Spark JDBC

object ReadData extends App {
  // 初始化 Spark 会话
  val spark: SparkSession = SparkSession.builder.appName("example").master("local").getOrCreate

  val jdbcURL = "jdbc:ch://localhost:8123/default"
  val query: String = "select * from example_table where id > 2"

  //---------------------------------------------------------------------------------------------------
  // 通过 jdbc 方法从 ClickHouse 读取表
  //---------------------------------------------------------------------------------------------------
  val connectionProperties = new Properties()
  connectionProperties.put("user", "default")
  connectionProperties.put("password", "123456")

  val df1: Dataset[Row] = spark.read.
    jdbc(jdbcURL, s"($query)", connectionProperties)

  df1.show()
  //---------------------------------------------------------------------------------------------------
  // 通过 load 方法从 ClickHouse 读取表
  //---------------------------------------------------------------------------------------------------
  val df2: Dataset[Row] = spark.read
    .format("jdbc")
    .option("url", jdbcURL)
    .option("user", "default")
    .option("password", "123456")
    .option("query", query)
    .load()

  df2.show()

  // 停止 Spark 会话// 停止 Spark 会话
  spark.stop()

}

from pyspark.sql import SparkSession

jar_files = [
    "jars/clickhouse-jdbc-X.X.X-SNAPSHOT-all.jar"
]

# 使用 JAR 初始化 Spark 会话
spark = SparkSession.builder \
    .appName("example") \
    .master("local") \
    .config("spark.jars", ",".join(jar_files)) \
    .getOrCreate()

url = "jdbc:ch://localhost:8123/default"
user = "your_user" 
password = "your_password"  
query = "select * from example_table where id > 2"
driver = "com.clickhouse.jdbc.ClickHouseDriver"

df = (spark.read
      .format('jdbc')
      .option('driver', driver)
      .option('url', url)
      .option('user', user)
      .option('password', password).option(
    'query', query).load())

df.show()

   CREATE TEMPORARY VIEW jdbcTable
           USING org.apache.spark.sql.jdbc
           OPTIONS (
                   url "jdbc:ch://localhost:8123/default", 
                   dbtable "schema.tablename",
                   user "username",
                   password "password",
                   driver "com.clickhouse.jdbc.ClickHouseDriver" 
           );
           
   SELECT * FROM jdbcTable;

object WriteData extends App {

  val spark: SparkSession = SparkSession.builder.appName("example").master("local").getOrCreate

  // JDBC 连接信息
  val jdbcUrl: String = "jdbc:ch://localhost:8123/default"
  val jdbcProperties: Properties = new Properties
  jdbcProperties.put("user", "default")
  jdbcProperties.put("password", "123456")

  // 创建一个示例 DataFrame

  val rows = Seq(Row(1, "John"), Row(2, "Doe"))

  val schema = List(
    StructField("id", DataTypes.IntegerType, nullable = false),
    StructField("name", StringType, nullable = true)
  )

  val df: DataFrame = spark.createDataFrame(
    spark.sparkContext.parallelize(rows),
    StructType(schema)
  )
  
  //---------------------------------------------------------------------------------------------------//---------------------------------------------------------------------------------------------------
  // 使用 jdbc 方法将 df 写入 ClickHouse
  //---------------------------------------------------------------------------------------------------//---------------------------------------------------------------------------------------------------

  df.write
    .mode(SaveMode.Append)
    .jdbc(jdbcUrl, "example_table", jdbcProperties)

  //---------------------------------------------------------------------------------------------------//---------------------------------------------------------------------------------------------------
  // 使用 save 方法将 df 写入 ClickHouse
  //---------------------------------------------------------------------------------------------------//---------------------------------------------------------------------------------------------------

  df.write
    .format("jdbc")
    .mode("append")
    .option("url", jdbcUrl)
    .option("dbtable", "example_table")
    .option("user", "default")
    .option("password", "123456")
    .save()

  // 停止 Spark session// 停止 Spark session
  spark.stop()

}

from pyspark.sql import SparkSession
from pyspark.sql import Row

jar_files = [
    "jars/clickhouse-jdbc-X.X.X-SNAPSHOT-all.jar"
]

# 使用 JAR 初始化 Spark session
spark = SparkSession.builder \
    .appName("example") \
    .master("local") \
    .config("spark.jars", ",".join(jar_files)) \
    .getOrCreate()

# 创建 DataFrame
data = [Row(id=11, name="John"), Row(id=12, name="Doe")]
df = spark.createDataFrame(data)

url = "jdbc:ch://localhost:8123/default"
user = "your_user" 
password = "your_password"  
driver = "com.clickhouse.jdbc.ClickHouseDriver"

# 将 DataFrame 写入 ClickHouse
df.write \
    .format("jdbc") \
    .option("driver", driver) \
    .option("url", url) \
    .option("user", user) \
    .option("password", password) \
    .option("dbtable", "example_table") \
    .mode("append") \
    .save()

   CREATE TEMPORARY VIEW jdbcTable
           USING org.apache.spark.sql.jdbc
           OPTIONS (
                   url "jdbc:ch://localhost:8123/default", 
                   dbtable "schema.tablename",
                   user "username",
                   password "password",
                   driver "com.clickhouse.jdbc.ClickHouseDriver" 
           );
   -- resultTable 可以通过 df.createTempView 或 Spark SQL 创建
   INSERT INTO TABLE jdbcTable
                SELECT * FROM resultTable;