Media Data之多媒体扫描过程分析（二）

此分析代码基于Android 6.0，转载请注明来源地址http://blog.csdn.net/lemon_blue/article/details/52262023

Media Data之多媒体扫描过程分析（一）
Media Data之多媒体扫描过程分析（三）

2.1.5 android_media_MediaScanner.cpp

对于android_media_MediaScanner.cpp来说，主要分析三个函数native_init,native_setup和processDirectory。

static void
android_media_MediaScanner_native_init(JNIEnv *env)
{
    ALOGV("native_init");
    jclass clazz = env->FindClass(kClassMediaScanner);
    if (clazz == NULL) {
        return;
    }
    //将之后创建的native对象的指针保存到MediaScanner.java的mNativeContext字段中
    fields.context = env->GetFieldID(clazz, "mNativeContext", "J");
    if (fields.context == NULL) {
        return;
    }
}

android_media_MediaScanner_native_init的功能主要是动态注册。

static void
android_media_MediaScanner_native_setup(JNIEnv *env, jobject thiz)
{
    //获取Stagefright的MediaScanner对象
    MediaScanner *mp = new StagefrightMediaScanner;
    if (mp == NULL) {
        jniThrowException(env, kRunTimeException, "Out of memory");
        return;
    }
    //将对象保存到mNativeContext中
    env->SetLongField(thiz, fields.context, (jlong)mp);
}

android_media_MediaScanner_native_setup方法的作用是创建native的MediaScanner对象，并且用的是StagefrightMediaScanner，等会分析。

static void
android_media_MediaScanner_processDirectory(
        JNIEnv *env, jobject thiz, jstring path, jobject client)
{
    //传入的参数path是需要扫描的路径，client是MediaScannerClient.java对象
    //获取之前保存到mNativeContext的StagefrightMediaScanner对象
    MediaScanner *mp = getNativeScanner_l(env, thiz);
    if (mp == NULL) {
        jniThrowException(env, kRunTimeException, "No scanner available");
        return;
    }
    if (path == NULL) {
        jniThrowException(env, kIllegalArgumentException, NULL);
        return;
    }
    const char *pathStr = env->GetStringUTFChars(path, NULL);
    if (pathStr == NULL) {  // Out of memory
        return;
    }
    //构造native层的MyMediaScannerClient对象，参数是java层的MyMediaScannerClient     
    //对象
    MyMediaScannerClient myClient(env, client);
    //调用native层processDirectory方法，参数是扫描路径和native的MyMediaScannerClient 
    //对象
    MediaScanResult result = mp->processDirectory(pathStr, myClient);
    if (result == MEDIA_SCAN_RESULT_ERROR) {
        ALOGE("An error occurred while scanning directory '%s'.", pathStr);
    }
    env->ReleaseStringUTFChars(path, pathStr);
}

android_media_MediaScanner_processDirectory方法的作用是启动native层processDirectory扫描方法，在配置过程稍显复杂，其一是java的MediaScanner的上下文环境传递给native额MediaScanner对象中，其二是native的MyMediaScannerClient对象与java的MyMediaScannerClient对象建立联系，方便将结果回调到java层。

2.1.6 MediaScanner.cpp

下面分析的是native层的相关处理，StagefrightMediaScanner.cpp继承自MediaScanner.cpp，在JNI调用的方法processDirectory也是由父类实现的。
先分析MediaScanner.cpp父类的方法。

MediaScanResult MediaScanner::processDirectory(
        const char *path, MediaScannerClient &client) {
    //前期的一些准备工作
    int pathLength = strlen(path);
    if (pathLength >= PATH_MAX) {
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    char* pathBuffer = (char *)malloc(PATH_MAX + 1);
    if (!pathBuffer) {
        return MEDIA_SCAN_RESULT_ERROR;
    }
    int pathRemaining = PATH_MAX - pathLength;
    strcpy(pathBuffer, path);
    if (pathLength > 0 && pathBuffer[pathLength - 1] != '/') {
        pathBuffer[pathLength] = '/';
        pathBuffer[pathLength + 1] = 0;
        --pathRemaining;
    }
    //设置native的MyMediaScannerClient对象的local信息
    client.setLocale(locale());
    //执行doProcessDirectory方法
    MediaScanResult result = doProcessDirectory(pathBuffer, pathRemaining, client, false);
    //释放资源
    free(pathBuffer);
    return result;
}

