Question

我在火花流应用程序中使用FlatMapGroupsWithStateFunction。

 FlatMapGroupsWithStateFunction<String, Row, SessionInfo, SessionUpdate> idstateUpdateFunction =
  new FlatMapGroupsWithStateFunction<String, Row, SessionInfo, SessionUpdate>() {.....}

会话更新类如下；

  public static class SessionUpdate implements Serializable {

private static final long serialVersionUID = -3858977319192658483L;

private String instanceId;

private ArrayList<GenericRowWithSchema> milestones = new ArrayList<GenericRowWithSchema>();

private Timestamp processingTimeoutTimestamp;

public SessionUpdate() { 
  super();
}

public SessionUpdate(String instanceId, ArrayList<GenericRowWithSchema> milestones, Timestamp processingTimeoutTimestamp) {
  super();
  this.instanceId = instanceId;
  this.milestones = milestones;
  this.processingTimeoutTimestamp = processingTimeoutTimestamp;
}

public String getInstanceId() {
  return instanceId;
}

public void setInstanceId(String instanceId) {
  this.instanceId = instanceId;
}

public ArrayList<GenericRowWithSchema> getMilestones() {
  return milestones;
}

public void setMilestones(ArrayList<GenericRowWithSchema> milestones) {
  this.milestones = milestones;
}

public Timestamp getProcessingTimeoutTimestamp() {
  return processingTimeoutTimestamp;
}

public void setProcessingTimeoutTimestamp(Timestamp processingTimeoutTimestamp) {
  this.processingTimeoutTimestamp = processingTimeoutTimestamp;
}

}

将以下内容添加到该类中后，我得到如下所述的异常；

private ArrayList<GenericRowWithSchema> milestones = new ArrayList<GenericRowWithSchema>();

例外：

ERROR  cannot resolve 'named_struct()' due to data type mismatch: input to function named_struct requires at least one argument;;

'SerializeFromObject [staticinvoke（类org.apache.spark.unsafe.types.UTF8String，StringType，fromString，assertnotnull（input [0，oracle.insight.spark.event_processor.EventProcessor $ SessionUpdate，true]）。getInstanceId，true ，false）AS instanceId＃62，mapobjects（MapObjects_loopValue2，MapObjects_loopIsNull2，ObjectType（类org.apache.spark.sql.catalyst.expressions.GenericRowWithSchema），如果（isnull（lambdavariable（MapObjects_loopValue2，MapObjects_loopIsNull2.ObjectType .sql.catalyst.expressions.GenericRowWithSchema），true）））null，否则named_struct（），assertnotnull（input [0，oracle.insight.spark.event_processor.EventProcessor $ SessionUpdate，true]）。getMilestones，None）AS里程碑＃63 ，staticinvoke（类org.apache.spark.sql.catalyst.util.DateTimeUtils $，TimestampType，来自JavaTimestamp，assertnotnull（input [0，oracle.insight.spark.event_processor.EventProcessor $ SessionUpdate，true]）。getProcessingTimeoutTimestamp，true，false ）AS processingTimeoutTimestamp＃64] +-FlatMapGroupsWithState，cast（值＃54作为字符串）.toString，createexternalrow（EventTime＃23.toString，InstanceID＃24.toString，Model＃25.toString，Milestone＃26.toString，Region＃27.toString，SalesOrganization＃28 .toString，ProductName＃29.toString，ReasonForQuoteReject＃30.toString，ReasonforRejectionBy＃31.toString，OpportunityAmount＃32.toJavaBigDecimal，Discount＃33.toJavaBigDecimal，TotalQuoteAmount＃34.toJavaBigDecimal，NetQuoteAmount＃35.toJavaBigDecimal，ApprovedDiscount＃36。，TotalOrderAmount＃37.toJavaBigDecimal，StructField（EventTime，StringType，true），StructField（InstanceID，StringType，true），StructField（Model，StringType，true），StructField（Milestone，StringType，true），StructField（Region，StringType，true）），StructField（SalesOrganization，StringType，true），StructField（ProductName，StringType，true），StructField（ReasonForQuoteReject，StringType，true），StructField（ReasonforRejectionBy，StringType，true），...还有6个字段），[value＃54 ]，[EventTime＃23，InstanceID＃24，Model＃25，里程碑＃26，地区＃27，SalesOrganization＃28，产品名称＃29，ReasonForQuoteReject＃30，ReasonforRejectionBy＃31，OpportunityAmount＃32，Discount＃33，TotalQuoteAmount＃34，NetQuoteAmount＃35，ApprovedDiscount＃36，TotalOrderAmount＃37]，obj＃ 61：oracle.insight.spark.event_processor.EventProcessor $ SessionUpdate，类[instanceId [0]：字符串，里程碑[0]：array>，processingTimeoutTimestamp [0]：timestamp]，Append，false，ProcessingTimeTimeout

任何提示我为什么会收到此异常？

谢谢。

由于数据类型不匹配'SerializeFromObject

0 个答案: