射获取ClassLoader的pathList对象,这里的ClassLoader实际是PathClassLoader,但是最终都会到BaseDexClassLoader的pathList。

获取到这个变量之后,先调用makeDexElements方法将生成的dex对象放到一个数组中,接着在expandFieldArray方法中将就的dex数组和新的dex数组合并:

private Object[] makeDexElements(Object dexPathList, ArrayList<File> files, File optimizedDirectory) throws IllegalAccessException, InvocationTargetException, NoSuchMethodException {
    Method makeDexElements = findMethod(dexPathList, "makeDexElements", ArrayList.class, File.class);

    return (Object[]) makeDexElements.invoke(dexPathList, files, optimizedDirectory);
}

这里的makeDexElements方法对应的是DexPathListmakeDexElements方法,最终目的是将dex对象添加到Element[]数组中,作为新的数组返回。

expandFieldArray对应的是DexPathListaddDexPath方法,将新旧Element[]数组合到一个数组中,旧的数组在前面。这样就导致了相同文件名的dex文件,最新修复了bug的dex不能立即生效。

private void expandFieldArray(Object instance, String fieldName, Object[] extraElements) throws NoSuchFieldException, IllegalArgumentException, IllegalAccessException {
    Field jlrField = findField(instance, fieldName);
    Object[] original = (Object[]) jlrField.get(instance);
    Object[] combined = (Object[]) Array.newInstance(original.getClass()
            .getComponentType(), original.length + extraElements.length);
    System.arraycopy(original, 0, combined, 0, original.length);
    System.arraycopy(extraElements, 0, combined, original.length,
            extraElements.length);
    jlrField.set(instance, combined);
    mLoaded = true;
    Log.d(TAG, "expandFieldArray");
}

对比看下DexPathList中的addDexPath方法:

public void addDexPath(String dexPath, File optimizedDirectory, boolean isTrusted) {
    final List<IOException> suppressedExceptionList = new ArrayList<IOException>();
    final Element[] newElements = makeDexElements(splitDexPath(dexPath), optimizedDirectory,
            suppressedExceptionList, definingContext, isTrusted);
    
    if (newElements != null && newElements.length > 0) {
        final Element[] oldElements = dexElements;
        dexElements = new Element[oldElements.length + newElements.length];
        System.arraycopy(
                oldElements, 0, dexElements, 0, oldElements.length);
        System.arraycopy(
                newElements, 0, dexElements, oldElements.length, newElements.length);
    }
}

可见我们自己的操作对应着DexPathListaddDexPath方法。

在我们自己的expandFieldArray方法最后通过执行jlrField.set(instance, combined);,将合并后的Element[]数组赋值给DexPathListElement[] dexElements

当上述操作完成之后,就要调用loadClass方法加载dex文件中的类了,这个方法在ClassLoader类中定义:

protected Class<?> loadClass(String name, boolean resolve) throws ClassNotFoundException
{
    Class<?> c = findLoadedClass(name);
    if (parent != null) {
        c = parent.loadClass(name, false);
    } else {
        c = findBootstrapClassOrNull(name);
    }
    if (c == null) {
        c = findClass(name);
    }
    return c;            
}

parentContext.getClassLoader所属的PathClassLoader传递,一直从BaseDexClassLoaderClassLoader

到这里要熟悉应用被创建初始化的流程了,这里先不引申过去,只需要知道这个parent是BootClassLoader类型。

@Override
protected Class<?> loadClass(String className, boolean resolve)
       throws ClassNotFoundException {
    Class<?> clazz = findLoadedClass(className);

    if (clazz == null) {
        clazz = findClass(className);
    }

    return clazz;
}

如果还找不到,就调用BaseDexClassLoaderfindClass方法了。

@Override
protected Class<?> findClass(String name) throws ClassNotFoundException {
    List<Throwable> suppressedExceptions = new ArrayList<Throwable>();
    Class c = pathList.findClass(name, suppressedExceptions);
    return c;
}

到这里如果还找不到就抛异常出来了,ClassNotFoundException

上边的流程就是各大博客上面说的双亲委派机制,父类先从已经加载的类里面找,找不到的话,再从自己BaseDexClassLoaderfindClass方法里面去找。

2.1.5 问题复盘

到这里,基本就将Dex加载的流程搞清楚了,但是这样的加载会导致新加载的Dex不能立即生效,必须重新启动应用之后才能生效。针对这种问题,可以将Dex热更新模块放到一个单独的进程中,当Dex加载完毕之后,调用killProcess方法自杀,然后由另一个进程拉活重启。