[SPARK-4075] [Deploy] Jar url validation is not enough for Jar file

In deploy.ClientArguments.isValidJarUrl, the url is checked as follows.

    def isValidJarUrl(s: String): Boolean = s.matches("(.+):(.+)jar")

So, it allows like 'hdfs:file.jar' (no authority).

Author: Kousuke Saruta <sarutak@oss.nttdata.co.jp>

Closes #2925 from sarutak/uri-syntax-check-improvement and squashes the following commits:

cf06173 [Kousuke Saruta] Improved URI syntax checking

(cherry picked from commit 098f83c7ccd7dad9f9228596da69fe5f55711a52)
Signed-off-by: Andrew Or <andrew@databricks.com>

1 files changed, 10 insertions, 1 deletions

diff --git a/core/src/main/scala/org/apache/spark/deploy/ClientArguments.scala b/core/src/main/scala/org/apache/spark/deploy/ClientArguments.scala
index 39150deab8..4e802e02c4 100644
--- a/core/src/main/scala/org/apache/spark/deploy/ClientArguments.scala
+++ b/core/src/main/scala/org/apache/spark/deploy/ClientArguments.scala
@@ -17,6 +17,8 @@
 
 package org.apache.spark.deploy
 
+import java.net.{URI, URISyntaxException}
+
 import scala.collection.mutable.ListBuffer
 
 import org.apache.log4j.Level
@@ -114,5 +116,12 @@ private[spark] class ClientArguments(args: Array[String]) {
 }
 
 object ClientArguments {
-  def isValidJarUrl(s: String): Boolean = s.matches("(.+):(.+)jar")
+  def isValidJarUrl(s: String): Boolean = {
+    try {
+      val uri = new URI(s)
+      uri.getScheme != null && uri.getAuthority != null && s.endsWith("jar")
+    } catch {
+      case _: URISyntaxException => false
+    }
+  }
 }