java.lang.UnsupportedOperationException：在 bean 类中不能有循环引用，但获得了类 java.time.ZoneOffset 的循环引用

Question

While executing the below command

StructType obs = Encoders.bean(Test.class).schema();

I am getting the below error

java.lang.UnsupportedOperationException: Cannot have circular references in bean class, but got the circular reference of class class java.time.ZoneOffset

class Test has OffsetDateTime field, which is causing the error. If I remove this field the error also goes away. Is there a way by which I can avoid this circular reference error and also use OffsetDateTime?

java.lang.UnsupportedOperationException: Cannot have circular references in bean class, but got the circular reference of class class java.time.ZoneOffset
    at org.apache.spark.sql.errors.QueryExecutionErrors$.cannotHaveCircularReferencesInBeanClassError(QueryExecutionErrors.scala:984)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:148)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.$anonfun$inferDataType$1(JavaTypeInference.scala:156)
    at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:286)
    at scala.collection.IndexedSeqOptimized.foreach(IndexedSeqOptimized.scala:36)
    at scala.collection.IndexedSeqOptimized.foreach$(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:198)
    at scala.collection.TraversableLike.map(TraversableLike.scala:286)
    at scala.collection.TraversableLike.map$(TraversableLike.scala:279)
    at scala.collection.mutable.ArrayOps$ofRef.map(ArrayOps.scala:198)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:154)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:134)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.$anonfun$inferDataType$1(JavaTypeInference.scala:156)
    at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:286)
    at scala.collection.IndexedSeqOptimized.foreach(IndexedSeqOptimized.scala:36)
    at scala.collection.IndexedSeqOptimized.foreach$(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:198)
    at scala.collection.TraversableLike.map(TraversableLike.scala:286)
    at scala.collection.TraversableLike.map$(TraversableLike.scala:279)
    at scala.collection.mutable.ArrayOps$ofRef.map(ArrayOps.scala:198)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:154)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.$anonfun$inferDataType$1(JavaTypeInference.scala:156)
    at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:286)
    at scala.collection.IndexedSeqOptimized.foreach(IndexedSeqOptimized.scala:36)
    at scala.collection.IndexedSeqOptimized.foreach$(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:198)
    at scala.collection.TraversableLike.map(TraversableLike.scala:286)
    at scala.collection.TraversableLike.map$(TraversableLike.scala:279)
    at scala.collection.mutable.ArrayOps$ofRef.map(ArrayOps.scala:198)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:154)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.$anonfun$inferDataType$1(JavaTypeInference.scala:156)
    at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:286)
    at scala.collection.IndexedSeqOptimized.foreach(IndexedSeqOptimized.scala:36)
    at scala.collection.IndexedSeqOptimized.foreach$(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:198)
    at scala.collection.TraversableLike.map(TraversableLike.scala:286)
    at scala.collection.TraversableLike.map$(TraversableLike.scala:279)
    at scala.collection.mutable.ArrayOps$ofRef.map(ArrayOps.scala:198)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:154)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.$anonfun$inferDataType$1(JavaTypeInference.scala:156)
    at scala.collection.TraversableLike.$anonfun$map$1(TraversableLike.scala:286)
    at scala.collection.IndexedSeqOptimized.foreach(IndexedSeqOptimized.scala:36)
    at scala.collection.IndexedSeqOptimized.foreach$(IndexedSeqOptimized.scala:33)
    at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:198)
    at scala.collection.TraversableLike.map(TraversableLike.scala:286)
    at scala.collection.TraversableLike.map$(TraversableLike.scala:279)
    at scala.collection.mutable.ArrayOps$ofRef.map(ArrayOps.scala:198)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:154)
    at org.apache.spark.sql.catalyst.JavaTypeInference$.inferDataType(JavaTypeInference.scala:70)
    at org.apache.spark.sql.catalyst.encoders.ExpressionEncoder$.javaBean(ExpressionEncoder.scala:66)
    at org.apache.spark.sql.Encoders$.bean(Encoders.scala:171)
    at org.apache.spark.sql.Encoders.bean(Encoders.scala)```

Answer 1

OffsetDateTime not supported by Encoders

As per Encoder doc , supported type for bean fields are:

supported types for java bean field: - primitive types: boolean, int, double, etc. - boxed types: Boolean, Integer, Double, etc. - String - java.math.BigDecimal, java.math.BigInteger - time related: java.sql.Date, java.sql.Timestamp, java.time.LocalDate, java.time.Instant - collection types: array, java.util.List, and map - nested java bean.

So it looks like it may not be possible to use OffsetDateTime as far as I can see. ZoneOffset is part of OffsetDateTime field so call Encoders.bean might be generating code which keeps referring ZoneOffset in circular manner.

Use Instant

You can convert a OffsetDateTime object to the more basic class, Instant .

Instant instant = odt.toInstant() ;

And back again.

ZoneOffset offset = ZoneOffset.ofHours( -8 ) ;
OffsetDateTime odt = instant.atOffset( offset ) ;