【发布时间】：2022-08-12 18:29:10
【问题描述】：
我正在使用一个新的Flink/Databricks connector，我试图了解单个 Flink 接收器是否可以创建多个增量表？
查看文档，对 path/s3/hdfs 的引用是单个位置
/**
* Convenience method for creating a {@link RowDataDeltaSinkBuilder} for {@link DeltaSink} to a
* Delta table.
*
* @param basePath root path of the Delta table
* @param conf Hadoop\'s conf object that will be used for creating instances of
* {@link io.delta.standalone.DeltaLog} and will be also passed to the
* {@link ParquetRowDataBuilder} to create {@link ParquetWriterFactory}
* @param rowType Flink\'s logical type to indicate the structure of the events in the stream
* @return builder for the DeltaSink
*/
public static RowDataDeltaSinkBuilder forRowData(
final Path basePath,
final Configuration conf,
final RowType rowType
) {
return new RowDataDeltaSinkBuilder(
basePath,
conf,
rowType,
false // mergeSchema
);
是否可以根据事件流动态更改 basePath？
-
我看到你在这里打开了一个类似的线程github.com/delta-io/connectors/issues/344你能根据斯科特的问题提供更多细节吗？
标签： apache-flink databricks delta-lake databricks-connect