Android Honeycomb加载键盘布局文件过程

原文地址：http://zhougaofeng.ixiezi.com/2011/04/19/honeycomb-keylayout/

Andriod启动过程中是如何正确加载.kl和.kcm的键盘布局文件？本文就从Honeycomb代码入手，详细介绍开机启动时键盘布局文件的加载过程。
Honeycom 相较与之前的版本，加入了一个.idc后缀的配置文件，使在不修改系统代码的前提下，我们就可以使用自定义的键盘布局文件，系统中与键盘布局相关的目录为 /system/usr/keychars，/system/usr/keylayout，/system/usr/idc

一、系统启动过程中SystemServer添加WindowManagerService
Slog.i(TAG, "Window Manager");
wm = WindowManagerService.main(context, power,
        factoryTest != SystemServer.FACTORY_TEST_LOW_LEVEL);
ServiceManager.addService(Context.WINDOW_SERVICE, wm);

((ActivityManagerService)ServiceManager.getService("activity"))
        .setWindowManager(wm);

二、WindowManagerService.java的构造函数，在加载键盘布局方面做了两件事情：1.初始化，构造一个InputManager实例；2.启动，由InputManager.java start()函数实现
private WindowManagerService(Context context, PowerManagerService pm,
            ……..
            ……..

        mInputManager = new InputManager(context, this); //构造InputManager实例

        PolicyThread thr = new PolicyThread(mPolicy, this, context, pm);
        thr.start();

        synchronized (thr) {
            while (!thr.mRunning) {
                try {
                    thr.wait();
                } catch (InterruptedException e) {
                }
            }
        }

        mInputManager.start(); //调用InputManager.java start()函数

        // Add ourself to the Watchdog monitors.
        Watchdog.getInstance().addMonitor(this);
}

三、InputManager.java是本地c代码的包装类，对com_android_server_InputManager.cpp接口函数进行包装，以提供其他java文件调取。
1.初始化，InputManager.java构造函数中的init()最后调用nativeInit(mCallbacks)，
public InputManager(Context context, WindowManagerService windowManagerService) {
    this.mContext = context;
    this.mWindowManagerService = windowManagerService;
   
    this.mCallbacks = new Callbacks();
   
    init(); //调用init()函数
}

private void init() {
    Slog.i(TAG, "Initializing input manager");
    nativeInit(mCallbacks); //java接口，由本地函数实现
}

2. 启动，InputManager.java的start()最后调用nativeStart()：
public void start() {
    Slog.i(TAG, "Starting input manager");
    nativeStart(); //java接口，由本地函数实现
}

四、com_android_server_InputManager.cpp实现InutManager.java的nativeInit(mCallbacks和nativeStart()， 当然还实现了其他功能的接口函数，这里不再介绍，对于android如何实现java和c之间的转换，我想对于了解jni的来说不难理解。不懂的可以看此 文章学习：http://hi.baidu.com/kellyvivian/blog/item /09cfb541179d2f3387947397.html
1.初始化，android_server_InputManager_nativeInit在被执行的时候会new一个NativeInputManager(callbacks)实例，NativeInputManager(callbacks)接着又会new一个InputManager(eventHub, this, this)实例
static void android_server_InputManager_nativeInit(JNIEnv* env, jclass clazz,
        jobject callbacks) {
    if (gNativeInputManager == NULL) {
        gNativeInputManager = new NativeInputManager(callbacks);
    } else {
        LOGE("Input manager already initialized.");
        jniThrowRuntimeException(env, "Input manager already initialized.");
    }
}

NativeInputManager::NativeInputManager(jobject callbacksObj) :
    mFilterTouchEvents(-1), mFilterJumpyTouchEvents(-1), mVirtualKeyQuietTime(-1),
    mMaxEventsPerSecond(-1) {
    JNIEnv* env = jniEnv();

    mCallbacksObj = env->NewGlobalRef(callbacksObj);

    …….
    sp<EventHub> eventHub = new EventHub();
    mInputManager = new InputManager(eventHub, this, this);
}

2.启动，android_server_InputManager_nativeStart中gNativeInputManager->getInputManager()->start()最终调用的是InputManager.cpp的start()函数
static void android_server_InputManager_nativeStart(JNIEnv* env, jclass clazz) {
    if (checkInputManagerUnitialized(env)) {
        return;
    }

    status_t result = gNativeInputManager->getInputManager()->start();
    if (result) {
        jniThrowRuntimeException(env, "Input manager could not be started.");
    }
}

