Revert "[HUDI-4324] Remove use_jdbc config from hudi sync (#6072)" (#6160)

This reverts commit 046044c83d.

conf/hudi-defaults.conf.template

@@ -20,7 +20,7 @@
 
 # Example:
 # hoodie.datasource.hive_sync.jdbcurl            jdbc:hive2://localhost:10000
-# hoodie.datasource.hive_sync.mode               jdbc
+# hoodie.datasource.hive_sync.use_jdbc           true
 # hoodie.datasource.hive_sync.support_timestamp  false
 # hoodie.index.type                              BLOOM
 # hoodie.metadata.enable                         false

docker/demo/config/hoodie-incr.properties

@@ -28,6 +28,5 @@ hoodie.deltastreamer.source.hoodieincr.path=/docker_hoodie_sync_valid_test
 hoodie.deltastreamer.source.hoodieincr.read_latest_on_missing_ckpt=true
 # hive sync
 hoodie.datasource.hive_sync.table=docker_hoodie_sync_valid_test_2
-hoodie.datasource.hive_sync.mode=jdbc
 hoodie.datasource.hive_sync.jdbcurl=jdbc:hive2://hiveserver:10000
 hoodie.datasource.hive_sync.partition_fields=partition

docker/demo/sparksql-incremental.commands

@@ -47,7 +47,6 @@ spark.sql("select key, `_hoodie_partition_path` as datestr, symbol, ts, open, cl
     option(HoodieWriteConfig.TBL_NAME.key(), "stock_ticks_derived_mor").
     option(HoodieSyncConfig.META_SYNC_TABLE_NAME.key(), "stock_ticks_derived_mor").
     option(HoodieSyncConfig.META_SYNC_DATABASE_NAME.key(), "default").
-    option(HiveSyncConfigHolder.HIVE_SYNC_MODE.key(), "jdbc").
     option(HiveSyncConfigHolder.HIVE_URL.key(), "jdbc:hive2://hiveserver:10000").
     option(HiveSyncConfigHolder.HIVE_USER.key(), "hive").
     option(HiveSyncConfigHolder.HIVE_PASS.key(), "hive").
@@ -80,7 +79,6 @@ spark.sql("select key, `_hoodie_partition_path` as datestr, symbol, ts, open, cl
     option(HoodieWriteConfig.TBL_NAME.key(), "stock_ticks_derived_mor_bs").
     option(HoodieSyncConfig.META_SYNC_TABLE_NAME.key(), "stock_ticks_derived_mor_bs").
     option(HoodieSyncConfig.META_SYNC_DATABASE_NAME.key(), "default").
-    option(HiveSyncConfigHolder.HIVE_SYNC_MODE.key(), "jdbc").
     option(HiveSyncConfigHolder.HIVE_URL.key(), "jdbc:hive2://hiveserver:10000").
     option(HiveSyncConfigHolder.HIVE_USER.key(), "hive").
     option(HiveSyncConfigHolder.HIVE_PASS.key(), "hive").

hudi-common/src/test/java/org/apache/hudi/common/util/TestDFSPropertiesConfiguration.java

@@ -185,7 +185,7 @@ public class TestDFSPropertiesConfiguration {
     DFSPropertiesConfiguration.refreshGlobalProps();
     assertEquals(5, DFSPropertiesConfiguration.getGlobalProps().size());
     assertEquals("jdbc:hive2://localhost:10000", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.datasource.hive_sync.jdbcurl"));
-    assertEquals("jdbc", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.datasource.hive_sync.mode"));
+    assertEquals("true", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.datasource.hive_sync.use_jdbc"));
     assertEquals("false", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.datasource.hive_sync.support_timestamp"));
     assertEquals("BLOOM", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.index.type"));
     assertEquals("true", DFSPropertiesConfiguration.getGlobalProps().get("hoodie.metadata.enable"));

hudi-common/src/test/resources/external-config/hudi-defaults.conf

@@ -20,7 +20,7 @@
 
 # Example:
 hoodie.datasource.hive_sync.jdbcurl            jdbc:hive2://localhost:10000
-hoodie.datasource.hive_sync.mode               jdbc
+hoodie.datasource.hive_sync.use_jdbc           true
 hoodie.datasource.hive_sync.support_timestamp  false
 hoodie.index.type                              BLOOM
 hoodie.metadata.enable                         true

hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/configuration/FlinkOptions.java

