Fix flattening Avro Maps with Utf8 keys

[chrishardis]

Fix issue #5884 by @varaga @jon-wei on master branch.
Successfully tested with Confluent 5.1.0 kafka and schema registry

[clintropolis]

Thanks for the contribution! 👍

It looks like your PR has some style issues and is failing CI

[ERROR] /home/travis/build/apache/incubator-druid/extensions-core/avro-extensions/src/main/java/org/apache/druid/data/input/avro/GenericAvroJsonProvider.java:142:77: '{' at column 77 should be on a new line. [LeftCurly]
[ERROR] /home/travis/build/apache/incubator-druid/extensions-core/avro-extensions/src/main/java/org/apache/druid/data/input/avro/GenericAvroJsonProvider.java:158:50: ',' is not followed by whitespace. [WhitespaceAfter]

Could you also please add a test case for this issue? It looks lgtm otherwise.

[clintropolis]

It would also appear that you have opened a duplicate PR, #7254.

[clintropolis]

Will this work if the map is empty?

It might be more straightforward to just do something like this in getMapValue

...
} else if (o instanceof Map) {
  final Map theMap = (Map) o;
  if (theMap.containsKey(s)) {
    return theMap.get(s);
  } else {
    final Utf8 utf8Key = new Utf8(s);
    return theMap.get(utf8Key);
  }
}
...

[chrishardis]

your suggestion is more elegant, I will commit it. Thks

[varaga]

Thanks @chrishardis for taking over the bug and committing it to master.
Would you be able to add some tests to cover the bug if it's feasible.

[chrishardis]

I spend my morning trying to run a test case, but it's not so easy has using the io.confluent.kafka.serializers.KafkaAvroSerializer without a kafkaproducer is quite an hard job. I test it many times on real Confluent 5.1 Kafka Cluster with Schema Registry.

[clintropolis]

I spend my morning trying to run a test case, but it's not so easy has using the io.confluent.kafka.serializers.KafkaAvroSerializer without a kafkaproducer is quite an hard job. I test it many times on real Confluent 5.1 Kafka Cluster with Schema Registry.

I think this issue would exist for any type of Avro indexing; like if you could create an Avro formatted file with a column that is a map type with utf8 keys, you could test with the Hadoop indexing tests if that would be easier.

That said, this change seems small and harmless enough that I think it would probably be ok without tests, so I will go ahead and approve.