
com.github.fsanaulla.chronicler.spark.rdd.package.scala Maven / Gradle / Ivy
The newest version!
/*
* Copyright 2018-2021 Faiaz Sanaulla
*
* 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.
*/
package com.github.fsanaulla.chronicler.spark
import com.github.fsanaulla.chronicler.core.alias.ErrorOr
import com.github.fsanaulla.chronicler.core.either
import com.github.fsanaulla.chronicler.core.model.InfluxWriter
import com.github.fsanaulla.chronicler.spark.core.{CallbackHandler, WriteConfig}
import com.github.fsanaulla.chronicler.sync.io.InfluxIO
import com.github.fsanaulla.chronicler.sync.shared.InfluxConfig
import org.apache.spark.rdd.RDD
import scala.reflect.ClassTag
import scala.util.{Failure, Success, Try}
package object rdd {
/** Extension that will provide static methods for saving RDDs to InfluxDB
*
* @param rdd
* - [[org.apache.spark.rdd.RDD]]
* @tparam T
* - inner type
*/
implicit final class RddOps[T](private val rdd: RDD[T]) extends AnyVal {
private def handleResponse(
handler: Option[CallbackHandler],
response: Try[ErrorOr[Int]]
): Unit =
if (handler.isDefined) {
val Some(rh) = handler
response match {
case Success(Right(code)) => rh.onSuccess(code)
// application level issues
case Success(Left(ex)) => rh.onApplicationFailure(ex)
// connection/network level issues
case Failure(ex) => rh.onNetworkFailure(ex)
}
}
/** Write [[org.apache.spark.rdd.RDD]] to InfluxDB, specifying database measurement
*
* @param dbName
* - database name
* @param measName
* - measurement name
* @param handler
* - defined callbacks for responses
* @param dataInfo
* - data characteristics
*/
def saveToInfluxDBMeas(
dbName: String,
measName: String,
handler: Option[CallbackHandler] = None,
dataInfo: WriteConfig = WriteConfig.default
)(implicit wr: InfluxWriter[T], conf: InfluxConfig, tt: ClassTag[T]): Unit = {
rdd.foreachPartition { partition =>
val client = InfluxIO(conf)
val meas = client.measurement[T](dbName, measName)
partition.sliding(dataInfo.batchSize, dataInfo.batchSize).foreach { batch =>
val response = meas.bulkWrite(
batch,
dataInfo.consistency,
dataInfo.precision,
dataInfo.retentionPolicy
)
// check if rh is defined
handleResponse(handler, response)
}
client.close()
}
}
/** Write [[org.apache.spark.rdd.RDD]] to InfluxDB, with measurements that generated dynamicly
*
* @param dbName
* - database name
* @param handler
* - defined callbacks for responses
* @param dataInfo
* - data characteristics
*/
def saveToInfluxDB(
dbName: String,
handler: Option[CallbackHandler] = None,
dataInfo: WriteConfig = WriteConfig.default
)(implicit wr: InfluxWriter[T], conf: InfluxConfig, tt: ClassTag[T]): Unit = {
rdd.foreachPartition { partition =>
val client = InfluxIO(conf)
val db = client.database(dbName)
partition.sliding(dataInfo.batchSize, dataInfo.batchSize).foreach { batch =>
val ethPoints = either.seq(batch.map(wr.write))
val response = ethPoints
.fold(Failure(_), Success(_))
.flatMap { rows =>
db.bulkWriteNative(
rows,
dataInfo.consistency,
dataInfo.precision,
dataInfo.retentionPolicy
)
}
handleResponse(handler, response)
}
client.close()
}
}
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy