【发布时间】:2019-08-10 09:03:20
【问题描述】:
我有一个工具,它使用org.apache.parquet.hadoop.ParquetWriter 将 CSV 数据文件转换为 parquet 数据文件。
我可以很好地编写基本的原始类型(INT32、DOUBLE、BINARY 字符串)。
我需要写 NULL 值,但我不知道怎么写。我试过用 ParquetWriter 简单地写null,它会引发异常。
如何使用org.apache.parquet.hadoop.ParquetWriter 写入NULL?有可以为空的类型吗?
我认为代码是不言自明的:
ArrayList<Type> fields = new ArrayList<>();
fields.add(new PrimitiveType(Type.Repetition.OPTIONAL, PrimitiveTypeName.INT32, "int32_col", null));
fields.add(new PrimitiveType(Type.Repetition.OPTIONAL, PrimitiveTypeName.DOUBLE, "double_col", null));
fields.add(new PrimitiveType(Type.Repetition.OPTIONAL, PrimitiveTypeName.BINARY, "string_col", null));
MessageType schema = new MessageType("input", fields);
Configuration configuration = new Configuration();
configuration.setQuietMode(true);
GroupWriteSupport.setSchema(schema, configuration);
SimpleGroupFactory f = new SimpleGroupFactory(schema);
ParquetWriter<Group> writer = new ParquetWriter<Group>(
new Path("output.parquet"),
new GroupWriteSupport(),
CompressionCodecName.SNAPPY,
ParquetWriter.DEFAULT_BLOCK_SIZE,
ParquetWriter.DEFAULT_PAGE_SIZE,
1048576,
true,
false,
ParquetProperties.WriterVersion.PARQUET_1_0,
configuration
);
// create row 1 with defined values
Group group1 = f.newGroup();
Integer int1 = 100;
Double double1 = 0.5;
String string1 = "string-value";
group1.add(0, int1);
group1.add(1, double1);
group1.add(2, string1);
writer.write(group1);
// create row 2 with NULL values -- does not work!
Group group2 = f.newGroup();
Integer int2 = null;
Double double2 = null;
String string2 = null;
group2.add(0, int2); // <-- throws NullPointerException
group2.add(1, double2); // <-- throws NullPointerException
group2.add(2, string2); // <-- throws NullPointerException
writer.write(group2);
writer.close();
【问题讨论】:
标签: java apache-spark hadoop parquet