Repository: spark
Updated Branches:
  refs/heads/branch-1.3 5b96b6933 -> 5185ea9b4


[MINOR] [BUILD] Use custom temp directory during build.

Even with all the efforts to cleanup the temp directories created by
unit tests, Spark leaves a lot of garbage in /tmp after a test run.
This change overrides java.io.tmpdir to place those files under the
build directory instead.

After an sbt full unit test run, I was left with > 400 MB of temp
files. Since they're now under the build dir, it's much easier to
clean them up.

Also make a slight change to a unit test to make it not pollute the
source directory with test data.

Author: Marcelo Vanzin <van...@cloudera.com>

Closes #6653 from vanzin/unit-test-tmp and squashes the following commits:

31e2dd5 [Marcelo Vanzin] Fix tests that depend on each other.
aa92944 [Marcelo Vanzin] [minor] [build] Use custom temp directory during build.

(cherry picked from commit b16b5434ff44c42e4b3a337f9af147669ba44896)
Signed-off-by: Sean Owen <so...@cloudera.com>


Project: http://git-wip-us.apache.org/repos/asf/spark/repo
Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/5185ea9b
Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/5185ea9b
Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/5185ea9b

Branch: refs/heads/branch-1.3
Commit: 5185ea9b4df3ee73807859b70ddfca8f02f1a659
Parents: 5b96b69
Author: Marcelo Vanzin <van...@cloudera.com>
Authored: Fri Jun 5 14:11:38 2015 +0200
Committer: Sean Owen <so...@cloudera.com>
Committed: Fri Jun 5 14:16:05 2015 +0200

----------------------------------------------------------------------
 .../spark/deploy/SparkSubmitUtilsSuite.scala    | 22 +++++++++++---------
 pom.xml                                         |  4 +++-
 project/SparkBuild.scala                        |  1 +
 3 files changed, 16 insertions(+), 11 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/spark/blob/5185ea9b/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
----------------------------------------------------------------------
diff --git 
a/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala 
b/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
index a49b4db..ceff5d7 100644
--- a/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
+++ b/core/src/test/scala/org/apache/spark/deploy/SparkSubmitUtilsSuite.scala
@@ -28,9 +28,12 @@ import org.scalatest.BeforeAndAfterAll
 
 import org.apache.spark.SparkFunSuite
 import org.apache.spark.deploy.SparkSubmitUtils.MavenCoordinate
+import org.apache.spark.util.Utils
 
 class SparkSubmitUtilsSuite extends SparkFunSuite with BeforeAndAfterAll {
 
+  private var tempIvyPath: String = _
+
   private val noOpOutputStream = new OutputStream {
     def write(b: Int) = {}
   }
@@ -47,6 +50,7 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with 
BeforeAndAfterAll {
     super.beforeAll()
     // We don't want to write logs during testing
     SparkSubmitUtils.printStream = new BufferPrintStream
+    tempIvyPath = Utils.createTempDir(namePrefix = "ivy").getAbsolutePath()
   }
 
   test("incorrect maven coordinate throws error") {
@@ -90,21 +94,20 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with 
BeforeAndAfterAll {
   }
 
   test("ivy path works correctly") {
-    val ivyPath = "dummy" + File.separator +  "ivy"
     val md = SparkSubmitUtils.getModuleDescriptor
     val artifacts = for (i <- 0 until 3) yield new MDArtifact(md, s"jar-$i", 
"jar", "jar")
-    var jPaths = SparkSubmitUtils.resolveDependencyPaths(artifacts.toArray, 
new File(ivyPath))
+    var jPaths = SparkSubmitUtils.resolveDependencyPaths(artifacts.toArray, 
new File(tempIvyPath))
     for (i <- 0 until 3) {
-      val index = jPaths.indexOf(ivyPath)
+      val index = jPaths.indexOf(tempIvyPath)
       assert(index >= 0)
-      jPaths = jPaths.substring(index + ivyPath.length)
+      jPaths = jPaths.substring(index + tempIvyPath.length)
     }
     val main = MavenCoordinate("my.awesome.lib", "mylib", "0.1")
     IvyTestUtils.withRepository(main, None, None) { repo =>
       // end to end
       val jarPath = SparkSubmitUtils.resolveMavenCoordinates(main.toString, 
Option(repo),
-        Option(ivyPath), true)
-      assert(jarPath.indexOf(ivyPath) >= 0, "should use non-default ivy path")
+        Option(tempIvyPath), true)
+      assert(jarPath.indexOf(tempIvyPath) >= 0, "should use non-default ivy 
path")
     }
   }
 
