danny0405 commented on code in PR #9590:
URL: https://github.com/apache/hudi/pull/9590#discussion_r1323835958
##########
hudi-common/src/main/java/org/apache/hudi/keygen/constant/KeyGeneratorType.java:
##########
@@ -33,31 +37,92 @@
public enum KeyGeneratorType {
@EnumFieldDescription("Simple key generator, which takes names of fields to
be used for recordKey and partitionPath as configs.")
- SIMPLE,
+ SIMPLE_KEYGEN("org.apache.hudi.keygen.SimpleKeyGenerator"),
+ @EnumFieldDescription("Simple key generator, which takes names of fields to
be used for recordKey and partitionPath as configs.")
+ SIMPLE_AVRO_KEYGEN("org.apache.hudi.keygen.SimpleAvroKeyGenerator"),
@EnumFieldDescription("Complex key generator, which takes names of fields to
be used for recordKey and partitionPath as configs.")
- COMPLEX,
+ COMPLEX_KEYGEN("org.apache.hudi.keygen.ComplexKeyGenerator"),
+ @EnumFieldDescription("Complex key generator, which takes names of fields to
be used for recordKey and partitionPath as configs.")
+ COMPLEX_AVRO_KEYGEN("org.apache.hudi.keygen.ComplexAvroKeyGenerator"),
@EnumFieldDescription("Timestamp-based key generator, that relies on
timestamps for partitioning field. Still picks record key by name.")
- TIMESTAMP,
+ TIMESTAMP_KEYGEN("org.apache.hudi.keygen.TimestampBasedKeyGenerator"),
+ @EnumFieldDescription("Timestamp-based key generator, that relies on
timestamps for partitioning field. Still picks record key by name.")
+
TIMESTAMP_AVRO_KEYGEN("org.apache.hudi.keygen.TimestampBasedAvroKeyGenerator"),
@EnumFieldDescription("This is a generic implementation type of KeyGenerator
where users can configure record key as a single field or "
+ " a combination of fields. Similarly partition path can be configured
to have multiple fields or only one field. "
+ " This KeyGenerator expects value for prop
\"hoodie.datasource.write.partitionpath.field\" in a specific format. "
+ " For example: "
+ " properties.put(\"hoodie.datasource.write.partitionpath.field\",
\"field1:PartitionKeyType1,field2:PartitionKeyType2\").")
- CUSTOM,
+ CUSTOM_KEYGEN("org.apache.hudi.keygen.CustomKeyGenerator"),
+ @EnumFieldDescription("This is a generic implementation type of KeyGenerator
where users can configure record key as a single field or "
+ + " a combination of fields. Similarly partition path can be configured
to have multiple fields or only one field. "
+ + " This KeyGenerator expects value for prop
\"hoodie.datasource.write.partitionpath.field\" in a specific format. "
+ + " For example: "
+ + " properties.put(\"hoodie.datasource.write.partitionpath.field\",
\"field1:PartitionKeyType1,field2:PartitionKeyType2\").")
+ CUSTOM_AVRO_KEYGEN("org.apache.hudi.keygen.CustomAvroKeyGenerator"),
@EnumFieldDescription("Simple Key generator for non-partitioned tables.")
- NON_PARTITION,
+ NON_PARTITION_KEYGEN("org.apache.hudi.keygen.NonpartitionedKeyGenerator"),
+ @EnumFieldDescription("Simple Key generator for non-partitioned tables.")
+
NON_PARTITION_AVRO_KEYGEN("org.apache.hudi.keygen.NonpartitionedAvroKeyGenerator"),
@EnumFieldDescription("Key generator for deletes using global indices.")
- GLOBAL_DELETE;
+ GLOBAL_DELETE_KEYGEN("org.apache.hudi.keygen.GlobalDeleteKeyGenerator"),
+ @EnumFieldDescription("Key generator for deletes using global indices.")
+
GLOBAL_DELETE_AVRO_KEYGEN("org.apache.hudi.keygen.GlobalAvroDeleteKeyGenerator"),
+
+ @EnumFieldDescription("Automatic record key generation.")
+
AUTO_RECORD_KEYGEN("org.apache.hudi.keygen.AutoRecordGenWrapperKeyGenerator"),
+ @EnumFieldDescription("Automatic record key generation.")
+
AUTO_RECORD_AVRO_KEYGEN("org.apache.hudi.keygen.AutoRecordGenWrapperAvroKeyGenerator"),
+
+ @EnumFieldDescription("A KeyGenerator which use the uuid as the record key.")
+ UUID_KEYGEN("org.apache.spark.sql.hudi.command.UuidKeyGenerator"),
+
+ @EnumFieldDescription("Custom key generator for the Hudi table metadata.")
+
HOODIE_METADATA_KEYGEN("org.apache.hudi.metadata.HoodieTableMetadataKeyGenerator"),
+
+ @EnumFieldDescription("Meant to be used internally for the spark sql MERGE
INTO command.")
+ MERGE_INTO_KEYGEN("org.apache.spark.sql.hudi.command.MergeIntoKeyGenerator"),
+
+ @EnumFieldDescription("Custom spark-sql specific KeyGenerator overriding
behavior handling TimestampType partition values.")
+ SQL_KEYGEN("org.apache.spark.sql.hudi.command.SqlKeyGenerator");
+
+ private final String className;
+
+ KeyGeneratorType(String className) {
+ this.className = className;
+ }
+
+ public String getClassName() {
+ return className;
+ }
+
+ public static KeyGeneratorType fromClassName(String className) {
+ for (KeyGeneratorType type : KeyGeneratorType.values()) {
+ if (type.getClassName().equals(className)) {
+ return type;
+ }
+ }
+ throw new IllegalArgumentException("No KeyGeneratorType found for class
name: " + className);
+ }
public static List<String> getNames() {
List<String> names = new ArrayList<>(KeyGeneratorType.values().length);
Arrays.stream(KeyGeneratorType.values())
.forEach(x -> names.add(x.name()));
return names;
}
+
+ public static String getKeyGeneratorClassName(HoodieConfig config) {
Review Comment:
Add annotation `@Nullable` to the method if there is any chance it could
return null value, or return `Option` instead.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
