Flink/Spark中ETL的简单模版

我们往往可以忽略外界的干扰因素，避免焦虑，专心做自己想做的事情，反正焦虑又解决不了问题

引言

使用flink或者spark的时候，写好固定的模版很重要，对于一下etl的实时任务，只需要执行一个map和fliter就解决了，简单抽象了一下，用的时候穿穿参数。

父类

abstract class BashLogLineParse extends Serializable{
  var line: String= ""
  def logParse():String
  def logFilter(mes:String):Boolean
}

etl

import java.text.SimpleDateFormat
import java.util.{Date, Locale}

import com.alibaba.fastjson.JSON
import com.oneniceapp.base.BashLogLineParse
import org.apache.commons.lang3.time.FastDateFormat

class StageDataOperationPushLog extends BashLogLineParse with Serializable{

  private val YYYYMMDDHHMM_TIME_FORMAT = FastDateFormat.getInstance("dd/MMM/yyyy:HH:mm:ss Z", Locale.ENGLISH)

  override def logParse(): String = {
    val line = this.line
    val log = JSON.parseObject(line)
    val datastr = log.getString("raw")

    val data = JSON.parseObject(datastr)
    val payloadstr = data.getString("payload")
    val payload = JSON.parseObject(payloadstr)
    val notice_id = payload.getString("notice_id")
    val user_id = data.getString("receiver")
    val raw_time = data.getLong("raw_time")
    val partition_date = new SimpleDateFormat("yyyy-MM-dd").format(new Date(raw_time*1000))

    data.remove("receiver")
    data.put("user_id", user_id)
    data.put("notice_id", notice_id)
    data.put("partition_date", partition_date)

    if(notice_id==null)
      ""
    else
      data.toJSONString
  }

  override def logFilter(mes:String):Boolean ={
    if(!mes.equals(""))
      true
    else
      false
  }

}

反射方法

public class CreateObject {
    public static Object createInstance(String className){

        try {
            Class clz = Class.forName(className);
            Object obj = clz.newInstance();
            return obj;
        } catch (Exception e) {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }

        return null;
    }
}

使用

StageDataOperationPushLog的全路径

 mapDate = transction.map(line=>{
        val ls = CreateObject
          .createInstance(conf.classFile)
          .asInstanceOf[BashLogLineParse]

        ls.line = line.message
        (line.topic, line.partition, line.offset, ls.logParse())
      })
        .filter(line=>{
          val ls = CreateObject
            .createInstance(conf.classFile)
            .asInstanceOf[BashLogLineParse]
          ls.logFilter(line._4)
        })