参考 Apache Spark 的源码自定义实现 Logging 日志打印工具

Posted 2021-12-18 Shockang

前言

本文隶属于专栏《1000个问题搞定大数据技术体系》，该专栏为笔者原创，引用请注明来源，不足和错误之处请在评论区帮忙指出，谢谢！

本专栏目录结构和参考文献请见1000个问题搞定大数据技术体系

正文

下面的日志打印工具可以无脑应用于任何的 scala 工程中。

package com.shockang.study.spark.internal

import org.apache.log4j._
import org.slf4j.Logger, LoggerFactory
import org.slf4j.bridge.SLF4JBridgeHandler
import org.slf4j.impl.StaticLoggerBinder

/**
 * 日志的工具特质。参考 Apache Spark 的源码实现： [[org.apache.spark.internal.Logging]]
 */
trait Logging 

  // log 字段设置为 transient
  // 这样实现了 Logging 接口的对象就可以序列化，并且可以在其他机器上使用。
  @transient private var log_ : Logger = null

  // 获得当前对象的日志名称
  protected def logName = 
    // 忽略 Scala 对象名称的 "$" 符号后面的内容
    this.getClass.getName.stripSuffix("$")
  

  // 得到或者创建当前对象的 logger
  protected def log: Logger = 
    if (log_ == null) 
      initializeLogIfNecessary(false)
      log_ = LoggerFactory.getLogger(logName)
    
    log_
  

  // =======================================================================
  // 处理单个字符串的方法
  // =======================================================================

  protected def logInfo(msg: => String): Unit = 
    if (log.isInfoEnabled) log.info(msg)
  

  protected def logDebug(msg: => String): Unit = 
    if (log.isDebugEnabled) log.debug(msg)
  

  protected def logTrace(msg: => String): Unit = 
    if (log.isTraceEnabled) log.trace(msg)
  

  protected def logWarning(msg: => String): Unit = 
    if (log.isWarnEnabled) log.warn(msg)
  

  protected def logError(msg: => String): Unit = 
    if (log.isErrorEnabled) log.error(msg)
  

  // =======================================================================
  // 处理字符串和 Throwable 的方法
  // =======================================================================

  protected def logInfo(msg: => String, throwable: Throwable): Unit = 
    if (log.isInfoEnabled) log.info(msg, throwable)
  

  protected def logDebug(msg: => String, throwable: Throwable): Unit = 
    if (log.isDebugEnabled) log.debug(msg, throwable)
  

  protected def logTrace(msg: => String, throwable: Throwable): Unit = 
    if (log.isTraceEnabled) log.trace(msg, throwable)
  

  protected def logWarning(msg: => String, throwable: Throwable): Unit = 
    if (log.isWarnEnabled) log.warn(msg, throwable)
  

  protected def logError(msg: => String, throwable: Throwable): Unit = 
    if (log.isErrorEnabled) log.error(msg, throwable)
  

  protected def isTraceEnabled(): Boolean = 
    log.isTraceEnabled
  

  // =======================================================================
  // 日志对象的初始化
  // =======================================================================

  protected def initializeLogIfNecessary(isInterpreter: Boolean): Unit = 
    initializeLogIfNecessary(isInterpreter, silent = false)
  

  protected def initializeLogIfNecessary(
                                          isInterpreter: Boolean,
                                          silent: Boolean = false): Boolean = 
    if (!Logging.initialized) 
      Logging.initLock.synchronized 
        if (!Logging.initialized) 
          initializeLogging(isInterpreter, silent)
          return true
        
      
    
    false
  

  // 测试使用
  private[spark] def initializeForcefully(isInterpreter: Boolean, silent: Boolean): Unit = 
    initializeLogging(isInterpreter, silent)
  

  private def initializeLogging(isInterpreter: Boolean, silent: Boolean): Unit = 
    // 如果 Log4j 1.2 使用了，但是没有初始化，加载默认的 log4j 配置文件。
    if (Logging.isLog4j12()) 
      val log4j12Initialized = LogManager.getRootLogger.getAllAppenders.hasMoreElements
      // scalastyle:off println
      if (!log4j12Initialized) 
        Logging.defaultLog4jConfig = true
        val defaultLogProps = "conf/log4j-defaults.properties"
        Option(getClass.getClassLoader.getResource(defaultLogProps)) match 
          case Some(url) =>
            PropertyConfigurator.configure(url)
            if (!silent) 
              System.err.println(s"使用默认的 log4j 配置文件: $defaultLogProps")
            
          case None =>
            System.err.println(s"无法加载配置文件： $defaultLogProps")
        
      

      val rootLogger = LogManager.getRootLogger()
      if (Logging.defaultRootLevel == null) 
        Logging.defaultRootLevel = rootLogger.getLevel()
      
      // scalastyle:on println
    
    Logging.initialized = true

    // 强制调用 slf4j 来初始化，避免通过多线程来触发：
    // http://mailman.qos.ch/pipermail/slf4j-dev/2010-April/002956.html
    log
  


private[spark] object Logging 
  @volatile private var initialized = false
  @volatile private var defaultRootLevel: Level = null
  @volatile private var defaultLog4jConfig = false

  val initLock = new Object()
  try 
    // 如果用户为了使用 JUL 进行日志打印，移除了 slf4j 到 JUL 的桥接方法，
    // 这里我们使用反射来处理这种场景。
    // scalastyle:off classforname
    val bridgeClass = Class.forName("org.slf4j.bridge.SLF4JBridgeHandler",
      true, Thread.currentThread().getContextClassLoader)
      .asInstanceOf[Class[SLF4JBridgeHandler]]
    // scalastyle:on classforname
    bridgeClass.getMethod("removeHandlersForRootLogger").invoke(null)
    val installed = bridgeClass.getMethod("isInstalled").invoke(null).asInstanceOf[Boolean]
    if (!installed) 
      bridgeClass.getMethod("install").invoke(null)
    
   catch 
    case e: ClassNotFoundException => // 无法进行日志打印所以直接静默失败
  

  /**
   * 将日志记录系统标记为未初始化。
   * 这将尽最大努力将日志记录系统重置为其初始状态，以便下一个使用日志记录的类再次触发初始化。
   */
  def uninitialize(): Unit = initLock.synchronized 
    if (isLog4j12()) 
      if (defaultLog4jConfig) 
        defaultLog4jConfig = false
        LogManager.resetConfiguration()
       else 
        val rootLogger = LogManager.getRootLogger()
        rootLogger.setLevel(defaultRootLevel)
      
    
    this.initialized = false
  

  private def isLog4j12(): Boolean = 
    // 这里将区别绑定的是 log4j 1.2（使用的 org.slf4j.impl.Log4jLoggerFactory）
    // 还是 log4j 2.0（使用的 org.apache.logging.slf4j.Log4jLoggerFactory）
    val binderClass = StaticLoggerBinder.getSingleton.getLoggerFactoryClassStr
    "org.slf4j.impl.Log4jLoggerFactory".equals(binderClass)
  

scalastyle 配置请参考我的这篇博客——Scala 的代码风格怎么统一？这份 scalastyle 配置你可以无脑复制