Schema Evolution & Management
Handle schema changes in source systems automatically with NATIS schema evolution policies.
On this page
Source system schemas change over time — new columns are added, data types change, and tables are restructured. NATIS Schema Evolution policies define how these changes are detected, propagated, and handled across your lakehouse.
Schema Evolution Policies
Policy | Behavior | Use Case — | — | — STRICT | Reject records that don't match the target schema | Production tables with strict contracts ADD_COLUMNS | Automatically add new columns detected in source | Evolving source schemas (default) EVOLVE | Add columns and widen compatible data types (e.g., INT→BIGINT) | Flexible analytics tables IGNORE_MISMATCH | Silently drop mismatched columns/values | Legacy sources with noisy schemas
Type Compatibility Rules
Use the Schema Diff tool (Data Explorer → Schema → Compare Versions) to visualize schema changes between pipeline runs before applying them to production tables.
- INT → BIGINT: automatically widened (safe)
- FLOAT → DOUBLE: automatically widened (safe)
- VARCHAR(100) → VARCHAR(200): widened if EVOLVE policy is set
- STRING → INT: rejected (incompatible; requires manual mapping)
- TIMESTAMP → DATE: rejected (loss of precision; requires explicit cast)
Was this page helpful?
Thanks for your feedback!