Android

关注公众号 jb51net

关闭
首页 > 软件编程 > Android > Android Google AutoService

Android Google AutoService框架使用详解

作者:devnn

AutoService是Google开发一个自动生成SPI清单文件的框架。看过一些基于APT的三方框架源码的读者应该有所了解。比如Arouter、EventBus等等

一般我们用它来自动帮我们注册APT文件(全称是Annotation Process Tool,或者叫注解处理器,AbstractProcessor的实现)。很多生成SPI文件的框架也是抄袭它的源码,可见它的作用还不小。

APT其实就是基于SPI一个工具,是JDK留给开发者的一个在编译前处理注解的接口。APT也是SPI的一个应用。关于SPI和APT下文会详细讲到。

先讲一下它是如何使用的。

AutoService的使用

AutoService框架的作用是自动生成SPI清单文件(META-INF/services下的文件)。不用它也行,如果不使用它就需要手动去创建这个文件、手动往这个文件里添加服务(接口实现)。

AutoService比较常用的场景是帮助注册APT(注解处理器)。下面以APT的例子来讲解它的使用。

开发APT需要在Java SE项目中开发,因为需要继承AbstractProcessor,AbstractProcessor作用在Java编译阶段。

先创建Java module,在Android Studio中也可以创建,然后在build.gradle中添加依赖,如下dependencies部分。

通过annotationProcessor添加注解处理器(AutoServiceProcessor.class),同时需要通过implementation添加annotation依赖,即AutoService.class。

plugins {
    id 'java-library'
}

dependencies {
    annotationProcessor 'com.google.auto.service:auto-service:1.0.1'
    //一般结合JavaPoet框架来生成Java代码,这里不对它进行阐述。
    //implementation 'com.squareup:javapoet:1.13.0' 
    implementation 'com.google.auto.service:auto-service-annotations:1.0.1'
}

然后在你处理注解处理器类上方添加@AutoService注解即可,value指定成javax.annotation.processing.Processor类,因为要生成的SPI清单文件(META-INF/services下的文件)名称是

javax.annotation.processing.Processor 这个Processor是Java内置的,Javac编译前默认的注解处理器接口。如果是我们自定义的接口就指定成自己的接口名。

@AutoService(value = {Processor.class})
public class MyProcessor extends AbstractProcessor {
    @Override
    public synchronized void init(ProcessingEnvironment processingEnv) {
        System.out.println("MyProcessor------------init---------------");
        super.init(processingEnv);
    }
    @Override
    public boolean process(Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
        System.out.println("MyProcessor------------process---------------");
        return false;
    }
}

AbstractProcessor是继承自Processor接口:

public abstract class AbstractProcessor implements Processor {
	...
}

AbstractProcessor这个类是JDK SE中的,Android Framework将它删除了(因为不需要也用不着),所以Android Module里面是不存在的。这也说明为什么创建Java SE项目来编写APT代码。

AutoService注解的声明如下,它的value是一个class集合,可以指定多个value。

@Documented
@Retention(CLASS)
@Target(TYPE)
public @interface AutoService {
  /** Returns the interfaces implemented by this service provider. */
  Class<?>[] value();
}

以上示例中MyProcessor的作用是处理项目的自定义注解,比如Arouter框架会利用它来处理@Aouter注解,并自动生成路由注册类。

编译这个Java项目后就会自动将MyProcessor添加到APT的SPI注册文件中。

要注意的是,这个时候MyProcessor是没有起作用的,init和process方法都不会执行。因为注解处理阶段它并不在SPI注册文件中,注解处理阶段完成后它才注册进去。将Java项目打包成jar,这个MyProcessor才会在SPI注册文件中。别的项目依赖这个jar,MyProcessor的代码才会执行。

以上是AutoService的使用。讲了这些,可能有人看不懂。没关系,先了解一下SPI技术。

关于SPI

什么是SPI呢,了解SPI是读懂AutoService的基础。

SPI是Service Provider Interface的简称,是JDK默认提供的一种将接口和实现类进行分离的机制。这种机制能将接口和实现进行解耦,大大提升系统的可扩展性。