MediaScanResult MediaScanner::doProcessDirectory(
        char *path, int pathRemaining, MediaScannerClient &client, bool noMedia) {
    // place to copy file or directory name
    char* fileSpot = path + strlen(path);
    struct dirent* entry;
    if (shouldSkipDirectory(path)) {
        ALOGD("Skipping: %s", path);
        return MEDIA_SCAN_RESULT_OK;
    }
    // Treat all files as non-media in directories that contain a  ".nomedia" file
    if (pathRemaining >= 8 /* strlen(".nomedia") */ ) {
        strcpy(fileSpot, ".nomedia");
        if (access(path, F_OK) == 0) {
            ALOGV("found .nomedia, setting noMedia flag");
            noMedia = true;
        }
        // restore path
        fileSpot[0] = 0;
    }
    //打开对应的文件夹路径
    DIR* dir = opendir(path);
    if (!dir) {
        ALOGW("Error opening directory '%s', skipping: %s.", path, strerror(errno));
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    MediaScanResult result = MEDIA_SCAN_RESULT_OK;
    //循环遍历所有文件
    while ((entry = readdir(dir))) {
        //调用doProcessDirectoryEntry方法
        if (doProcessDirectoryEntry(path, pathRemaining, client, noMedia, entry, fileSpot)
                == MEDIA_SCAN_RESULT_ERROR) {
            result = MEDIA_SCAN_RESULT_ERROR;
            break;
        }
    }
    //关闭文件夹
    closedir(dir);
    return result;
}

MediaScanResult MediaScanner::doProcessDirectoryEntry(
        char *path, int pathRemaining, MediaScannerClient &client, bool noMedia,
        struct dirent* entry, char* fileSpot) {
    struct stat statbuf;
    //枚举目录中的文件和子文件夹信息
    const char* name = entry->d_name;
    // ignore "." and ".."
    if (name[0] == '.' && (name[1] == 0 || (name[1] == '.' && name[2] == 0))) {
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    int nameLength = strlen(name);
    if (nameLength + 1 > pathRemaining) {
        // path too long!
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    strcpy(fileSpot, name);

    int type = entry->d_type;
    if (type == DT_UNKNOWN) {
        // If the type is unknown, stat() the file instead.
        // This is sometimes necessary when accessing NFS mounted filesystems, but
        // could be needed in other cases well.
        //执行stat方法，获取文件的所有属性，成功返回0失败返回-1
        if (stat(path, &statbuf) == 0) {
            if (S_ISREG(statbuf.st_mode)) {
                type = DT_REG;
            } else if (S_ISDIR(statbuf.st_mode)) {
                type = DT_DIR;
            }
        } else {
            ALOGD("stat() failed for %s: %s", path, strerror(errno) );
        }
    }
    if (type == DT_DIR) {
        bool childNoMedia = noMedia;
        // set noMedia flag on directories with a name that starts with '.'
        // for example, the Mac ".Trashes" directory
        if (name[0] == '.')
            childNoMedia = true;
        // report the directory to the client
        if (stat(path, &statbuf) == 0) {
            //调用MyMediaScannerClient的scanFile函数
            status_t status = client.scanFile(path, statbuf.st_mtime, 0,
                    true /*isDirectory*/, childNoMedia);
            if (status) {
                //返回值是checkAndClearExceptionFromCallback，如果是true就出错
                return MEDIA_SCAN_RESULT_ERROR;
            }
        }
        // and now process its contents
        strcat(fileSpot, "/");
        MediaScanResult result = doProcessDirectory(path, pathRemaining - nameLength - 1,
                client, childNoMedia);
        if (result == MEDIA_SCAN_RESULT_ERROR) {
            return MEDIA_SCAN_RESULT_ERROR;
        }
    } else if (type == DT_REG) {
        stat(path, &statbuf);
        status_t status = client.scanFile(path, statbuf.st_mtime, statbuf.st_size,
                false /*isDirectory*/, noMedia);
        if (status) {
            return MEDIA_SCAN_RESULT_ERROR;
        }
    }
    return MEDIA_SCAN_RESULT_OK;
}

