hvanhovell commented on code in PR #40368: URL: https://github.com/apache/spark/pull/40368#discussion_r1141997143
########## connector/connect/server/src/main/scala/org/apache/spark/sql/connect/service/SparkConnectAddArtifactsHandler.scala: ########## @@ -0,0 +1,239 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You 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 org.apache.spark.sql.connect.service + +import com.google.common.io.CountingOutputStream +import io.grpc.stub.StreamObserver +import java.nio.file.{Files, Path, Paths} +import java.util.zip.{CheckedOutputStream, CRC32} +import scala.collection.mutable +import scala.util.control.NonFatal + +import org.apache.spark.connect.proto +import org.apache.spark.connect.proto.{AddArtifactsRequest, AddArtifactsResponse} +import org.apache.spark.connect.proto.AddArtifactsResponse.ArtifactSummary +import org.apache.spark.sql.connect.artifact.SparkConnectArtifactManager +import org.apache.spark.util.Utils + +/** + * Handles [[AddArtifactsRequest]]s for the [[SparkConnectService]]. + * + * @param responseObserver + */ +class SparkConnectAddArtifactsHandler(val responseObserver: StreamObserver[AddArtifactsResponse]) + extends StreamObserver[AddArtifactsRequest] { + + // Temporary directory where artifacts are rebuilt from the bytes sent over the wire. + protected val stagingDir: Path = Utils.createTempDir().toPath + protected val stagedArtifacts: mutable.Buffer[StagedArtifact] = + mutable.Buffer.empty[StagedArtifact] + // If not null, indicates the currently active chunked artifact that is being rebuilt from + // several [[AddArtifactsRequest]]s. + private var chunkedArtifact: StagedChunkedArtifact = _ + private var holder: SessionHolder = _ + private def artifactManager: SparkConnectArtifactManager = + SparkConnectArtifactManager.getOrCreateArtifactManager + + override def onNext(req: AddArtifactsRequest): Unit = { + if (this.holder == null) { + this.holder = SparkConnectService.getOrCreateIsolatedSession( + req.getUserContext.getUserId, + req.getSessionId) + } + + if (req.hasBeginChunk) { + // The beginning of a multi-chunk artifact. + require(chunkedArtifact == null) + chunkedArtifact = writeArtifactToFile(req.getBeginChunk) + } else if (req.hasChunk) { + // We are currently processing a multi-chunk artifact + require(chunkedArtifact != null && !chunkedArtifact.isFinished) + chunkedArtifact.write(req.getChunk) + + if (chunkedArtifact.isFinished) { + chunkedArtifact.close() + // Unset the currently active chunked artifact. + chunkedArtifact = null + } + } else if (req.hasBatch) { + // Each artifact in the batch is single-chunked. + req.getBatch.getArtifactsList.forEach(artifact => writeArtifactToFile(artifact).close()) + } else { + throw new UnsupportedOperationException(s"Unsupported data transfer request: $req") + } + } + + override def onError(throwable: Throwable): Unit = { + Utils.deleteRecursively(stagingDir.toFile) + responseObserver.onError(throwable) + } + + protected def addStagedArtifactToArtifactManager(artifact: StagedArtifact): Unit = { + artifactManager.addArtifact(holder.session, artifact.path, artifact.stagedPath) + } + + /** + * Process all the staged artifacts built in this stream. + * + * @return + */ + protected def flushStagedArtifacts(): Seq[ArtifactSummary] = { + // Non-lazy transformation when using Buffer. + stagedArtifacts.map { artifact => + // We do not store artifacts that fail the CRC. The failure is reported in the artifact + // summary and it is up to the client to decide whether to retry sending the artifact. + if (artifact.getCrcStatus.contains(true)) { + addStagedArtifactToArtifactManager(artifact) + } + artifact.summary() + }.toSeq + } + + protected def cleanUpStagedArtifacts(): Unit = Utils.deleteRecursively(stagingDir.toFile) + + override def onCompleted(): Unit = { + val artifactSummaries = flushStagedArtifacts() + // Add the artifacts to the session. Review Comment: I mean update the comment :) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
