search

databricks / iceberg-kafka-connect

Apache License 2.0
219 stars 49 forks source link

Ability to treat semi-structured fields as maps instead of structs #192

Open kainoa21 opened 9 months ago

kainoa21 commented 9 months ago

It is common to have map type data in a kafka stream with a large (potentially unbounded) set of keys. Currently, this can result in struct type columns with a large number of fields which is problematic.

okayhooni commented 9 months ago

I also hope that feature..!

related issue: https://github.com/tabular-io/iceberg-kafka-connect/issues/141