从上面的分析中看到调用到了MyMediaScannerClient的scanFile函数，下面分析这个函数

virtual status_t scanFile(const char* path, long long lastModified,
        long long fileSize, bool isDirectory, bool noMedia)
{
    jstring pathStr;
    if ((pathStr = mEnv->NewStringUTF(path)) == NULL) {
        mEnv->ExceptionClear();
        return NO_MEMORY;
    }
   //此处的mClient是java层的MyMediaScannerClient，调用的也是java层的scanFile方法
    mEnv->CallVoidMethod(mClient, mScanFileMethodID, pathStr, lastModified,
            fileSize, isDirectory, noMedia);
    mEnv->DeleteLocalRef(pathStr);
    return checkAndClearExceptionFromCallback(mEnv, "scanFile");
}

可以看出在native层的MyMediaScannerClient调用的是java层MyMediaScannerClient的scanFile函数，下面分析java层的逻辑。

public void scanFile(String path, long lastModified, long fileSize,
        boolean isDirectory, boolean noMedia) {
    // This is the callback funtion from native codes.
    //调用了doScanFile方法
    doScanFile(path, null, lastModified, fileSize, isDirectory, false, noMedia);
}
public Uri doScanFile(String path, String mimeType, long lastModified,
                long fileSize, boolean isDirectory, boolean scanAlways, boolean noMedia) {
            //参数scanAlways控制是否强制扫描
            Uri result = null;
            try {
                // beginFile方法的作用主要是1. 生成FileEntry，2.判断是否有修改文件
                FileEntry entry = beginFile(path, mimeType, lastModified,
                        fileSize, isDirectory, noMedia);
                // if this file was just inserted via mtp, set the rowid to zero
                // (even though it already exists in the database), to trigger
                // the correct code path for updating its entry
                if (mMtpObjectHandle != 0) {
                    entry.mRowId = 0;
                }
                // rescan for metadata if file was modified since last scan
                if (entry != null && (entry.mLastModifiedChanged || scanAlways)) {
                    if (noMedia) {
                        //不是media的情况
                        result = endFile(entry, false, false, false, false, false);
                    } else {
                        //重新扫描获取的信息
                        String lowpath = path.toLowerCase(Locale.ROOT);
                        boolean ringtones = (lowpath.indexOf(RINGTONES_DIR) > 0);
                        boolean notifications = (lowpath.indexOf(NOTIFICATIONS_DIR) > 0);
                        boolean alarms = (lowpath.indexOf(ALARMS_DIR) > 0);
                        boolean podcasts = (lowpath.indexOf(PODCAST_DIR) > 0);
                        boolean music = (lowpath.indexOf(MUSIC_DIR) > 0) ||
                            (!ringtones && !notifications && !alarms && !podcasts);
                        boolean isaudio = MediaFile.isAudioFileType(mFileType);
                        boolean isvideo = MediaFile.isVideoFileType(mFileType);
                        boolean isimage = MediaFile.isImageFileType(mFileType); 
                        if (isaudio || isvideo || isimage) {
                        //如过类型是音频、视频和图片的话，对路径进行处理
                        //If the given path exists on emulated external storage, 
                        //return the translated backing path hosted on internal storage.
                            path = Environment.maybeTranslateEmulatedPathToInternal
                                        (new File(path)).getAbsolutePath();
                        }
                        // we only extract metadata for audio and video files
                        if (isaudio || isvideo) {
                        //调用processFile方法，把MyMediaScannerClient作为参数传入
                        // processFile方法是native方法，稍后分析
                            processFile(path, mimeType, this);
                        }
                        if (isimage) {
                            //如果是图片，单独处理，调用processImageFile方法
                            //Decode a file path into a bitmap.
                            processImageFile(path);
                        }
                    // endFile方法是更新数据库
                    result = endFile(entry, ringtones, notifications, alarms, music, podcasts);
                    }
                }
            } catch (RemoteException e) {
                Log.e(TAG, "RemoteException in MediaScanner.scanFile()", e);
            }
            return result;
        }

