Hi Dan
I think using Flink SQL should be able to meet your needs.
You can write a Flink Jar program. Accept different directories, schemas,
mappings, and sink tables to generate DDL and DML.
Assuming you have two directories:
directory1 -> f1, f2, f3, f4 -> iceberg1
directory2 -> f1, f2, f3 ->
Thanks Feng,
I think my challenge (and why I expected I’d need to use Java) is that there
will be parquet files with different schemas landing in the s3 bucket - so I
don’t want to hard-code the schema in a sql table definition.
I’m not sure if this is even possible? Maybe I would have to write