@@ -756,6 +756,12 @@ public class FlinkOptions extends HoodieConfig {
       .defaultValue(false)
       .withDescription("Assume partitioning is yyyy/mm/dd, default false");
 
+  public static final ConfigOption<Boolean> HIVE_SYNC_USE_JDBC = ConfigOptions
+      .key("hive_sync.use_jdbc")
+      .booleanType()
+      .defaultValue(true)
+      .withDescription("Use JDBC when hive synchronization is enabled, default true");
+
   public static final ConfigOption<Boolean> HIVE_SYNC_AUTO_CREATE_DB = ConfigOptions
       .key("hive_sync.auto_create_db")
       .booleanType()

hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/sink/utils/HiveSyncContext.java

@@ -43,6 +43,7 @@ import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_TABLE_PROPERTIES;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_TABLE_SERDE_PROPERTIES;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_URL;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USER;
+import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USE_JDBC;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USE_PRE_APACHE_INPUT_FORMAT;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.METASTORE_URIS;
 import static org.apache.hudi.sync.common.HoodieSyncConfig.META_SYNC_ASSUME_DATE_PARTITION;
@@ -104,6 +105,7 @@ public class HiveSyncContext {
     props.setPropertyIfNonNull(HIVE_TABLE_SERDE_PROPERTIES.key(), conf.getString(FlinkOptions.HIVE_SYNC_TABLE_SERDE_PROPERTIES));
     props.setPropertyIfNonNull(META_SYNC_PARTITION_FIELDS.key(), String.join(",", FilePathUtils.extractHivePartitionFields(conf)));
     props.setPropertyIfNonNull(META_SYNC_PARTITION_EXTRACTOR_CLASS.key(), conf.getString(FlinkOptions.HIVE_SYNC_PARTITION_EXTRACTOR_CLASS_NAME));
+    props.setPropertyIfNonNull(HIVE_USE_JDBC.key(), String.valueOf(conf.getBoolean(FlinkOptions.HIVE_SYNC_USE_JDBC)));
     props.setPropertyIfNonNull(META_SYNC_USE_FILE_LISTING_FROM_METADATA.key(), String.valueOf(conf.getBoolean(FlinkOptions.METADATA_ENABLED)));
     props.setPropertyIfNonNull(HIVE_IGNORE_EXCEPTIONS.key(), String.valueOf(conf.getBoolean(FlinkOptions.HIVE_SYNC_IGNORE_EXCEPTIONS)));
     props.setPropertyIfNonNull(HIVE_SUPPORT_TIMESTAMP_TYPE.key(), String.valueOf(conf.getBoolean(FlinkOptions.HIVE_SYNC_SUPPORT_TIMESTAMP)));

hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/streamer/FlinkStreamerConfig.java

@@ -321,6 +321,9 @@ public class FlinkStreamerConfig extends Configuration {
   @Parameter(names = {"--hive-sync-assume-date-partitioning"}, description = "Assume partitioning is yyyy/mm/dd, default false")
   public Boolean hiveSyncAssumeDatePartition = false;
 
+  @Parameter(names = {"--hive-sync-use-jdbc"}, description = "Use JDBC when hive synchronization is enabled, default true")
+  public Boolean hiveSyncUseJdbc = true;
+
   @Parameter(names = {"--hive-sync-auto-create-db"}, description = "Auto create hive database if it does not exists, default true")
   public Boolean hiveSyncAutoCreateDb = true;
 
