Allow Avro type with no fields

[Fryuni]

We have one record field that can be of different types depending of the what the message is about. All of they are mapped correctly to BigQuery except this one:

{
          "type": "record",
          "name": "NothingChanged",
          "fields": []
}

which results in this error:

java.lang.IllegalArgumentException: The RECORD field must have at least one sub-field
	at com.google.cloud.bigquery.Field$Builder.setType(Field.java:149)
	at com.google.cloud.bigquery.Field.newBuilder(Field.java:285)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertStruct(BigQuerySchemaConverter.java:175)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertField(BigQuerySchemaConverter.java:127)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertStruct(BigQuerySchemaConverter.java:169)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertField(BigQuerySchemaConverter.java:127)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertSchema(BigQuerySchemaConverter.java:109)
	at com.wepay.kafka.connect.bigquery.convert.BigQuerySchemaConverter.convertSchema(BigQuerySchemaConverter.java:43)
	at com.wepay.kafka.connect.bigquery.SchemaManager.constructTableInfo(SchemaManager.java:68)
	at com.wepay.kafka.connect.bigquery.SchemaManager.createTable(SchemaManager.java:49)
	at com.wepay.kafka.connect.bigquery.BigQuerySinkConnector.ensureExistingTables(BigQuerySinkConnector.java:117)
	at com.wepay.kafka.connect.bigquery.BigQuerySinkConnector.ensureExistingTables(BigQuerySinkConnector.java:140)
	at com.wepay.kafka.connect.bigquery.BigQuerySinkConnector.start(BigQuerySinkConnector.java:159)
	at org.apache.kafka.connect.runtime.WorkerConnector.doStart(WorkerConnector.java:110)
	at org.apache.kafka.connect.runtime.WorkerConnector.start(WorkerConnector.java:135)
	at org.apache.kafka.connect.runtime.WorkerConnector.transitionTo(WorkerConnector.java:195)
	at org.apache.kafka.connect.runtime.Worker.startConnector(Worker.java:257)
	at org.apache.kafka.connect.runtime.distributed.DistributedHerder.startConnector(DistributedHerder.java:1183)
	at org.apache.kafka.connect.runtime.distributed.DistributedHerder.access$1400(DistributedHerder.java:125)
	at org.apache.kafka.connect.runtime.distributed.DistributedHerder$15.call(DistributedHerder.java:1199)
	at org.apache.kafka.connect.runtime.distributed.DistributedHerder$15.call(DistributedHerder.java:1195)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)

Could record types with 0 fields simply be ignored while generating the schema for BigQuery?

[anderseriksson]

We have a similar situation with a struct without fields. It seems version 1.6.5 (with some extra commits) of the connector is actually ignoring this field in BigQuery...