Spark sql之集成Hbase-connectors

源码地址

GitHub - apache/hbase-connectors: Apache HBase Connectors

Hbase官方文档

https://hbase.apache.org/book.html#_sparksqldataframes

编译源码：

• 1.下载源码
• 2.下载maven（3.6以上版本）
• 3.编译父工程

cd hbase-connectors-master
mvn clean install -DskipTests //编译 跳过测试，此阶段会很长时间

• 4.编译子工程

cd /hbase-connectors-master/spark

//版本按自己的改
mvn -Dspark.version=3.0.3 -Dscala.version=2.12.10 -Dhadoop-three.version=3.1.1 -Dscala.binary.version=2.12 -Dhbase.version=2.3.5 clean install

• 5.将jar包加入本地仓库

//加入本地仓库
mvn install:install-file -DgroupId=cn.huorong.hbase.connectors.spark -DartifactId=hbase-spark -Dversion=2.3.5_spark-3.0.3 -Dpackaging=jar -Dfile=C:\Maven\repository3.8\huorong\hbase-spark-1.0.1-SNAPSHOT.jar

//在pom文件引用
<dependency>
	<groupId>cn.huorong.hbase.connectors.spark</groupId>
	<artifactId>hbase-spark</artifactId>
	<version>2.3.5_spark-3.0.3 </version>
</dependency>

spark sql集成hbase demo

• HBaseRecord

package hbase_connector

case class HBaseRecord(col0: String,
                       col1: Boolean,
                       col2: Double,
                       col3: Float,
                       col4: Int,
                       col5: Long,
                       col6: Short,
                       col7: String,
                       col8: Byte)
object HBaseRecord
{
  def apply(i: Int, t: String): HBaseRecord = {
    val s = s"""row${"%03d".format(i)}"""
    HBaseRecord(s,
      i % 2 == 0,
      i.toDouble,
      i.toFloat,
      i,
      i.toLong,
      i.toShort,
      s"String$i: $t",
      i.toByte)
  }
}

• df_hbase

object df_hbase {
  def main(args: Array[String]): Unit = {
    val session: SparkSession = SparkSession.builder().appName(this.getClass.getSimpleName).master("local").getOrCreate()
    val sc: SparkContext = session.sparkContext

    val hbaseConf: Configuration = HBaseConfiguration.create()
    hbaseConf.set("hbase.zookeeper.properties.quorum","192.168.1.76,192.168.1.73,192.168.1.77")
    hbaseConf.set("hbase.zookeeper.properties.property.clientPort","2181")

    //一定要创建 hbasecontext
    val hbaseContext = new HBaseContext(sc, hbaseConf, null)

    val records: immutable.IndexedSeq[HBaseRecord] = (0 to 255).map { i => HBaseRecord(i, "extra") }

    //根据rdd创建df
    val recordDF: DataFrame = session.createDataFrame(records)
    recordDF.show(false)


    //定义映射的catalog
    val catalog = s"""{
                 |"table":{"namespace":"default", "name":"table1"},
                 |"rowkey":"key",
                 |"columns":{
                 |"col0":{"cf":"rowkey", "col":"key", "type":"string"},
                 |"col1":{"cf":"cf1", "col":"col1", "type":"boolean"},
                 |"col2":{"cf":"cf2", "col":"col2", "type":"double"},
                 |"col3":{"cf":"cf3", "col":"col3", "type":"float"},
                 |"col4":{"cf":"cf4", "col":"col4", "type":"int"},
                 |"col5":{"cf":"cf5", "col":"col5", "type":"bigint"},
                 |"col6":{"cf":"cf6", "col":"col6", "type":"smallint"},
                 |"col7":{"cf":"cf7", "col":"col7", "type":"string"},
                 |"col8":{"cf":"cf8", "col":"col8", "type":"tinyint"}
                 |}
                 |}""".stripMargin

    //写入数据
    recordDF.write
      .format("org.apache.hadoop.hbase.spark")
      .option(HBaseTableCatalog.tableCatalog,catalog)
      .option(HBaseTableCatalog.newTable,"5") //写入到5个分区内
      .mode(SaveMode.Overwrite)//覆盖模式
      .save()

    //读取数据
    val table1DF: DataFrame = session.read
      .format("org.apache.hadoop.hbase.spark")
      .option(HBaseTableCatalog.tableCatalog, catalog)
      .option("hbase.spark.pushdown.columnfilter", false) //使用filter过滤列时需要加上
      .load()
    table1DF.show(400,false)
		
+------+-----+----+----+----+----+----+---------------+----+
|col0  |col1 |col2|col3|col4|col5|col6|col7           |col8|
+------+-----+----+----+----+----+----+---------------+----+
|row000|true |0.0 |0.0 |0   |0   |0   |String0: extra |0   |
|row001|false|1.0 |1.0 |1   |1   |1   |String1: extra |1   |
|row002|true |2.0 |2.0 |2   |2   |2   |String2: extra |2   |
|row003|false|3.0 |3.0 |3   |3   |3   |String3: extra |3   |
|row004|true |4.0 |4.0 |4   |4   |4   |String4: extra |4   |
|row005|false|5.0 |5.0 |5   |5   |5   |String5: extra |5   |
|row006|true |6.0 |6.0 |6   |6   |6   |String6: extra |6   |
|row007|false|7.0 |7.0 |7   |7   |7   |String7: extra |7   |
|row008|true |8.0 |8.0 |8   |8   |8   |String8: extra |8   |
|row009|false|9.0 |9.0 |9   |9   |9   |String9: extra |9   |
|row010|true |10.0|10.0|10  |10  |10  |String10: extra|10  |
|row011|false|11.0|11.0|11  |11  |11  |String11: extra|11  |
|row012|true |12.0|12.0|12  |12  |12  |String12: extra|12  |
|row013|false|13.0|13.0|13  |13  |13  |String13: extra|13  |
|row014|true |14.0|14.0|14  |14  |14  |String14: extra|14  |
|row015|false|15.0|15.0|15  |15  |15  |String15: extra|15  |
|row016|true |16.0|16.0|16  |16  |16  |String16: extra|16  |
|row017|false|17.0|17.0|17  |17  |17  |String17: extra|17  |
|row018|true |18.0|18.0|18  |18  |18  |String18: extra|18  |
|row019|false|19.0|19.0|19  |19  |19  |String19: extra|19  |
+------+-----+----+----+----+----+----+---------------+----+
only showing top 20 rows

列过滤

table1DF.filter("col4==20").show(false) //报错
table1DF.filter(row=>{ row.getAs[Int]("col4") == 20 }) //不报错


加上，都不报错了
.option("hbase.spark.pushdown.columnfilter", false) //使用filter过滤列时需要加上

范围查找

session.read
.format("org.apache.hadoop.hbase.spark")
.option(HBaseTableCatalog.tableCatalog, catalog)
.option(HBaseSparkConf.TIMERANGE_START,"1661238243000")
.option(HBaseSparkConf.TIMERANGE_END,"1661238244000")