@@ -417,6 +420,7 @@ public class FlinkStreamerConfig extends Configuration {
     conf.setString(FlinkOptions.HIVE_SYNC_PARTITION_FIELDS, config.hiveSyncPartitionFields);
     conf.setString(FlinkOptions.HIVE_SYNC_PARTITION_EXTRACTOR_CLASS_NAME, config.hiveSyncPartitionExtractorClass);
     conf.setBoolean(FlinkOptions.HIVE_SYNC_ASSUME_DATE_PARTITION, config.hiveSyncAssumeDatePartition);
+    conf.setBoolean(FlinkOptions.HIVE_SYNC_USE_JDBC, config.hiveSyncUseJdbc);
     conf.setBoolean(FlinkOptions.HIVE_SYNC_AUTO_CREATE_DB, config.hiveSyncAutoCreateDb);
     conf.setBoolean(FlinkOptions.HIVE_SYNC_IGNORE_EXCEPTIONS, config.hiveSyncIgnoreExceptions);
     conf.setBoolean(FlinkOptions.HIVE_SYNC_SKIP_RO_SUFFIX, config.hiveSyncSkipRoSuffix);

hudi-kafka-connect/demo/config-sink-hive.json

@@ -21,6 +21,7 @@
 		"hoodie.datasource.hive_sync.table": "huditesttopic",
 		"hoodie.datasource.hive_sync.partition_fields": "date",
 		"hoodie.datasource.hive_sync.partition_extractor_class": "org.apache.hudi.hive.MultiPartKeysValueExtractor",
+		"hoodie.datasource.hive_sync.use_jdbc": "false",
 		"hoodie.datasource.hive_sync.mode": "hms",
 		"dfs.client.use.datanode.hostname": "true",
 		"hive.metastore.uris": "thrift://hivemetastore:9083",

hudi-kafka-connect/src/main/java/org/apache/hudi/connect/writers/KafkaConnectConfigs.java

@@ -172,6 +172,7 @@ public class KafkaConnectConfigs extends HoodieConfig {
   public static final String HIVE_URL = "hoodie.datasource.hive_sync.jdbcurl";
   public static final String HIVE_PARTITION_FIELDS = "hoodie.datasource.hive_sync.partition_fields";
   public static final String HIVE_PARTITION_EXTRACTOR_CLASS = "hoodie.datasource.hive_sync.partition_extractor_class";
+  public static final String HIVE_USE_JDBC = "hoodie.datasource.hive_sync.use_jdbc";
   public static final String HIVE_SYNC_MODE = "hoodie.datasource.hive_sync.mode";
   public static final String HIVE_AUTO_CREATE_DATABASE = "hoodie.datasource.hive_sync.auto_create_database";
   public static final String HIVE_IGNORE_EXCEPTIONS = "hoodie.datasource.hive_sync.ignore_exceptions";

hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/DataSourceOptions.scala

@@ -437,6 +437,10 @@ object DataSourceWriteOptions {
   val HIVE_ASSUME_DATE_PARTITION: ConfigProperty[String] = HoodieSyncConfig.META_SYNC_ASSUME_DATE_PARTITION
   @Deprecated
   val HIVE_USE_PRE_APACHE_INPUT_FORMAT: ConfigProperty[String] = HiveSyncConfigHolder.HIVE_USE_PRE_APACHE_INPUT_FORMAT
+
+  /** @deprecated Use {@link HIVE_SYNC_MODE} instead of this config from 0.9.0 */
+  @Deprecated
+  val HIVE_USE_JDBC: ConfigProperty[String] = HiveSyncConfigHolder.HIVE_USE_JDBC
   @Deprecated
   val HIVE_AUTO_CREATE_DATABASE: ConfigProperty[String] = HiveSyncConfigHolder.HIVE_AUTO_CREATE_DATABASE
   @Deprecated
@@ -496,6 +500,9 @@ object DataSourceWriteOptions {
   /** @deprecated Use {@link HIVE_USE_PRE_APACHE_INPUT_FORMAT} and its methods instead */
   @Deprecated
   val HIVE_USE_PRE_APACHE_INPUT_FORMAT_OPT_KEY = HiveSyncConfigHolder.HIVE_USE_PRE_APACHE_INPUT_FORMAT.key()
+  /** @deprecated Use {@link HIVE_USE_JDBC} and its methods instead */
+  @Deprecated
+  val HIVE_USE_JDBC_OPT_KEY = HiveSyncConfigHolder.HIVE_USE_JDBC.key()
   /** @deprecated Use {@link HIVE_AUTO_CREATE_DATABASE} and its methods instead */
   @Deprecated
   val HIVE_AUTO_CREATE_DATABASE_OPT_KEY = HiveSyncConfigHolder.HIVE_AUTO_CREATE_DATABASE.key()
@@ -686,6 +693,9 @@ object DataSourceWriteOptions {
   val DEFAULT_HIVE_ASSUME_DATE_PARTITION_OPT_VAL = HoodieSyncConfig.META_SYNC_ASSUME_DATE_PARTITION.defaultValue()
   @Deprecated
   val DEFAULT_USE_PRE_APACHE_INPUT_FORMAT_OPT_VAL = "false"
+  /** @deprecated Use {@link HIVE_USE_JDBC} and its methods instead */
+  @Deprecated
+  val DEFAULT_HIVE_USE_JDBC_OPT_VAL = HiveSyncConfigHolder.HIVE_USE_JDBC.defaultValue()
   /** @deprecated Use {@link HIVE_AUTO_CREATE_DATABASE} and its methods instead */
   @Deprecated
   val DEFAULT_HIVE_AUTO_CREATE_DATABASE_OPT_KEY = HiveSyncConfigHolder.HIVE_AUTO_CREATE_DATABASE.defaultValue()

hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/HoodieWriterUtils.scala

@@ -78,6 +78,7 @@ object HoodieWriterUtils {
     hoodieConfig.setDefaultValue(HoodieSyncConfig.META_SYNC_PARTITION_FIELDS)
     hoodieConfig.setDefaultValue(HoodieSyncConfig.META_SYNC_PARTITION_EXTRACTOR_CLASS)
     hoodieConfig.setDefaultValue(HIVE_STYLE_PARTITIONING)
+    hoodieConfig.setDefaultValue(HiveSyncConfigHolder.HIVE_USE_JDBC)
     hoodieConfig.setDefaultValue(HiveSyncConfigHolder.HIVE_CREATE_MANAGED_TABLE)
     hoodieConfig.setDefaultValue(HiveSyncConfigHolder.HIVE_SYNC_AS_DATA_SOURCE_TABLE)
     hoodieConfig.setDefaultValue(ASYNC_COMPACT_ENABLE)

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/HiveSyncConfig.java

@@ -45,6 +45,7 @@ import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_TABLE_PROPERTIES;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_TABLE_SERDE_PROPERTIES;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_URL;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USER;
+import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USE_JDBC;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USE_PRE_APACHE_INPUT_FORMAT;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.METASTORE_URIS;
 
@@ -94,6 +95,9 @@ public class HiveSyncConfig extends HoodieSyncConfig {
             + "com.uber.hoodie to org.apache.hudi. Stop using this after you migrated the table definition to "
             + "org.apache.hudi input format.")
     public Boolean usePreApacheInputFormat;
+    @Deprecated
+    @Parameter(names = {"--use-jdbc"}, description = "Hive jdbc connect url")
+    public Boolean useJdbc;
     @Parameter(names = {"--metastore-uris"}, description = "Hive metastore uris")
     public String metastoreUris;
     @Parameter(names = {"--sync-mode"}, description = "Mode to choose for Hive ops. Valid values are hms,glue,jdbc and hiveql")
@@ -138,6 +142,7 @@ public class HiveSyncConfig extends HoodieSyncConfig {
       props.setPropertyIfNonNull(HIVE_PASS.key(), hivePass);
       props.setPropertyIfNonNull(HIVE_URL.key(), jdbcUrl);
       props.setPropertyIfNonNull(HIVE_USE_PRE_APACHE_INPUT_FORMAT.key(), usePreApacheInputFormat);
+      props.setPropertyIfNonNull(HIVE_USE_JDBC.key(), useJdbc);
       props.setPropertyIfNonNull(HIVE_SYNC_MODE.key(), syncMode);
       props.setPropertyIfNonNull(METASTORE_URIS.key(), metastoreUris);
       props.setPropertyIfNonNull(HIVE_AUTO_CREATE_DATABASE.key(), autoCreateDatabase);

