Right now, the Hive connector mixes partition keys (like year or country in the folder path) with the actual data columns inside the files. It’s annoying because you can't tell which columns are for filtering/partitioning and which ones are just regular data. This makes it hard to write efficient queries or know how the table is actually organized.
Describe the solution you'd like
Make the connector "smart" enough to label partition keys. When I look at the table schema, it should clearly show which columns are partitions. Ideally, it should prevent users from accidentally trying to update a partition key like a normal piece of data.
Right now, the Hive connector mixes partition keys (like year or country in the folder path) with the actual data columns inside the files. It’s annoying because you can't tell which columns are for filtering/partitioning and which ones are just regular data. This makes it hard to write efficient queries or know how the table is actually organized.
Describe the solution you'd like
Make the connector "smart" enough to label partition keys. When I look at the table schema, it should clearly show which columns are partitions. Ideally, it should prevent users from accidentally trying to update a partition key like a normal piece of data.