Hi Impala Devs, I would like to share the design document 'Design Proposal: Pluggable Query-Shape Reporter for Iceberg Tables' for discussion: JIRA: IMPALA-15447 <https://docs.google.com/document/d/1esQWpVWJs13tb5VgvHGU-KSn9tX73rhm8av9K0sPzwI/edit?tab=t.0>
*Motivation* We need to know query patterns that depend heavily on actual query workloads. Analyzing this pattern requires determining which columns are repeatedly used as equi-join keys and sort keys, as these exist only within the query plan inside the engine. The proposed SPI addresses this by capturing structural query shapes directly after planning, which can help services drive educated optimizations downstream. Please feel free to review the document and share your feedback. Regards, Prabhat Mishra