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/HiveSyncConfigHolder.java

@@ -52,6 +52,15 @@ public class HiveSyncConfigHolder {
       .withDocumentation("Flag to choose InputFormat under com.uber.hoodie package instead of org.apache.hudi package. "
           + "Use this when you are in the process of migrating from "
           + "com.uber.hoodie to org.apache.hudi. Stop using this after you migrated the table definition to org.apache.hudi input format");
+  /**
+   * @deprecated Use {@link #HIVE_SYNC_MODE} instead of this config from 0.9.0
+   */
+  @Deprecated
+  public static final ConfigProperty<String> HIVE_USE_JDBC = ConfigProperty
+      .key("hoodie.datasource.hive_sync.use_jdbc")
+      .defaultValue("true")
+      .deprecatedAfter("0.9.0")
+      .withDocumentation("Use JDBC when hive synchronization is enabled");
   public static final ConfigProperty<String> METASTORE_URIS = ConfigProperty
       .key("hoodie.datasource.hive_sync.metastore.uris")
       .defaultValue("thrift://localhost:9083")
@@ -100,7 +109,7 @@ public class HiveSyncConfigHolder {
       .withDocumentation("The number of partitions one batch when synchronous partitions to hive.");
   public static final ConfigProperty<String> HIVE_SYNC_MODE = ConfigProperty
       .key("hoodie.datasource.hive_sync.mode")
-      .defaultValue("jdbc")
+      .noDefaultValue()
       .withDocumentation("Mode to choose for Hive ops. Valid values are hms, jdbc and hiveql.");
   public static final ConfigProperty<Boolean> HIVE_SYNC_BUCKET_SYNC = ConfigProperty
       .key("hoodie.datasource.hive_sync.bucket_sync")

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/HoodieHiveSyncClient.java

@@ -21,6 +21,7 @@ package org.apache.hudi.hive;
 import org.apache.hudi.common.table.TableSchemaResolver;
 import org.apache.hudi.common.table.timeline.HoodieInstant;
 import org.apache.hudi.common.util.Option;
+import org.apache.hudi.common.util.StringUtils;
 import org.apache.hudi.common.util.collection.Pair;
 import org.apache.hudi.hive.ddl.DDLExecutor;
 import org.apache.hudi.hive.ddl.HMSDDLExecutor;
@@ -48,6 +49,7 @@ import java.util.stream.Collectors;
 
 import static org.apache.hudi.hadoop.utils.HoodieHiveUtils.GLOBALLY_CONSISTENT_READ_TIMESTAMP;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_SYNC_MODE;
+import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USE_JDBC;
 import static org.apache.hudi.sync.common.HoodieSyncConfig.META_SYNC_DATABASE_NAME;
 import static org.apache.hudi.sync.common.util.TableUtils.tableId;
 