从上面的分析可以看到，其实又调用到了processFile方法中，他也是一个native方法，需要再回到jni层继续分析此方法。

static void
android_media_MediaScanner_processFile(
        JNIEnv *env, jobject thiz, jstring path,
        jstring mimeType, jobject client)
{
    // Lock already hold by processDirectory
    //获取的还是native层的MediaScanner对象，实际类型是StagefrightMediaScanner对象
    MediaScanner *mp = getNativeScanner_l(env, thiz);
    const char *pathStr = env->GetStringUTFChars(path, NULL);
    if (pathStr == NULL) {  // Out of memory
        return;
    }
    //构造了新的native层的MyMediaScannerClient对象，传入的还是java层的MyMediaScannerClient对象
    MyMediaScannerClient myClient(env, client);
    //调用的是StagefrightMediaScanner对象的processFile方法，等会分析
    MediaScanResult result = mp->processFile(pathStr, mimeTypeStr, myClient);
    if (result == MEDIA_SCAN_RESULT_ERROR) {
        ALOGE("An error occurred while scanning file '%s'.", pathStr);
    }
    env->ReleaseStringUTFChars(path, pathStr);
    if (mimeType) {
        env->ReleaseStringUTFChars(mimeType, mimeTypeStr);
    }
}

从上面的分析可以看出，调用了StagefrightMediaScanner对象的processFile方法，下面分析此方法。

MediaScanResult StagefrightMediaScanner::processFile(
        const char *path, const char *mimeType,
        MediaScannerClient &client) {
    //调用native层的MyMediaScannerClient对象进行local信息，语言设置
    client.setLocale(locale());
    //beginFile方法是由MyMediaScannerClient的父类实现的，其实谷歌并没有实现此方法 
    client.beginFile();
    //具体的方法是调用processFileInternal实现的
    MediaScanResult result = processFileInternal(path, mimeType, client);
    //根据设置的区域信息来对字符串进行转换
    client.endFile();
    return result;
}

MediaScanResult StagefrightMediaScanner::processFileInternal(
        const char *path, const char * /* mimeType */,
        MediaScannerClient &client) {
    //获取扩展名信息
    const char *extension = strrchr(path, '.');
    if (!extension) {
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    //对扩展名不符合的跳过扫描
    if (!FileHasAcceptableExtension(extension)
        && !AVUtils::get()->isEnhancedExtension(extension)) {
        return MEDIA_SCAN_RESULT_SKIPPED;
    }
    // MediaMetadataRetriever将一个输入媒体文件中设置帧和元数据
    sp mRetriever(new MediaMetadataRetriever);
    //打开资源
    int fd = open(path, O_RDONLY | O_LARGEFILE);
    status_t status;
    if (fd < 0) {
        // couldn't open it locally, maybe the media server can?
        //打开资源失败
        status = mRetriever->setDataSource(NULL /* httpService */, path);
    } else {
        //设置资源
        status = mRetriever->setDataSource(fd, 0, 0x7ffffffffffffffL);
        close(fd);
    }
    if (status) {
        return MEDIA_SCAN_RESULT_ERROR;
    }