五、InputManager.cpp中主要有三个函数：initialize()初始化函数，在构造函数中调用；start()开启线程函数；stop()取消线程函数，在虚构函数中调用。
1.初始化，InputManager.cpp构造函数调用initialize()，期间new一个InputReaderThread线程
InputManager::InputManager(
        const sp<EventHubInterface>& eventHub,
        const sp<InputReaderPolicyInterface>& readerPolicy,
        const sp<InputDispatcherPolicyInterface>& dispatcherPolicy) {
    mDispatcher = new InputDispatcher(dispatcherPolicy);
    mReader = new InputReader(eventHub, readerPolicy, mDispatcher);
    initialize();
}

void InputManager::initialize() {
    mReaderThread = new InputReaderThread(mReader);
    mDispatcherThread = new InputDispatcherThread(mDispatcher);
}

2.启动，mReaderThread->run("InputReader", PRIORITY_URGENT_DISPLAY)开启初始化时new的InputReaderThread线程
status_t InputManager::start() {
    ……..

    result = mReaderThread->run("InputReader", PRIORITY_URGENT_DISPLAY);
    if (result) {
        LOGE("Could not start InputReader thread due to error %d.", result);

        mDispatcherThread->requestExit();
        return result;
    }

    return OK;
}

六、InputReader.cpp中定义了InputReaderThread类，继承于Thread类
1.初始化，InputReaderThread构造函数，初始化一个Thread类
InputReaderThread::InputReaderThread(const sp<InputReaderInterface>& reader) :
        Thread(/*canCallJava*/ true), mReader(reader) {
}

2.启动，run启动线程，Thread run()方法又调用InputReaderThread 的虚函数threadLoop()，接着调用InputReader的loopOnce()方法，最后调用EventHub.cpp的getEvent(& rawEvent)方法
bool InputReaderThread::threadLoop() {
    mReader->loopOnce();
    return true;
}
void InputReader::loopOnce() {
    RawEvent rawEvent;
    mEventHub->getEvent(& rawEvent);

#if DEBUG_RAW_EVENTS
    LOGD("Input event: device=%d type=0x%x scancode=%d keycode=%d value=%d",
            rawEvent.deviceId, rawEvent.type, rawEvent.scanCode, rawEvent.keyCode,
            rawEvent.value);
#endif

    process(& rawEvent);
}

七、EventHub.cpp是android输入系统的硬件抽象层，维护输入设备的运行，包括Keyboard、 TouchScreen、TraceBall等。
EventHub.cpp中依次执行getEvent()–>openPlatformInput()–>scanDir(DEVICE_PATH)–> openDevice(devname)

bool EventHub::openPlatformInput(void) {
    /*
     * Open platform-specific input device(s).
     */
    int res, fd;
    ………
    // Reserve fd index 0 for inotify.
    struct pollfd pollfd;
    pollfd.fd = fd;
    pollfd.events = POLLIN;
    pollfd.revents = 0;
    mFds.push(pollfd);
    mDevices.push(NULL);

    res = scanDir(DEVICE_PATH); //DEVICE_PATH = "/dev/input"
    if(res < 0) {
        LOGE("scan dir failed for %s\n", DEVICE_PATH);
    }

    return true;
}

int EventHub::scanDir(const char *dirname)
{
    ……
        openDevice(devname);
    }
    closedir(dir);
    return 0;
}

openDevice方法会打开/dev/input目录下的所有设备文件，读取name、version、id等设备信息，然后执行loadConfiguration()方法，如果键盘设备就会执行loadKeyMap()这个方法
int EventHub::openDevice(const char *devicePath) {
    ……

    // Load the configuration file for the device.
    loadConfiguration(device);

    ……

    if ((device->classes & INPUT_DEVICE_CLASS_KEYBOARD) != 0) {
        // Load the keymap for the device.
        status_t status = loadKeyMap(device);

