simonbence commented on a change in pull request #4223:
URL: https://github.com/apache/nifi/pull/4223#discussion_r422978152
##########
File path:
nifi-nar-bundles/nifi-extension-utils/nifi-record-utils/nifi-avro-record-utils/src/main/java/org/apache/nifi/avro/AvroTypeUtil.java
##########
@@ -256,6 +256,12 @@ private static Schema buildAvroSchema(final DataType
dataType, final String fiel
case LONG:
schema = Schema.create(Type.LONG);
break;
+ case BIGDECIMAL:
+ // One more byte than below to allow the dot in the string
representation
+ schema = Schema.createFixed(fieldName + "Type", null,
"org.apache.nifi",39);
Review comment:
Based what I saw within org.apache.avro.Schema, other than creating
schema of a composite type or one based on the Type enum within (which does not
support BigDecimal), creating a fixed one is supported by it. As we do use it
in
[convertToAvroObject](https://github.com/apache/nifi/blob/master/nifi-nar-bundles/nifi-extension-utils/nifi-record-utils/nifi-avro-record-utils/src/main/java/org/apache/nifi/avro/AvroTypeUtil.java#L691-L698)
as well it seems like a working approach.
As for the 38, the idea was to be consistent with ORC, but you are right, we
do not need to bound ourselves for the smallest common set but we might say in
case of Avro, we pick a bigger limit.
----------------------------------------------------------------
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]