    const char *value;
    if ((value = mRetriever->extractMetadata(
                    METADATA_KEY_MIMETYPE)) != NULL) {
        //设置类型
        status = client.setMimeType(value);
        if (status) {
            return MEDIA_SCAN_RESULT_ERROR;
        }
    }
    //构造元数据的tag
    struct KeyMap {
        const char *tag;
        int key;
    };
    static const KeyMap kKeyMap[] = {
        { "tracknumber", METADATA_KEY_CD_TRACK_NUMBER },
        { "discnumber", METADATA_KEY_DISC_NUMBER },
        { "album", METADATA_KEY_ALBUM },
        { "artist", METADATA_KEY_ARTIST },
        { "albumartist", METADATA_KEY_ALBUMARTIST },
        { "composer", METADATA_KEY_COMPOSER },
        { "genre", METADATA_KEY_GENRE },
        { "title", METADATA_KEY_TITLE },
        { "year", METADATA_KEY_YEAR },
        { "duration", METADATA_KEY_DURATION },
        { "writer", METADATA_KEY_WRITER },
        { "compilation", METADATA_KEY_COMPILATION },
        { "isdrm", METADATA_KEY_IS_DRM },
        { "width", METADATA_KEY_VIDEO_WIDTH },
        { "height", METADATA_KEY_VIDEO_HEIGHT },
    };
    static const size_t kNumEntries = sizeof(kKeyMap) / sizeof(kKeyMap[0]);
    //循环遍历
    for (size_t i = 0; i < kNumEntries; ++i) {
        const char *value;
        if ((value = mRetriever->extractMetadata(kKeyMap[i].key)) != NULL) {
            //设置tag和value到MyMediaScannerClient中，稍后分析
            status = client.addStringTag(kKeyMap[i].tag, value);
            if (status != OK) {
                return MEDIA_SCAN_RESULT_ERROR;
            }
        }
    }
    return MEDIA_SCAN_RESULT_OK;
}

从上面的分析中，设置tag和value是通过MyMediaScannerClient调用的，在MyMediaScannerClient的父类MediaScannerClient有addStringTag方法，在方法中又调用了子类MyMediaScannerClient的handleStringTag方法。

status_t MediaScannerClient::addStringTag(const char* name, const char* value)
{
    //调用子类的handleStringTag方法
    handleStringTag(name, value);
    return OK;
}

virtual status_t handleStringTag(const char* name, const char* value)
{
    jstring nameStr, valueStr;
    //获取字符串的值
    if ((nameStr = mEnv->NewStringUTF(name)) == NULL) {
        mEnv->ExceptionClear();
        return NO_MEMORY;
    }
    char *cleaned = NULL;
    //如果value的值不是utf-8编码，则需要特殊处理
    if (!isValidUtf8(value)) {
        cleaned = strdup(value);
        char *chp = cleaned;
        char ch;
        while ((ch = *chp)) {
            if (ch & 0x80) {
                *chp = '?';
            }
            chp++;
        }
        value = cleaned;
    }
    //将处理完成的值赋值到新的字符串valueStr中
    valueStr = mEnv->NewStringUTF(value);
    //释放资源
    free(cleaned);
    if (valueStr == NULL) {
        mEnv->DeleteLocalRef(nameStr);
        mEnv->ExceptionClear();
        return NO_MEMORY;
    }
    //调用java层MyMediaScanner的handleStringTag方法
    mEnv->CallVoidMethod(
        mClient, mHandleStringTagMethodID, nameStr, valueStr);
    mEnv->DeleteLocalRef(nameStr);
    mEnv->DeleteLocalRef(valueStr);
    return checkAndClearExceptionFromCallback(mEnv, "handleStringTag");
}

此时在native层中又去调用java层的方法了，此处调用的是handleStringTag方法。

public void handleStringTag(String name, String value) {
    if (name.equalsIgnoreCase("title") || name.startsWith("title;")) {
        // Don't trim() here, to preserve the special \001 character
        // used to force sorting. The media provider will trim() before
        // inserting the title in to the database.
        //将tag信息中的value值都赋值到了成员变量中
        mTitle = value;
    } else if (name.equalsIgnoreCase("artist") || name.startsWith("artist;")) {
        mArtist = value.trim();
    } else if (name.equalsIgnoreCase("albumartist") || name.startsWith("albumartist;")
            || name.equalsIgnoreCase("band") || name.startsWith("band;")) {
        mAlbumArtist = value.trim();
    ... ...
}