        ……
        }

        ……
}

Honeycomb与之前版本不同之处是加入loadConfiguration()方 法，它获取与当前设备驱动Vendor、Product、Version匹配的配置文件名，或者是Vendor、Product匹配的配置文件名，具体可 查看Input.cpp中getInputDeviceConfigurationFilePathByDeviceIdentifie和 getInputDeviceConfigurationFilePathByName方法。
如： kernel/ drivers/input/keyboard/atkbd.c键盘驱动中定义了 input_dev->id.vendor = 0×0001; input_dev->id.product = 0×0001; input_dev->id.version = 0xab41，那么与之对应的配置名为Vendor_0001_Product_0001_Version_ad41.idc，返回这个文件的全路径并赋 值给device->configurationFile。如果/system/user/idc下存在此文件，接下来调用 PropertyMap.cpp的load()方法解析该配置文件并将解析后的信息保存到device->configuration中。
void EventHub::loadConfiguration(Device* device) {
    device->configurationFile = getInputDeviceConfigurationFilePathByDeviceIdentifier(
            device->identifier, INPUT_DEVICE_CONFIGURATION_FILE_TYPE_CONFIGURATION);
    if (device->configurationFile.isEmpty()) {
        LOGD("No input device configuration file found for device ‘%s’.",
                device->identifier.name.string());
    } else {
        status_t status = PropertyMap::load(device->configurationFile,
                &device->configuration);
        if (status) {
            LOGE("Error loading input device configuration file for device ‘%s’.  "
                    "Using default configuration.",
                    device->identifier.name.string());
        }
    }
}

EventHub.cpp中loadKeyMap又调用了Keyboard.cpp的KeyMap::load()方法
status_t EventHub::loadKeyMap(Device* device) {
    return device->keyMap.load(device->identifier, device->configuration);
}

八、在Keyboard.cpp的load方法中，首先判断deviceConfiguration参数是否为空，deviceConfiguration的赋值就是上面loadConfiguration()方法所做的工作。
如 果有.idc的配置文件，那么获取key为keyboard.layout的value给keyLayoutName和key为 keyboard.characterMap的value给keyCharacterMapName，最后调用loadKeyLayout和 loadKeyCharacterMap方法加载此键盘布局文件；如果没有对应的.idc配置文件，则deviceConfiguration为空，就会 接着执行probeKeyMap(deviceIdenfifier, String8("Generic"))方法
status_t KeyMap::load(const InputDeviceIdentifier& deviceIdenfifier,
        const PropertyMap* deviceConfiguration) {
    // Use the configured key layout if available.
    if (deviceConfiguration) {
        String8 keyLayoutName;
        if (deviceConfiguration->tryGetProperty(String8("keyboard.layout"),
                keyLayoutName)) {
            status_t status = loadKeyLayout(deviceIdenfifier, keyLayoutName);
            if (status == NAME_NOT_FOUND) {
                LOGE("Configuration for keyboard device ‘%s’ requested keyboard layout ‘%s’ but "
                        "it was not found.",
                        deviceIdenfifier.name.string(), keyLayoutName.string());
            }
        }

        String8 keyCharacterMapName;
        if (deviceConfiguration->tryGetProperty(String8("keyboard.characterMap"),
                keyCharacterMapName)) {
            status_t status = loadKeyCharacterMap(deviceIdenfifier, keyCharacterMapName);
            if (status == NAME_NOT_FOUND) {
                LOGE("Configuration for keyboard device ‘%s’ requested keyboard character "
                        "map ‘%s’ but it was not found.",
                        deviceIdenfifier.name.string(), keyLayoutName.string());
            }
        }

        if (isComplete()) {
            return OK;
        }
    }

    ……
    if (probeKeyMap(deviceIdenfifier, String8("Generic"))) {
        return OK;
    }
    ……
}

probeKeyMap方法判断名为Gerneric的布局文件是否存在，若存在就会调用loadKeyLayout和loadKeyCharacterMap方法加载此键盘布局文件
bool KeyMap::probeKeyMap(const InputDeviceIdentifier& deviceIdentifier,
        const String8& keyMapName) {
    if (!haveKeyLayout()) {
        loadKeyLayout(deviceIdentifier, keyMapName);
    }
    if (!haveKeyCharacterMap()) {
        loadKeyCharacterMap(deviceIdentifier, keyMapName);
    }
    return isComplete();
}

至此，Android Honeycomb已经正确加载了键盘布局文件，那么我们如何定制和使用自己的键盘布局文件呢？

附件：qwerty.idc配置文件内容
# Copyright (C) 2010 The Android Open Source Project
#
# 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.

#
# Emulator keyboard configuration file #1.
#

touch.deviceType = touchScreen
touch.orientationAware = 1

keyboard.layout = qwerty
keyboard.characterMap = qwerty
keyboard.orientationAware = 1
keyboard.builtIn = 1

cursor.mode = navigation
cursor.orientationAware = 1