README

Overview

SAP HANA is an in-memory, column-oriented relational database that powers SAP Business Suite, SAP S/4HANA, and bespoke analytics built on the HANA Cloud / HANA on-premise platform. Learn more in the official SAP HANA documentation.

The DataHub integration for SAP HANA covers core metadata entities such as datasets (tables, regular views, and calculation views), schema fields, and containers. It also captures table- and column-level lineage from calculation views, table-level and procedure-to-procedure lineage from stored procedures, data profiling, and stateful deletion detection.

Concept Mapping

SAP HANA Concept	DataHub Entity (Subtype)	Notes
Tenant database	Platform Instance	Top-level scope; all URNs include the configured platform instance.
Schema	Container (SCHEMA)	Top-level namespace inside the tenant database.
Table (row / column / virtual)	Dataset (TABLE)	Includes column-store, row-store, and virtual tables. Schema, descriptions, and tags (where available) are extracted.
Regular View	Dataset (VIEW)	Standard SQL views; the view definition is captured for lineage.
Calculation View	Dataset (Calculation View)	Opt-in via include_calculation_views: true. Activated calc views are pulled from _SYS_REPO.ACTIVE_OBJECT; column-level lineage is parsed from the XML.
SqlScriptView Calculation View	Dataset (Calculation View)	Calc views whose body is HANA SQLScript. Table-level lineage is extracted from the procedure body; column-level lineage from SQLScript is out of scope.
Stored Procedure (SQLScript)	DataFlow (Procedures Container) + DataJob (Stored Procedure)	Opt-in via include_stored_procedures (on by default). One DataFlow per schema; each procedure becomes a DataJob. Table-level lineage is parsed from the body.
Column / field	SchemaField	Native type, nullability, and (for calc views) comments are extracted.
Table- and column-level lineage	Lineage edges	Extracted from view definitions and calc-view XML; column-level lineage flows through the SqlParsingAggregator like Snowflake / Redshift.
Query history / usage	DatasetUsageStatistics + Operations + query-derived lineage	Opt-in via include_query_usage. Mines _SYS_STATISTICS.HOST_SQL_PLAN_CACHE (requires the statistics service and the MONITORING role).