到此文件的读取过程分析完成了，这些成员变量装填完成之后就会调用到endFile方法中，进行更新数据库了。

2.2 IPC方式

由于发广播的方式无法实时地获取连接的状态，所以Android又提供了一种查询方法，就是通过IPC，也就是进程间通信的方式去启动扫描，然后获取扫描的状态。

2.2.1流程图

2.2.2MediaScannerConnection.java

/**
 * MediaScannerConnection provides a way for applications to pass a
 * newly created or downloaded media file to the media scanner service.
 * The media scanner service will read metadata from the file and add
 * the file to the media content provider.
 * The MediaScannerConnectionClient provides an interface for the
 * media scanner service to return the Uri for a newly scanned file
 * to the client of the MediaScannerConnection class.
 */

通过注释可以看出，MediaScannerConnection可以提供另一种非发广播的方式去主动扫描文件，他的调用过程是跨进程的，扫描的结果会通过回调函数获得。
在MediaScannerConnection内部提供了两种方式去供客户端使用，一种是实现接口和回调方法，另一种是使用代理模式所提供的静态方法。

（1）实现接口
首先通过构造方法新建实例，并且设置相关的成员变量。然后在客户端处调用connect方法，去绑定service，并且调用requestScanFile方法去跨进程调用MediaScannerService中的方法。当连接到MediaScannerService后回调客户端onMediaScannerConnected方法，当MediaScannerService扫描完成后，回调客户端onScanCompleted方法，整个过程完成。

//监听扫描完成的接口
public interface OnScanCompletedListener {
    public void onScanCompleted(String path, Uri uri);
}
//客户端需要实现的接口，同时也是在服务端所获取的客户端的实例
public interface MediaScannerConnectionClient extends OnScanCompletedListener {
    public void onMediaScannerConnected();
    public void onScanCompleted(String path, Uri uri);
}
//构造方法，传入的参数是客户端的上下文环境和客户端的实例
public MediaScannerConnection(Context context, MediaScannerConnectionClient client) {
    mContext = context;
    mClient = client;
}
// ServiceConnection的回调方法，当service连接时回调
public void onServiceConnected(ComponentName className, IBinder service) {
    synchronized (this) {
        //获取IMediaScannerService的实例mService
        mService = IMediaScannerService.Stub.asInterface(service);
        if (mService != null && mClient != null) {
            //当service连接上时，回调到客户端的onMediaScannerConnected方法
            mClient.onMediaScannerConnected();
        }
    }
}
// IMediaScannerListener是AIDL文件，只有一个方法scanCompleted
//这里获取了服务端IMediaScannerListener的实例
private final IMediaScannerListener.Stub mListener = new IMediaScannerListener.Stub() {
    public void scanCompleted(String path, Uri uri) {
        MediaScannerConnectionClient client = mClient;
        if (client != null) {
            //当回调到scanCompleted时，调用客户端的onScanCompleted方法
            client.onScanCompleted(path, uri);
        }
    }
};
//此方法是在客户端处调用，传入需要扫描的路径和文件类型
public void scanFile(String path, String mimeType) {
    synchronized (this) {
        if (mService == null || !mConnected) {
            throw new IllegalStateException("not connected to MediaScannerService");
        }
        try {
            //调用IMediaScannerService的方法
            mService.requestScanFile(path, mimeType, mListener);
        } catch (RemoteException e) {
        }
    }
}
//在客户端调用方法，bindService到MediaScannerService
public void connect() {
    synchronized (this) {
        if (!mConnected) {
            Intent intent = new Intent(IMediaScannerService.class.getName());
            intent.setComponent(
                    new ComponentName("com.android.providers.media",
                            "com.android.providers.media.MediaScannerService"));
            mContext.bindService(intent, this, Context.BIND_AUTO_CREATE);
            mConnected = true;
        }
    }
}

MediaScannerConnection部分分析完成，可以看出在connect方法中去绑定了远程的MediaScannerService，接下来分析在MediaScannerService完成的操作。

