Github user tdas commented on a diff in the pull request:
https://github.com/apache/spark/pull/11034#discussion_r51957616
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/FileStreamSource.scala
---
@@ -0,0 +1,151 @@
+/*
+ * 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.execution.streaming
+
+import java.io.{BufferedWriter, OutputStreamWriter}
+
+import scala.collection.mutable.ArrayBuffer
+
+import org.apache.hadoop.fs.{FileStatus, FileSystem, Path}
+
+import org.apache.spark.Logging
+import org.apache.spark.sql.{DataFrame, SQLContext}
+import org.apache.spark.sql.types.{StringType, StructType}
+import org.apache.spark.util.collection.OpenHashSet
+
+/**
+ * A very simple source that reads text files from the given directory as
they appear.
+ */
+class FileStreamSource(
+ sqlContext: SQLContext,
+ metadataPath: String,
+ path: String,
+ dataSchema: Option[StructType],
+ dataFrameBuilder: Array[String] => DataFrame) extends Source with
Logging {
+
+ import sqlContext.implicits._
+
+ /** Returns the schema of the data from this source */
+ override lazy val schema: StructType = {
+ dataSchema.getOrElse {
+ val filesPresent = fetchAllFiles()
+ if (filesPresent.isEmpty) {
+ new StructType().add("value", StringType)
+ } else {
+ // There are some existing files. Use them to infer the schema
+ dataFrameBuilder(filesPresent.toArray).schema
+ }
+ }
+ }
+
+ /** Returns the maximum offset that can be retrieved from the source. */
+ def fetchMaxOffset(): LongOffset = synchronized {
+ val filesPresent = fetchAllFiles()
+ val newFiles = new ArrayBuffer[String]()
+ filesPresent.foreach { file =>
+ if (!seenFiles.contains(file)) {
+ logDebug(s"new file: $file")
+ newFiles.append(file)
+ seenFiles.add(file)
+ } else {
+ logDebug(s"old file: $file")
+ }
+ }
+
+ if (newFiles.nonEmpty) {
+ maxBatchFile += 1
+ writeBatch(maxBatchFile, newFiles)
+ }
+
+ new LongOffset(maxBatchFile)
+ }
+
+ def currentOffset: LongOffset = synchronized {
+ new LongOffset(maxBatchFile)
+ }
+
+ /**
+ * Returns the next batch of data that is available after `start`, if
any is available.
+ */
+ override def getNextBatch(start: Option[Offset]): Option[Batch] = {
+ val startId =
start.map(_.asInstanceOf[LongOffset].offset).getOrElse(-1L)
+ val end = fetchMaxOffset()
+ val endId = end.offset
+
+ val batchFiles = (startId + 1 to endId).filter(_ >= 0).map(i =>
s"$metadataPath/$i")
+ if (batchFiles.nonEmpty) {
+ logDebug(s"Producing files from batches ${startId + 1}:$endId")
+ logDebug(s"Batch files: $batchFiles")
+
+ // Probably does not need to be a spark job...
+ val files = sqlContext
+ .read
+ .text(batchFiles: _*)
+ .as[String]
+ .collect()
+ logDebug(s"Streaming ${files.mkString(", ")}")
+ Some(new Batch(end, dataFrameBuilder(files)))
+ } else {
+ None
+ }
+ }
+
+ private def sparkContext = sqlContext.sparkContext
+
+ private val fs = FileSystem.get(sparkContext.hadoopConfiguration)
+ private val existingBatchFiles = fetchAllBatchFiles()
+ private val existingBatchIds =
existingBatchFiles.map(_.getPath.getName.toInt)
+ private var maxBatchFile = if (existingBatchIds.isEmpty) -1 else
existingBatchIds.max
--- End diff --
nit: Isnt this `maxBatchId` rather than the file (since its an int, rather
than a string)?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
