Github user dongjoon-hyun commented on a diff in the pull request:
https://github.com/apache/spark/pull/21495#discussion_r193936286
--- Diff:
repl/scala-2.11/src/main/scala/org/apache/spark/repl/SparkILoopInterpreter.scala
---
@@ -21,8 +21,22 @@ import scala.collection.mutable
import scala.tools.nsc.Settings
import scala.tools.nsc.interpreter._
-class SparkILoopInterpreter(settings: Settings, out: JPrintWriter) extends
IMain(settings, out) {
- self =>
+class SparkILoopInterpreter(settings: Settings, out: JPrintWriter,
initializeSpark: () => Unit)
+ extends IMain(settings, out) { self =>
+
+ /**
+ * We override `initializeSynchronous` to initialize Spark *after*
`intp` is properly initialized
+ * and *before* the REPL sees any files in the private `loadInitFiles`
functions, so that
+ * the Spark context is visible in those files.
+ *
+ * This is a bit of a hack, but there isn't another hook available to us
at this point.
+ *
+ * See the discussion in Scala community
https://github.com/scala/bug/issues/10913 for detail.
+ */
+ override def initializeSynchronous(): Unit = {
+ super.initializeSynchronous()
+ initializeSpark()
--- End diff --
Can we upgrade to the corresponding `jline` version together in this PR,
@dbtsai ?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
