From 3d3ecd77411bfdd1c72be51616c46a6caf839be2 Mon Sep 17 00:00:00 2001 From: OopsOutOfMemory Date: Fri, 6 Feb 2015 13:20:10 -0800 Subject: [SPARK-5586][Spark Shell][SQL] Make `sqlContext` available in spark shell Result is like this ``` 15/02/05 13:41:22 INFO SparkILoop: Created spark context.. Spark context available as sc. 15/02/05 13:41:22 INFO SparkILoop: Created sql context.. SQLContext available as sqlContext. scala> sq sql sqlContext sqlParser sqrt ``` Author: OopsOutOfMemory Closes #4387 from OopsOutOfMemory/sqlContextInShell and squashes the following commits: c7f5203 [OopsOutOfMemory] auto-import sql() function e160697 [OopsOutOfMemory] Merge branch 'sqlContextInShell' of https://github.com/OopsOutOfMemory/spark into sqlContextInShell 37c0a16 [OopsOutOfMemory] auto detect hive support a9c59d9 [OopsOutOfMemory] rename and reduce range of imports 6b9e309 [OopsOutOfMemory] Merge branch 'master' into sqlContextInShell cae652f [OopsOutOfMemory] make sqlContext available in spark shell --- .../main/scala/org/apache/spark/repl/SparkILoop.scala | 19 +++++++++++++++++++ .../scala/org/apache/spark/repl/SparkILoopInit.scala | 10 ++++++++++ .../src/main/scala/org/apache/spark/repl/Main.scala | 18 ++++++++++++++++++ .../main/scala/org/apache/spark/repl/SparkILoop.scala | 12 +++++++++++- 4 files changed, 58 insertions(+), 1 deletion(-) (limited to 'repl') diff --git a/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoop.scala b/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoop.scala index 72c1a98999..b4db3df795 100644 --- a/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoop.scala +++ b/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoop.scala @@ -45,6 +45,7 @@ import scala.reflect.api.{Mirror, TypeCreator, Universe => ApiUniverse} import org.apache.spark.Logging import org.apache.spark.SparkConf import org.apache.spark.SparkContext +import org.apache.spark.sql.SQLContext import org.apache.spark.util.Utils /** The Scala interactive shell. It provides a read-eval-print loop @@ -130,6 +131,7 @@ class SparkILoop( // NOTE: Must be public for visibility @DeveloperApi var sparkContext: SparkContext = _ + var sqlContext: SQLContext = _ override def echoCommandMessage(msg: String) { intp.reporter printMessage msg @@ -1016,6 +1018,23 @@ class SparkILoop( sparkContext } + @DeveloperApi + def createSQLContext(): SQLContext = { + val name = "org.apache.spark.sql.hive.HiveContext" + val loader = Utils.getContextOrSparkClassLoader + try { + sqlContext = loader.loadClass(name).getConstructor(classOf[SparkContext]) + .newInstance(sparkContext).asInstanceOf[SQLContext] + logInfo("Created sql context (with Hive support)..") + } + catch { + case cnf: java.lang.ClassNotFoundException => + sqlContext = new SQLContext(sparkContext) + logInfo("Created sql context..") + } + sqlContext + } + private def getMaster(): String = { val master = this.master match { case Some(m) => m diff --git a/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoopInit.scala b/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoopInit.scala index 99bd777c04..0cf2de6d39 100644 --- a/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoopInit.scala +++ b/repl/scala-2.10/src/main/scala/org/apache/spark/repl/SparkILoopInit.scala @@ -127,7 +127,17 @@ private[repl] trait SparkILoopInit { _sc } """) + command(""" + @transient val sqlContext = { + val _sqlContext = org.apache.spark.repl.Main.interp.createSQLContext() + println("SQL context available as sqlContext.") + _sqlContext + } + """) command("import org.apache.spark.SparkContext._") + command("import sqlContext.implicits._") + command("import sqlContext.sql") + command("import org.apache.spark.sql.Dsl._") } } diff --git a/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala b/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala index 69e44d4f91..dc25692749 100644 --- a/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala +++ b/repl/scala-2.11/src/main/scala/org/apache/spark/repl/Main.scala @@ -19,6 +19,7 @@ package org.apache.spark.repl import org.apache.spark.util.Utils import org.apache.spark._ +import org.apache.spark.sql.SQLContext import scala.tools.nsc.Settings import scala.tools.nsc.interpreter.SparkILoop @@ -34,6 +35,7 @@ object Main extends Logging { "-Yrepl-outdir", s"${outputDir.getAbsolutePath}", "-Yrepl-sync"), true) val classServer = new HttpServer(conf, outputDir, new SecurityManager(conf)) var sparkContext: SparkContext = _ + var sqlContext: SQLContext = _ var interp = new SparkILoop // this is a public var because tests reset it. def main(args: Array[String]) { @@ -74,6 +76,22 @@ object Main extends Logging { sparkContext } + def createSQLContext(): SQLContext = { + val name = "org.apache.spark.sql.hive.HiveContext" + val loader = Utils.getContextOrSparkClassLoader + try { + sqlContext = loader.loadClass(name).getConstructor(classOf[SparkContext]) + .newInstance(sparkContext).asInstanceOf[SQLContext] + logInfo("Created sql context (with Hive support)..") + } + catch { + case cnf: java.lang.ClassNotFoundException => + sqlContext = new SQLContext(sparkContext) + logInfo("Created sql context..") + } + sqlContext + } + private def getMaster: String = { val master = { val envMaster = sys.env.get("MASTER") diff --git a/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoop.scala b/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoop.scala index 2507273059..1bd2a69914 100644 --- a/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoop.scala +++ b/repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoop.scala @@ -66,8 +66,18 @@ class SparkILoop(in0: Option[BufferedReader], protected val out: JPrintWriter) println("Spark context available as sc.") _sc } - """) + """) + command( """ + @transient val sqlContext = { + val _sqlContext = org.apache.spark.repl.Main.createSQLContext() + println("SQL context available as sqlContext.") + _sqlContext + } + """) command("import org.apache.spark.SparkContext._") + command("import sqlContext.implicits._") + command("import sqlContext.sql") + command("import org.apache.spark.sql.Dsl._") } } -- cgit v1.2.3