@@ -70,19 +72,23 @@ public class HoodieHiveSyncClient extends HoodieSyncClient {
     // Support JDBC, HiveQL and metastore based implementations for backwards compatibility. Future users should
     // disable jdbc and depend on metastore client for all hive registrations
     try {
-      HiveSyncMode syncMode = HiveSyncMode.of(config.getStringOrDefault(HIVE_SYNC_MODE));
-      switch (syncMode) {
-        case HMS:
-          ddlExecutor = new HMSDDLExecutor(config);
-          break;
-        case HIVEQL:
-          ddlExecutor = new HiveQueryDDLExecutor(config);
-          break;
-        case JDBC:
-          ddlExecutor = new JDBCExecutor(config);
-          break;
-        default:
-          throw new HoodieHiveSyncException("Invalid sync mode given " + config.getString(HIVE_SYNC_MODE));
+      if (!StringUtils.isNullOrEmpty(config.getString(HIVE_SYNC_MODE))) {
+        HiveSyncMode syncMode = HiveSyncMode.of(config.getString(HIVE_SYNC_MODE));
+        switch (syncMode) {
+          case HMS:
+            ddlExecutor = new HMSDDLExecutor(config);
+            break;
+          case HIVEQL:
+            ddlExecutor = new HiveQueryDDLExecutor(config);
+            break;
+          case JDBC:
+            ddlExecutor = new JDBCExecutor(config);
+            break;
+          default:
+            throw new HoodieHiveSyncException("Invalid sync mode given " + config.getString(HIVE_SYNC_MODE));
+        }
+      } else {
+        ddlExecutor = config.getBoolean(HIVE_USE_JDBC) ? new JDBCExecutor(config) : new HiveQueryDDLExecutor(config);
       }
       this.client = Hive.get(config.getHiveConf()).getMSC();
     } catch (Exception e) {

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/ddl/HiveQueryDDLExecutor.java

@@ -47,7 +47,7 @@ import java.util.stream.Collectors;
 import static org.apache.hudi.sync.common.util.TableUtils.tableId;
 
 /**
- * This class offers DDL executor backed by the HiveQL Driver.
+ * This class offers DDL executor backed by the hive.ql Driver This class preserves the old useJDBC = false way of doing things.
  */
 public class HiveQueryDDLExecutor extends QueryBasedDDLExecutor {
 

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/ddl/JDBCExecutor.java

@@ -43,7 +43,7 @@ import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_USER;
 import static org.apache.hudi.hive.util.HiveSchemaUtil.HIVE_ESCAPE_CHARACTER;
 
 /**
- * This class offers DDL executor backed by the jdbc.
+ * This class offers DDL executor backed by the jdbc This class preserves the old useJDBC = true way of doing things.
  */
 public class JDBCExecutor extends QueryBasedDDLExecutor {
 

hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/replication/HiveSyncGlobalCommitParams.java

@@ -32,7 +32,6 @@ import java.io.IOException;
 import java.io.InputStream;
 import java.util.Properties;
 
-import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_SYNC_MODE;
 import static org.apache.hudi.hive.HiveSyncConfigHolder.HIVE_URL;
 import static org.apache.hudi.sync.common.HoodieSyncConfig.META_SYNC_BASE_PATH;
 
@@ -93,7 +92,6 @@ public class HiveSyncGlobalCommitParams {
     String jdbcUrl = forRemote ? loadedProps.getProperty(REMOTE_HIVE_SERVER_JDBC_URLS)
             : loadedProps.getProperty(LOCAL_HIVE_SERVER_JDBC_URLS, loadedProps.getProperty(HIVE_URL.key()));
     props.setPropertyIfNonNull(HIVE_URL.key(), jdbcUrl);
-    props.setProperty(HIVE_SYNC_MODE.key(), "jdbc");
     LOG.info("building hivesync config forRemote: " + forRemote + " " + jdbcUrl + " "
         + basePath);
     return props;

hudi-sync/hudi-hive-sync/src/test/java/org/apache/hudi/hive/TestHiveSyncTool.java

@@ -120,7 +120,7 @@ public class TestHiveSyncTool {
     return opts;
   }
 
-  // (useSchemaFromCommitMetadata, syncAsDataSource, syncMode)
+  // (useJdbc, useSchemaFromCommitMetadata, syncAsDataSource)
   private static Iterable<Object[]> syncDataSourceTableParams() {
     List<Object[]> opts = new ArrayList<>();
     for (Object mode : SYNC_MODES) {

hudi-utilities/src/main/java/org/apache/hudi/utilities/HoodieDropPartitionsTool.java

