37112fcfcd
## What changes were proposed in this pull request? This PR proposes to fix two. **Skip a property without a getter in beans** Currently, if we use a JavaBean without the getter as below: ```java public static class BeanWithoutGetter implements Serializable { private String a; public void setA(String a) { this.a = a; } } BeanWithoutGetter bean = new BeanWithoutGetter(); List<BeanWithoutGetter> data = Arrays.asList(bean); spark.createDataFrame(data, BeanWithoutGetter.class).show(); ``` - Before It throws an exception as below: ``` java.lang.NullPointerException at org.spark_project.guava.reflect.TypeToken.method(TypeToken.java:465) at org.apache.spark.sql.catalyst.JavaTypeInference$$anonfun$2.apply(JavaTypeInference.scala:126) at org.apache.spark.sql.catalyst.JavaTypeInference$$anonfun$2.apply(JavaTypeInference.scala:125) ``` - After ``` ++ || ++ || ++ ``` **Supports empty bean in encoder creation** ```java public static class EmptyBean implements Serializable {} EmptyBean bean = new EmptyBean(); List<EmptyBean> data = Arrays.asList(bean); spark.createDataset(data, Encoders.bean(EmptyBean.class)).show(); ``` - Before throws an exception as below: ``` java.lang.UnsupportedOperationException: Cannot infer type for class EmptyBean because it is not bean-compliant at org.apache.spark.sql.catalyst.JavaTypeInference$.org$apache$spark$sql$catalyst$JavaTypeInference$$serializerFor(JavaTypeInference.scala:436) at org.apache.spark.sql.catalyst.JavaTypeInference$.serializerFor(JavaTypeInference.scala:341) ``` - After ``` ++ || ++ || ++ ``` ## How was this patch tested? Unit test in `JavaDataFrameSuite`. Author: hyukjinkwon <gurwls223@gmail.com> Closes #17013 from HyukjinKwon/SPARK-19666. |
||
---|---|---|
.. | ||
benchmarks | ||
src | ||
pom.xml |