SPI机制约定:当一个Jar包需要提供一个接口的实现类时,这个Jar包需要在META-INF/services目录里同时创建一个以服务接口命名的文件。该文件里就是实现该服务接口的具体实现类。而当外部程序装配这个模块的时候,就能通过该Jar包META-INF/services/里的配置文件找到具体的实现类名,并装载实例化,完成模块的注入。

SPI示例

比如有一个接口IMyService

package com.devnn.demo.interface
public interface IMyService {
    void hello();
}

它的实现类有:

package com.devnn.demo.impl
import com.devnn.demo.interfaces.devnnService;
public class MyServiceImpl_1 implements IMyService {
    @Override
    public void hello() {
        System.out.println("Hi,I am MyServiceImpl_1");
    }
}
package com.devnn.demo.impl;
import com.devnn.demo.interfaces.devnnService;
public class MyServiceImpl_2 implements IMyService {
    @Override
    public void hello() {
        System.out.println("Hi,I am MyServiceImpl_2");
    }
}

resource/META-INF/services目录下创建文件com.devnn.demo.interface.IMyService,内容为所有实现类的完整名称:

com.devnn.demo.impl.MyServiceImpl_1
com.devnn.demo.impl.MyServiceImpl_2

项目结构:

加载IMyService接口的所有子类:

public class SPI_Demo {
    public static void main(String[] agrs) {
       //使用jdk提供的类ServiceLoader来加载IMyService的子类
       ServiceLoader<IMyService> loaders = ServiceLoader.load(IMyService.class);
       //遍历并调用子类方法
        for (IMyService service : loaders) { 
            service.hello();
        }
    }
}

运行就会打印:

Hi,I am MyServiceImpl_1
Hi,I am MyServiceImpl_2

是不是很神奇,通过一个接口,就可以找到它的实现类,这就是SPI的作用。

APT技术

然后再说下APT,开头说了APT是SPI的一个应用。为什么这么说呢?APT其实就是Java给我们提供的内置的SPI接口,作用是在编译java前处理java源码中的注解。

APT的服务接口就是这个

javax.annotation.processing.Processor

跟META_INF/service下的文件名是一致的。

Java编译器读取这个清单文件,加载实现这个接口的所有类,完成用户的注解处理逻辑。

AutoService源码

然后再回到AutoService,结合源码对它进行剖析,AutoService主要代码就一个类,即AutoServiceProcessor.java,为了方便阅读,笔者先将它原封不动copy在这里,后面再对它进行解析。

