1
0

[HUDI-92] Provide reasonable names for Spark DAG stages in HUDI. (#1289)

This commit is contained in:
Prashant Wason
2020-07-19 10:29:25 -07:00
committed by GitHub
parent 1aae437257
commit b71f25f210
25 changed files with 79 additions and 10 deletions

View File

@@ -86,7 +86,7 @@ public class TestHoodieBloomIndex extends HoodieClientTestHarness {
@BeforeEach
public void setUp() throws Exception {
initSparkContexts("TestHoodieBloomIndex");
initSparkContexts();
initPath();
initFileSystem();
// We have some records to be tagged (two different partitions)

View File

@@ -71,7 +71,7 @@ public class TestHoodieGlobalBloomIndex extends HoodieClientTestHarness {
@BeforeEach
public void setUp() throws Exception {
initSparkContexts("TestHoodieGlobalBloomIndex");
initSparkContexts();
initPath();
// We have some records to be tagged (two different partitions)
String schemaStr = FileIOUtils.readAsUTFString(getClass().getResourceAsStream("/exampleSchema.txt"));

View File

@@ -60,7 +60,7 @@ public class TestHoodieCommitArchiveLog extends HoodieClientTestHarness {
public void init() throws Exception {
initDFS();
initPath();
initSparkContexts("TestHoodieCommitArchiveLog");
initSparkContexts();
hadoopConf = dfs.getConf();
hadoopConf.addResource(dfs.getConf());
dfs.mkdirs(new Path(basePath));

View File

@@ -57,7 +57,7 @@ public class TestHoodieMergeHandle extends HoodieClientTestHarness {
@BeforeEach
public void setUp() throws Exception {
initSparkContexts("TestHoodieMergeHandle");
initSparkContexts();
initPath();
initFileSystem();
initTestDataGenerator();

View File

@@ -66,7 +66,7 @@ public class TestHoodieCompactor extends HoodieClientTestHarness {
@BeforeEach
public void setUp() throws Exception {
// Initialize a local spark env
initSparkContexts("TestHoodieCompactor");
initSparkContexts();
// Create a temp folder as the base path
initPath();

View File

@@ -41,6 +41,8 @@ import org.apache.hudi.table.HoodieTable;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.SQLContext;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.TestInfo;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@@ -56,7 +58,8 @@ import java.util.concurrent.atomic.AtomicInteger;
public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness implements Serializable {
private static final Logger LOG = LoggerFactory.getLogger(HoodieClientTestHarness.class);
private String testMethodName;
protected transient JavaSparkContext jsc = null;
protected transient Configuration hadoopConf = null;
protected transient SQLContext sqlContext;
@@ -82,6 +85,15 @@ public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness im
protected transient MiniDFSCluster dfsCluster;
protected transient DistributedFileSystem dfs;
@BeforeEach
public void setTestMethodName(TestInfo testInfo) {
if (testInfo.getTestMethod().isPresent()) {
testMethodName = testInfo.getTestMethod().get().getName();
} else {
testMethodName = "Unknown";
}
}
/**
* Initializes resource group for the subclasses of {@link HoodieClientTestBase}.
*/
@@ -113,7 +125,7 @@ public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness im
*/
protected void initSparkContexts(String appName) {
// Initialize a local spark env
jsc = new JavaSparkContext(HoodieClientTestUtils.getSparkConfForTest(appName));
jsc = new JavaSparkContext(HoodieClientTestUtils.getSparkConfForTest(appName + "#" + testMethodName));
jsc.setLogLevel("ERROR");
hadoopConf = jsc.hadoopConfiguration();
@@ -122,11 +134,11 @@ public abstract class HoodieClientTestHarness extends HoodieCommonTestHarness im
}
/**
* Initializes the Spark contexts ({@link JavaSparkContext} and {@link SQLContext}) with a default name
* <b>TestHoodieClient</b>.
* Initializes the Spark contexts ({@link JavaSparkContext} and {@link SQLContext})
* with a default name matching the name of the class.
*/
protected void initSparkContexts() {
initSparkContexts("TestHoodieClient");
initSparkContexts(this.getClass().getSimpleName());
}
/**

View File

@@ -146,9 +146,30 @@ public class HoodieClientTestUtils {
new RandomAccessFile(path, "rw").setLength(length);
}
/**
* Returns a Spark config for this test.
*
* The following properties may be set to customize the Spark context:
* SPARK_EVLOG_DIR: Local directory where event logs should be saved. This
* allows viewing the logs with spark-history-server.
*
* @note When running the tests using maven, use the following syntax to set
* a property:
* mvn -DSPARK_XXX=yyy ...
*
* @param appName A name for the Spark application. Shown in the Spark web UI.
* @return A Spark config
*/
public static SparkConf getSparkConfForTest(String appName) {
SparkConf sparkConf = new SparkConf().setAppName(appName)
.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer").setMaster("local[8]");
String evlogDir = System.getProperty("SPARK_EVLOG_DIR");
if (evlogDir != null) {
sparkConf.set("spark.eventLog.enabled", "true");
sparkConf.set("spark.eventLog.dir", evlogDir);
}
return HoodieReadClient.addHoodieSupport(sparkConf);
}