davidov541 commented on a change in pull request #933: HIVE-21218: Adding 
support for Confluent Kafka Avro message format
URL: https://github.com/apache/hive/pull/933#discussion_r387878744
 
 

 ##########
 File path: kafka-handler/src/java/org/apache/hadoop/hive/kafka/KafkaSerDe.java
 ##########
 @@ -133,12 +134,40 @@
       Preconditions.checkArgument(!schemaFromProperty.isEmpty(), "Avro Schema 
is empty Can not go further");
       Schema schema = AvroSerdeUtils.getSchemaFor(schemaFromProperty);
       LOG.debug("Building Avro Reader with schema {}", schemaFromProperty);
-      bytesConverter = new AvroBytesConverter(schema);
+      bytesConverter = getByteConverterForAvroDelegate(schema, tbl);
     } else {
       bytesConverter = new BytesWritableConverter();
     }
   }
 
+  enum BytesConverterType {
+    CONFLUENT,
+    SKIP,
+    NONE;
+
+    static BytesConverterType fromString(String value) {
+      try {
+        return BytesConverterType.valueOf(value.trim().toUpperCase());
+      } catch (Exception e){
+        return NONE;
+      }
+    }
+  }
+
+  BytesConverter getByteConverterForAvroDelegate(Schema schema, Properties 
tbl) {
+    String avroBytesConverterProperty = tbl.getProperty(AvroSerdeUtils
+                                                            
.AvroTableProperties.AVRO_SERDE_TYPE
+                                                            .getPropName(), 
BytesConverterType.NONE.toString());
+    BytesConverterType avroByteConverterType = 
BytesConverterType.fromString(avroBytesConverterProperty);
+    Integer avroSkipBytes = 
Integer.getInteger(tbl.getProperty(AvroSerdeUtils.AvroTableProperties.AVRO_SERDE_SKIP_BYTES
 
 Review comment:
   I think I'm confused what you're asking for then. The initialize function 
takes in a java.util.Properties object that has properties that have been set 
for the serde in the DDL for the table. It reads a few from that object, and 
then passes it to getByteConverterForAvroDelegate, where it is also used in the 
code added here. The usage of the properties object here matches what is being 
done in initialize, and seems to match what I would expect. These aren't 
pulling system properties of the JVM, or at least are not necessarily doing so, 
instead reading from the Properties object passed to us.
   
   Does that make sense, or am I way off base?

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to