/*
 * Copyright 2008 Google LLC
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package com.google.auto.service.processor;
import static com.google.auto.common.AnnotationMirrors.getAnnotationValue;
import static com.google.auto.common.MoreElements.getAnnotationMirror;
import static com.google.auto.common.MoreStreams.toImmutableSet;
import static com.google.common.base.Throwables.getStackTraceAsString;
import com.google.auto.common.MoreElements;
import com.google.auto.common.MoreTypes;
import com.google.auto.service.AutoService;
import com.google.common.annotations.VisibleForTesting;
import com.google.common.collect.HashMultimap;
import com.google.common.collect.ImmutableList;
import com.google.common.collect.ImmutableSet;
import com.google.common.collect.Multimap;
import com.google.common.collect.Sets;
import java.io.IOException;
import java.io.OutputStream;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
import java.util.SortedSet;
import javax.annotation.processing.AbstractProcessor;
import javax.annotation.processing.Filer;
import javax.annotation.processing.RoundEnvironment;
import javax.annotation.processing.SupportedOptions;
import javax.lang.model.SourceVersion;
import javax.lang.model.element.AnnotationMirror;
import javax.lang.model.element.AnnotationValue;
import javax.lang.model.element.Element;
import javax.lang.model.element.PackageElement;
import javax.lang.model.element.TypeElement;
import javax.lang.model.type.DeclaredType;
import javax.lang.model.type.TypeMirror;
import javax.lang.model.util.SimpleAnnotationValueVisitor8;
import javax.lang.model.util.Types;
import javax.tools.Diagnostic.Kind;
import javax.tools.FileObject;
import javax.tools.StandardLocation;
/**
 * Processes {@link AutoService} annotations and generates the service provider
 * configuration files described in {@link java.util.ServiceLoader}.
 * <p>
 * Processor Options:<ul>
 *   <li>{@code -Adebug} - turns on debug statements</li>
 *   <li>{@code -Averify=true} - turns on extra verification</li>
 * </ul>
 */
@SupportedOptions({"debug", "verify"})
public class AutoServiceProcessor extends AbstractProcessor {
  @VisibleForTesting
  static final String MISSING_SERVICES_ERROR = "No service interfaces provided for element!";
  private final List<String> exceptionStacks = Collections.synchronizedList(new ArrayList<>());
  /**
   * Maps the class names of service provider interfaces to the
   * class names of the concrete classes which implement them.
   * <p>
   * For example,
   *   {@code "com.google.apphosting.LocalRpcService" ->
   *   "com.google.apphosting.datastore.LocalDatastoreService"}
   */
  private final Multimap<String, String> providers = HashMultimap.create();
  @Override
  public ImmutableSet<String> getSupportedAnnotationTypes() {
    return ImmutableSet.of(AutoService.class.getName());
  }
  @Override
  public SourceVersion getSupportedSourceVersion() {
    return SourceVersion.latestSupported();
  }
  /**
   * <ol>
   *  <li> For each class annotated with {@link AutoService}<ul>
   *      <li> Verify the {@link AutoService} interface value is correct
   *      <li> Categorize the class by its service interface
   *      </ul>
   *
   *  <li> For each {@link AutoService} interface <ul>
   *       <li> Create a file named {@code META-INF/services/<interface>}
   *       <li> For each {@link AutoService} annotated class for this interface <ul>
   *           <li> Create an entry in the file
   *           </ul>
   *       </ul>
   * </ol>
   */
  @Override
  public boolean process(Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
    try {
      processImpl(annotations, roundEnv);
    } catch (RuntimeException e) {
      // We don't allow exceptions of any kind to propagate to the compiler
      String trace = getStackTraceAsString(e);
      exceptionStacks.add(trace);
      fatalError(trace);
    }
    return false;
  }
  ImmutableList<String> exceptionStacks() {
    return ImmutableList.copyOf(exceptionStacks);
  }
  private void processImpl(Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
    if (roundEnv.processingOver()) {
      generateConfigFiles();
    } else {
      processAnnotations(annotations, roundEnv);
    }
  }
  private void processAnnotations(
      Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
    Set<? extends Element> elements = roundEnv.getElementsAnnotatedWith(AutoService.class);
    log(annotations.toString());
    log(elements.toString());
    for (Element e : elements) {
      // TODO(gak): check for error trees?
      TypeElement providerImplementer = MoreElements.asType(e);
      AnnotationMirror annotationMirror = getAnnotationMirror(e, AutoService.class).get();
      Set<DeclaredType> providerInterfaces = getValueFieldOfClasses(annotationMirror);
      if (providerInterfaces.isEmpty()) {
        error(MISSING_SERVICES_ERROR, e, annotationMirror);
        continue;
      }
      for (DeclaredType providerInterface : providerInterfaces) {
        TypeElement providerType = MoreTypes.asTypeElement(providerInterface);
        log("provider interface: " + providerType.getQualifiedName());
        log("provider implementer: " + providerImplementer.getQualifiedName());
        if (checkImplementer(providerImplementer, providerType, annotationMirror)) {
          providers.put(getBinaryName(providerType), getBinaryName(providerImplementer));
        } else {
          String message =
              "ServiceProviders must implement their service provider interface. "
                  + providerImplementer.getQualifiedName()
                  + " does not implement "
                  + providerType.getQualifiedName();
          error(message, e, annotationMirror);
        }
      }
    }
  }
  private void generateConfigFiles() {
    Filer filer = processingEnv.getFiler();
    for (String providerInterface : providers.keySet()) {
      String resourceFile = "META-INF/services/" + providerInterface;
      log("Working on resource file: " + resourceFile);
      try {
        SortedSet<String> allServices = Sets.newTreeSet();
        try {
          // would like to be able to print the full path
          // before we attempt to get the resource in case the behavior
