New feature
GA
SharePoint connector
SharePoint connector is a Databricks data engineering capability, introduced June 2025.
The managed Lakeflow Connect file connector that ingests SharePoint files into Delta tables - unstructured files as binary, structured formats parsed into tables, or file metadata only - plus SharePoint list data.
- Unstructured files load into memory as one record each, so a file over 100 MB can sink the update; the docs suggest a row_filter on length to skip the big ones.
- It is all or nothing per folder: there is no picking individual files, only a folder, drive, subsite or site.
- It has a sibling - the standard SharePoint connector for hand-built Spark and SQL pipelines - and both reached GA in the same release note.
Limitations: No SCD type 2: unstructured ingestion is SCD type 1 only and structured ingestion is append-only, so updates and deletes to existing structured files are missed. No individual file selection, no file-level ACLs, no files linked from another document library, and files over 100 MB risk failing. List ingestion (Beta) is not tenant-wide, skips OneDrive and attachments, and has no UI authoring.
Open in REbricked →- Category
- Data engineering
- Also known as
- Microsoft SharePoint connector, Managed SharePoint connector, SharePoint
- Verified
- 2026-10-20
Sources
- Official Official Databricks / Microsoft docs
- Official Databricks release notes: Microsoft SharePoint connector (Beta) (June 11, 2025)
- Official Databricks release notes: Managed SharePoint connector adds structured file ingestion and file metadata support (June 2026)
- Official Databricks release notes: SharePoint connector (GA) (August 4, 2026)
- Official Databricks docs: Microsoft SharePoint connector limitations
- Official Databricks docs: Ingest files from SharePoint (standard connector)