@Override
public IBinder onBind(Intent intent){
    return mBinder;
}

//在绑定之后获取到了服务端的实例，实现requestScanFile的具体方法
private final IMediaScannerService.Stub mBinder = 
        new IMediaScannerService.Stub() {
    //此处是requestScanFile实现的具体方法
    public void requestScanFile(String path, String mimeType, IMediaScannerListener listener){
        Bundle args = new Bundle();
        //将相关的参数都放入到了bundle中
        args.putString("filepath", path);
        args.putString("mimetype", mimeType);
        if (listener != null) {
            args.putIBinder("listener", listener.asBinder());
        }
        // 用startService的启动方式去启动，传入bundle
        startService(new Intent(MediaScannerService.this,
                MediaScannerService.class).putExtras(args));
    }
    //此处是scanFile实现的具体方法
    public void scanFile(String path, String mimeType) {
        requestScanFile(path, mimeType, null);
    }
};

//在onStartCommand方法中将intent的值发送到了ServiceHandler处理
private final class ServiceHandler extends Handler {
    @Override
    public void handleMessage(Message msg)
    {
        Bundle arguments = (Bundle) msg.obj;
        String filePath = arguments.getString("filepath");

        try {
            if (filePath != null) {
                //从intent中获取IBinder对象
                IBinder binder = arguments.getIBinder("listener");
                //获取IMediaScannerListener的实例
                IMediaScannerListener listener = (binder == null ? null :
                       IMediaScannerListener.Stub.asInterface(binder));
                Uri uri = null;
                try {
                    uri = scanFile(filePath, arguments.getString("mimetype"));
                } catch (Exception e) {
                    Log.e(TAG, "Exception scanning file", e);
                }
                if (listener != null) {
                    //查询完成后，回调到IMediaScannerListener，客户端处也随之回调
                    listener.scanCompleted(filePath, uri);
                }
            ... ...

在MediaScannerService的主要作用就是接受intent，调用scanFile方法扫描，扫描完成之后调用回调方法，给客户端回调。

（2） 静态方法实现
在MediaScannerConnection也可以通过提供的静态方法去实现扫描。其原理就是实现代理模式，远程代理客户端的实例进行相关操作，客户端只需要传入相应的参数即可，不需要手动连接service等操作，比较方便实用。

public static void scanFile(Context context, String[] paths, String[] mimeTypes,
        OnScanCompletedListener callback) {
    //实例化ClientProxy，并给构造函数传参
    ClientProxy client = new ClientProxy(paths, mimeTypes, callback);
    //实例化MediaScannerConnection，并给构造函数传参
    MediaScannerConnection connection = new MediaScannerConnection(context, client);
    client.mConnection = connection;
    //调用connect函数
    connection.connect();
}

//客户端的远程代理类
static class ClientProxy implements MediaScannerConnectionClient {
    final String[] mPaths;
    final String[] mMimeTypes;
    final OnScanCompletedListener mClient;
    MediaScannerConnection mConnection;
    int mNextPath;
    //构造函数，配置参数
    ClientProxy(String[] paths, String[] mimeTypes, OnScanCompletedListener client) {
        mPaths = paths;
        mMimeTypes = mimeTypes;
        mClient = client;
    }
    //实现回调方法
    public void onMediaScannerConnected() {
        scanNextPath();
    }
    public void onScanCompleted(String path, Uri uri) {
        if (mClient != null) {
            mClient.onScanCompleted(path, uri);
        }
        scanNextPath();
    }
    //因为传入的路径是数组，进行循环扫描
    void scanNextPath() {
        if (mNextPath >= mPaths.length) {
            mConnection.disconnect();
            return;
        }
        String mimeType = mMimeTypes != null ? mMimeTypes[mNextPath] : null;
        mConnection.scanFile(mPaths[mNextPath], mimeType);
        mNextPath++;
    }
}

所以对于客户端来说，实现此静态方法去扫描，只需要传入上下文，查询的路径（可以是多个路径，用数组表示），文件类型和监听器即可，不需要考虑其他，比较方便使用。