          // of filer.getResource does change to match the spec, but there's
          // no good way to resolve CLASS_OUTPUT without first getting a resource.
          FileObject existingFile =
              filer.getResource(StandardLocation.CLASS_OUTPUT, "", resourceFile);
          log("Looking for existing resource file at " + existingFile.toUri());
          Set<String> oldServices = ServicesFiles.readServiceFile(existingFile.openInputStream());
          log("Existing service entries: " + oldServices);
          allServices.addAll(oldServices);
        } catch (IOException e) {
          // According to the javadoc, Filer.getResource throws an exception
          // if the file doesn't already exist.  In practice this doesn't
          // appear to be the case.  Filer.getResource will happily return a
          // FileObject that refers to a non-existent file but will throw
          // IOException if you try to open an input stream for it.
          log("Resource file did not already exist.");
        }
        Set<String> newServices = new HashSet<>(providers.get(providerInterface));
        if (!allServices.addAll(newServices)) {
          log("No new service entries being added.");
          continue;
        }
        log("New service file contents: " + allServices);
        FileObject fileObject =
            filer.createResource(StandardLocation.CLASS_OUTPUT, "", resourceFile);
        try (OutputStream out = fileObject.openOutputStream()) {
          ServicesFiles.writeServiceFile(allServices, out);
        }
        log("Wrote to: " + fileObject.toUri());
      } catch (IOException e) {
        fatalError("Unable to create " + resourceFile + ", " + e);
        return;
      }
    }
  }
  /**
   * Verifies {@link ServiceProvider} constraints on the concrete provider class. Note that these
   * constraints are enforced at runtime via the ServiceLoader, we're just checking them at compile
   * time to be extra nice to our users.
   */
  private boolean checkImplementer(
      TypeElement providerImplementer,
      TypeElement providerType,
      AnnotationMirror annotationMirror) {
    String verify = processingEnv.getOptions().get("verify");
    if (verify == null || !Boolean.parseBoolean(verify)) {
      return true;
    }
    // TODO: We're currently only enforcing the subtype relationship
    // constraint. It would be nice to enforce them all.
    Types types = processingEnv.getTypeUtils();
    if (types.isSubtype(providerImplementer.asType(), providerType.asType())) {
      return true;
    }
    // Maybe the provider has generic type, but the argument to @AutoService can't be generic.
    // So we allow that with a warning, which can be suppressed with @SuppressWarnings("rawtypes").
    // See https://github.com/google/auto/issues/870.
    if (types.isSubtype(providerImplementer.asType(), types.erasure(providerType.asType()))) {
      if (!rawTypesSuppressed(providerImplementer)) {
        warning(
            "Service provider "
                + providerType
                + " is generic, so it can't be named exactly by @AutoService."
                + " If this is OK, add @SuppressWarnings(\"rawtypes\").",
            providerImplementer,
            annotationMirror);
      }
      return true;
    }
    return false;
  }
  private static boolean rawTypesSuppressed(Element element) {
    for (; element != null; element = element.getEnclosingElement()) {
      SuppressWarnings suppress = element.getAnnotation(SuppressWarnings.class);
      if (suppress != null && Arrays.asList(suppress.value()).contains("rawtypes")) {
        return true;
      }
    }
    return false;
  }
  /**
   * Returns the binary name of a reference type. For example,
   * {@code com.google.Foo$Bar}, instead of {@code com.google.Foo.Bar}.
   *
   */
  private String getBinaryName(TypeElement element) {
    return getBinaryNameImpl(element, element.getSimpleName().toString());
  }
  private String getBinaryNameImpl(TypeElement element, String className) {
    Element enclosingElement = element.getEnclosingElement();
    if (enclosingElement instanceof PackageElement) {
      PackageElement pkg = MoreElements.asPackage(enclosingElement);
      if (pkg.isUnnamed()) {
        return className;
      }
      return pkg.getQualifiedName() + "." + className;
    }
    TypeElement typeElement = MoreElements.asType(enclosingElement);
    return getBinaryNameImpl(typeElement, typeElement.getSimpleName() + "$" + className);
  }
  /**
   * Returns the contents of a {@code Class[]}-typed "value" field in a given {@code
   * annotationMirror}.
   */
  private ImmutableSet<DeclaredType> getValueFieldOfClasses(AnnotationMirror annotationMirror) {
    return getAnnotationValue(annotationMirror, "value")
        .accept(
            new SimpleAnnotationValueVisitor8<ImmutableSet<DeclaredType>, Void>(ImmutableSet.of()) {
              @Override
              public ImmutableSet<DeclaredType> visitType(TypeMirror typeMirror, Void v) {
                // TODO(ronshapiro): class literals may not always be declared types, i.e.
                // int.class, int[].class
                return ImmutableSet.of(MoreTypes.asDeclared(typeMirror));
              }
              @Override
              public ImmutableSet<DeclaredType> visitArray(
                  List<? extends AnnotationValue> values, Void v) {
                return values.stream()
                    .flatMap(value -> value.accept(this, null).stream())
                    .collect(toImmutableSet());
              }
            },
            null);
  }
  private void log(String msg) {
    if (processingEnv.getOptions().containsKey("debug")) {
      processingEnv.getMessager().printMessage(Kind.NOTE, msg);
    }
  }
  private void warning(String msg, Element element, AnnotationMirror annotation) {
    processingEnv.getMessager().printMessage(Kind.WARNING, msg, element, annotation);
  }
  private void error(String msg, Element element, AnnotationMirror annotation) {
    processingEnv.getMessager().printMessage(Kind.ERROR, msg, element, annotation);
  }
  private void fatalError(String msg) {
    processingEnv.getMessager().printMessage(Kind.ERROR, "FATAL ERROR: " + msg);
  }
}