@@ -123,13 +126,12 @@ class SparkSubmitUtilsSuite extends SparkFunSuite with 
BeforeAndAfterAll {
       assert(jarPath.indexOf("mylib") >= 0, "should find artifact")
     }
     // Local ivy repository with modified home
-    val dummyIvyPath = "dummy" + File.separator + "ivy"
-    val dummyIvyLocal = new File(dummyIvyPath, "local" + File.separator)
+    val dummyIvyLocal = new File(tempIvyPath, "local" + File.separator)
     IvyTestUtils.withRepository(main, None, Some(dummyIvyLocal), true) { repo 
=>
       val jarPath = SparkSubmitUtils.resolveMavenCoordinates(main.toString, 
None,
-        Some(dummyIvyPath), true)
+        Some(tempIvyPath), true)
       assert(jarPath.indexOf("mylib") >= 0, "should find artifact")
-      assert(jarPath.indexOf(dummyIvyPath) >= 0, "should be in new ivy path")
+      assert(jarPath.indexOf(tempIvyPath) >= 0, "should be in new ivy path")
     }
   }
 

http://git-wip-us.apache.org/repos/asf/spark/blob/5185ea9b/pom.xml
----------------------------------------------------------------------
diff --git a/pom.xml b/pom.xml
index 101cd6e..2883099 100644
--- a/pom.xml
+++ b/pom.xml
@@ -172,7 +172,7 @@
     <parquet.deps.scope>compile</parquet.deps.scope>
 
     <!--
-      Overridable test home. So that you can call individual pom files 
directory without
+      Overridable test home. So that you can call individual pom files 
directly without
       things breaking.
     -->
     <spark.test.home>${session.executionRootDirectory}</spark.test.home>
@@ -1200,6 +1200,7 @@
             </environmentVariables>
             <systemProperties>
               <java.awt.headless>true</java.awt.headless>
+              <java.io.tmpdir>${project.build.directory}/tmp</java.io.tmpdir>
               
<spark.test.home>${session.executionRootDirectory}</spark.test.home>
               <spark.testing>1</spark.testing>
               <spark.ui.enabled>false</spark.ui.enabled>
@@ -1230,6 +1231,7 @@
             </environmentVariables>
             <systemProperties>
               <java.awt.headless>true</java.awt.headless>
+              <java.io.tmpdir>${project.build.directory}/tmp</java.io.tmpdir>
               <spark.test.home>${spark.test.home}</spark.test.home>
               <spark.testing>1</spark.testing>
               <spark.ui.enabled>false</spark.ui.enabled>

http://git-wip-us.apache.org/repos/asf/spark/blob/5185ea9b/project/SparkBuild.scala
----------------------------------------------------------------------
diff --git a/project/SparkBuild.scala b/project/SparkBuild.scala
index 2082691..72269e8 100644
--- a/project/SparkBuild.scala
+++ b/project/SparkBuild.scala
@@ -428,6 +428,7 @@ object TestSettings {
     // launched by the tests have access to the correct test-time classpath.
     envVars in Test += ("SPARK_DIST_CLASSPATH" ->
       (fullClasspath in 
Test).value.files.map(_.getAbsolutePath).mkString(":").stripSuffix(":")),
+    javaOptions in Test += s"-Djava.io.tmpdir=$sparkHome/target/tmp",
     javaOptions in Test += "-Dspark.test.home=" + sparkHome,
     javaOptions in Test += "-Dspark.testing=1",
     javaOptions in Test += "-Dspark.port.maxRetries=100",


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org

Reply via email to