@@ -174,6 +174,8 @@ public class HoodieDropPartitionsTool implements Serializable {
     public String hiveURL = "jdbc:hive2://localhost:10000";
     @Parameter(names = {"--hive-partition-field"}, description = "Comma separated list of field in the hive table to use for determining hive partition columns.", required = false)
     public String hivePartitionsField = "";
+    @Parameter(names = {"--hive-sync-use-jdbc"}, description = "Use JDBC when hive synchronization.", required = false)
+    public boolean hiveUseJdbc = true;
     @Parameter(names = {"--hive-metastore-uris"}, description = "hive meta store uris to use.", required = false)
     public String hiveHMSUris = null;
     @Parameter(names = {"--hive-sync-mode"}, description = "Mode to choose for Hive ops. Valid values are hms, jdbc and hiveql.", required = false)
@@ -213,7 +215,7 @@ public class HoodieDropPartitionsTool implements Serializable {
           + "   --hive-pass-word " + "Masked" + ", \n"
           + "   --hive-jdbc-url " + hiveURL + ", \n"
           + "   --hive-partition-field " + hivePartitionsField + ", \n"
-          + "   --hive-sync-mode " + hiveSyncMode + ", \n"
+          + "   --hive-sync-use-jdbc " + hiveUseJdbc + ", \n"
           + "   --hive-metastore-uris " + hiveHMSUris + ", \n"
           + "   --hive-sync-ignore-exception " + hiveSyncIgnoreException + ", \n"
           + "   --hive-partition-value-extractor-class " + partitionValueExtractorClass + ", \n"
@@ -245,7 +247,7 @@ public class HoodieDropPartitionsTool implements Serializable {
           && Objects.equals(hivePassWord, config.hivePassWord)
           && Objects.equals(hiveURL, config.hiveURL)
           && Objects.equals(hivePartitionsField, config.hivePartitionsField)
-          && Objects.equals(hiveSyncMode, config.hiveSyncMode)
+          && Objects.equals(hiveUseJdbc, config.hiveUseJdbc)
           && Objects.equals(hiveHMSUris, config.hiveHMSUris)
           && Objects.equals(partitionValueExtractorClass, config.partitionValueExtractorClass)
           && Objects.equals(sparkMaster, config.sparkMaster)
@@ -259,7 +261,7 @@ public class HoodieDropPartitionsTool implements Serializable {
     public int hashCode() {
       return Objects.hash(basePath, runningMode, tableName, partitions, instantTime,
           syncToHive, hiveDataBase, hiveTableName, hiveUserName, hivePassWord, hiveURL,
-          hivePartitionsField, hiveSyncMode, hiveHMSUris, partitionValueExtractorClass,
+          hivePartitionsField, hiveUseJdbc, hiveHMSUris, partitionValueExtractorClass,
           sparkMaster, sparkMemory, propsFilePath, configs, hiveSyncIgnoreException, help);
     }
   }
@@ -348,6 +350,7 @@ public class HoodieDropPartitionsTool implements Serializable {
     props.put(DataSourceWriteOptions.HIVE_PASS().key(), cfg.hivePassWord);
     props.put(DataSourceWriteOptions.HIVE_URL().key(), cfg.hiveURL);
     props.put(DataSourceWriteOptions.HIVE_PARTITION_FIELDS().key(), cfg.hivePartitionsField);
+    props.put(DataSourceWriteOptions.HIVE_USE_JDBC().key(), cfg.hiveUseJdbc);
     props.put(DataSourceWriteOptions.HIVE_SYNC_MODE().key(), cfg.hiveSyncMode);
     props.put(DataSourceWriteOptions.HIVE_IGNORE_EXCEPTIONS().key(), cfg.hiveSyncIgnoreException);
     props.put(DataSourceWriteOptions.HIVE_PASS().key(), cfg.hivePassWord);