AutoService源码分析

主要逻辑在process方法中,通过实现AbstractProcessor的process方法来实现功能。

process委托给了processImpl:

 private void processImpl(Set<? extends TypeElement> annotations, RoundEnvironment roundEnv) {
   if (roundEnv.processingOver()) { //本轮注解处理完毕
      generateConfigFiles();//生成SPI注册文件
    } else { //未处理完毕,继续处理
      processAnnotations(annotations, roundEnv);//整理需要注册的文件,放入缓存
    }
 }

再看processAnnotations方法,笔者已经加了注释:

private void processAnnotations(
      Set<? extends TypeElement> annotations, RoundEnvironment roundEnv){
	//获取所有加了AutoService注解的类
    Set<? extends Element> elements = roundEnv.getElementsAnnotatedWith(AutoService.class);
    for (Element e : elements) {
      //将Element转成TypeElement
      TypeElement providerImplementer = MoreElements.asType(e);
      //获取AutoServce注解指定的value
      AnnotationMirror annotationMirror = getAnnotationMirror(e, AutoService.class).get();
      //获取value集合
      Set<DeclaredType> providerInterfaces = getValueFieldOfClasses(annotationMirror);
      //如果没有指定value,报错
      if (providerInterfaces.isEmpty()) {
        error(MISSING_SERVICES_ERROR, e, annotationMirror);
        continue;
      }
      //遍历所有的value,获取value的完整类名(例如javax.annotation.processing.Processor)
      for (DeclaredType providerInterface : providerInterfaces) {
        TypeElement providerType = MoreTypes.asTypeElement(providerInterface);
		//判断是否是继承关系,是则放入providers缓存起来,否则报错
        if (checkImplementer(providerImplementer, providerType, annotationMirror)) {
          providers.put(getBinaryName(providerType), getBinaryName(providerImplementer));
        } else {
         //报错代码,略
        }
      }
    }
  }

