[hudi] branch master updated: [HUDI-6264] Assert that bootstrap base path is different than path (#8808)

[yihua]

This is an automated email from the ASF dual-hosted git repository.

yihua pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/hudi.git


The following commit(s) were added to refs/heads/master by this push:
     new 51d28fd75f5 [HUDI-6264] Assert that bootstrap base path is different 
than path (#8808)
51d28fd75f5 is described below

commit 51d28fd75f5deb9f2f5a96f3522cf613ed0e95ee
Author: Jon Vexler <jbvex...@gmail.com>
AuthorDate: Tue May 30 12:27:50 2023 -0400

    [HUDI-6264] Assert that bootstrap base path is different than path (#8808)
    
    Co-authored-by: Jonathan Vexler <=>
    Co-authored-by: Y Ethan Guo <ethan.guoyi...@gmail.com>
---
 .../src/main/java/org/apache/hudi/cli/commands/BootstrapCommand.java  | 4 ++++
 .../src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala         | 3 ++-
 .../spark/sql/hudi/command/procedures/RunBootstrapProcedure.scala     | 4 ++++
 .../org/apache/hudi/utilities/deltastreamer/BootstrapExecutor.java    | 4 ++++
 4 files changed, 14 insertions(+), 1 deletion(-)

diff --git 
a/hudi-cli/src/main/java/org/apache/hudi/cli/commands/BootstrapCommand.java 
b/hudi-cli/src/main/java/org/apache/hudi/cli/commands/BootstrapCommand.java
index 98cf9fc0d90..4f046df6198 100644
--- a/hudi-cli/src/main/java/org/apache/hudi/cli/commands/BootstrapCommand.java
+++ b/hudi-cli/src/main/java/org/apache/hudi/cli/commands/BootstrapCommand.java
@@ -82,6 +82,10 @@ public class BootstrapCommand {
           defaultValue = "") final String[] configs)
       throws IOException, InterruptedException, URISyntaxException {
 
+    if (targetPath.equals(srcPath)) {
+      throw new IllegalArgumentException("srcPath and targetPath must be 
different");
+    }
+
     String sparkPropertiesPath =
         
Utils.getDefaultPropertiesFile(JavaConverters.mapAsScalaMapConverter(System.getenv()).asScala());
 
diff --git 
a/hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala
 
b/hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala
index f353f78b063..8b35bf43403 100644
--- 
a/hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala
+++ 
b/hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala
@@ -54,7 +54,7 @@ import 
org.apache.hudi.internal.schema.utils.AvroSchemaEvolutionUtils.reconcileN
 import org.apache.hudi.internal.schema.utils.{AvroSchemaEvolutionUtils, 
SerDeHelper}
 import org.apache.hudi.keygen.constant.KeyGeneratorOptions
 import org.apache.hudi.keygen.factory.HoodieSparkKeyGeneratorFactory
-import 
org.apache.hudi.keygen.factory.HoodieSparkKeyGeneratorFactory.{createKeyGenerator,
 getKeyGeneratorClassName}
+import 
org.apache.hudi.keygen.factory.HoodieSparkKeyGeneratorFactory.getKeyGeneratorClassName
 import org.apache.hudi.keygen.{BaseKeyGenerator, KeyGenUtils, 
SparkKeyGeneratorInterface, TimestampBasedAvroKeyGenerator, 
TimestampBasedKeyGenerator}
 import org.apache.hudi.metrics.Metrics
 import org.apache.hudi.sync.common.HoodieSyncConfig
@@ -703,6 +703,7 @@ object HoodieSparkSqlWriter {
     val bootstrapBasePath = hoodieConfig.getStringOrThrow(BASE_PATH,
       s"'${BASE_PATH.key}' is required for '${BOOTSTRAP_OPERATION_OPT_VAL}'" +
         " operation'")
+    assert(!path.equals(bootstrapBasePath), "Bootstrap base path and Hudi 
table base path must be different")
     val bootstrapIndexClass = hoodieConfig.getStringOrDefault(INDEX_CLASS_NAME)
 
     var schema: String = null
diff --git 
a/hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/procedures/RunBootstrapProcedure.scala
 
b/hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/procedures/RunBootstrapProcedure.scala
index 458d618e92b..c2f18edaeeb 100644
--- 
a/hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/procedures/RunBootstrapProcedure.scala
+++ 
b/hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/command/procedures/RunBootstrapProcedure.scala
@@ -95,6 +95,10 @@ class RunBootstrapProcedure extends BaseProcedure with 
ProcedureBuilder with Log
 
     properties.setProperty(HoodieBootstrapConfig.BASE_PATH.key, bootstrapPath)
 
+    if (bootstrapPath.equals(basePath)) {
+      throw new IllegalArgumentException("bootstrap_path and base_path must be 
different")
+    }
+
     if (!StringUtils.isNullOrEmpty(keyGeneratorClass) && 
KeyGeneratorType.getNames.contains(keyGeneratorClass.toUpperCase(Locale.ROOT))) 
{
       properties.setProperty(HoodieWriteConfig.KEYGENERATOR_TYPE.key, 
keyGeneratorClass.toUpperCase(Locale.ROOT))
     }
diff --git 
a/hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/BootstrapExecutor.java
 
b/hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/BootstrapExecutor.java
index 07cfbeb847d..fc7ecc0f156 100644
--- 
a/hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/BootstrapExecutor.java
+++ 
b/hudi-utilities/src/main/java/org/apache/hudi/utilities/deltastreamer/BootstrapExecutor.java
@@ -202,6 +202,10 @@ public class BootstrapExecutor implements Serializable {
       }
     }
 
+    if (cfg.targetBasePath.equals(bootstrapBasePath)) {
+      throw new IllegalArgumentException("Bootstrap source base path and Hudi 
table base path must be different");
+    }
+
     HoodieTableMetaClient.PropertyBuilder builder = 
HoodieTableMetaClient.withPropertyBuilder()
         .fromProperties(props)
         .setTableType(cfg.tableType)