注解处理完毕,就会生成SPI注册文件。如果SPI路径上文件已经存在,先要把已存在的SPI清单读进内存,再把新的provider加进去,然后全部写出,覆盖原来的文件。这部分逻辑如下:

  private void generateConfigFiles() {
    Filer filer = processingEnv.getFiler();//获取文件工具类,processingEnv是AbstractProcessor的成员变量,直接拿来用。
	//遍历之前解析的providers缓存
    for (String providerInterface : providers.keySet()) {
     //providerInterface就是value字段指定的接口,例如javax.annotation.processing.Processor
      String resourceFile = "META-INF/services/" + providerInterface;
      log("Working on resource file: " + resourceFile);
      try {
        SortedSet<String> allServices = Sets.newTreeSet();
        try {
        //已经存在的SPI文件
          FileObject existingFile =
              filer.getResource(StandardLocation.CLASS_OUTPUT, "", resourceFile);
          //SPI文件中的service条目清单
          Set<String> oldServices = ServicesFiles.readServiceFile(existingFile.openInputStream());
          log("Existing service entries: " + oldServices);
          allServices.addAll(oldServices);
        } catch (IOException e) {
          log("Resource file did not already exist.");
        }
		//新的service条目清单
        Set<String> newServices = new HashSet<>(providers.get(providerInterface));
        //如果已经存在,则不处理
        if (!allServices.addAll(newServices)) {
          log("No new service entries being added.");
          continue;
        }
		//以下是将缓存的services写入文件中。
        log("New service file contents: " + allServices);
        FileObject fileObject =
            filer.createResource(StandardLocation.CLASS_OUTPUT, "", resourceFile);
        try (OutputStream out = fileObject.openOutputStream()) {
          ServicesFiles.writeServiceFile(allServices, out);
        }
        log("Wrote to: " + fileObject.toUri());
      } catch (IOException e) {
        fatalError("Unable to create " + resourceFile + ", " + e);
        return;
      }
    }
  }

可见AutoServiceProcessor的主要功能就是将加了AutoService注解的类,加到SPI注册文件中。SPI文件名称(或者叫服务)可以通过value指定。

下面将AutoService从mavenCentral仓库中下载下来(一个jar包),解压查看它的内容:

可以看到它里面内容并不多,主要就是一个AutoServiceProcessor类和一个APT清单文件。打开这个清单文件,里面就是AutoServiceProcessor类的全路径:

所以我们将AutoService加到java项目中,其实就是引入了AutoServiceProcessor这个注解处理器,帮助我们处理@AutoService注解,将我们的服务(一般是APT类,也可以是其它的类,通过value指定)自动注册进SPI文件中。

看到这里,不知道读者有没有领悟。

AutoService是一个注解处理器,我们自己开发的APT也是注解处理器,它们都是注解处理器,AutoSevice是自动帮我们注册注解处理器的注解处理器。是不是有点绕?

当然AutoService的作用不仅在于注册APT,还可以注册其它服务。只是注册APT我们比较常见。

再举一个AutoService的使用场景:

在组件化架构app中,有一个主Module和若干业务Module,如何在主Module中初始化各个业务Module?这可以使用SPI技术,在业务Module中创建一个初始化类实现一个共同的接口,然后在这个类上加AutoService注解,在主Module中就可以通过SPI机制加载这些业务Module的初始化类,调用初始化接口。

AutoService不仅是一个自动注册APT的框架,它还是一个SPI技术的模板,有时候我们需要自己开发一个基于APT同时又要注册自定义service的框架,它的源码是一个很好的参考。AutoServiceProcessor里面的大部分代码是可以复制拿来用。再比如,ServiceFiles.java是SPI资源文件读取和写入的工具类,直接复制到我们项目中即可。

到此这篇关于Android Google AutoService框架使用详解的文章就介绍到这了,更多相关Android Google AutoService内容请搜索脚本之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持脚本之家!

您可能感兴趣的文章:
阅读全文