SharePoint

SharePoint Online is Microsoft's document management platform.

Bruin supports SharePoint as a source for Ingestr assets. You can ingest Excel and CSV files from a SharePoint Online document library into a destination supported by ingestr.

Configuration

Step 1: Add a connection to .bruin.yml

connections:
  sharepoint:
    - name: my-sharepoint
      tenant_id: your-tenant-id
      client_id: your-client-id
      client_secret: your-client-secret
      hostname: example.sharepoint.com
      site: sites/Example
      library: Documents # optional
      max_file_size: 104857600 # optional, bytes
      max_files: 10000 # optional, use 0 for unlimited

• tenant_id: Azure AD tenant ID for the app registration.
• client_id: app registration client ID.
• client_secret: app registration client secret.
• hostname: SharePoint tenant hostname, such as example.sharepoint.com.
• site: server-relative site path, such as sites/Example.
• library: optional document library name. If omitted, ingestr uses the site's default Documents library.
• max_file_size: optional maximum bytes for a single downloaded file. Use 0 for unlimited.
• max_files: optional maximum number of files a glob may match. Defaults to 10000; use 0 for unlimited.

Authentication uses the OAuth2 client-credentials flow. The app registration needs application permission to read the site's files, such as Sites.Read.All or Files.Read.All, granted with admin consent.

Step 2: Create an ingestr asset

name: raw.sharepoint_budget
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/budget.xlsx#sheet=North,date_cols=Date"
  destination: duckdb

• source_connection: the name of the SharePoint connection defined in .bruin.yml.
• source_table: the file path in the document library, optionally followed by ingestr hints such as sheet, sheets, skip, drop_empty, date_cols, xlsx, or csv.

Example assets

These examples use DuckDB as the destination. Replace connection and destination with the destination connection you want to load into.

Single Excel sheet:

name: raw.sharepoint_products
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/products.xlsx#xlsx,sheet=Sheet1"
  destination: duckdb

Excel sheet with skipped rows and date conversion:

name: raw.sharepoint_forecast
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/parameters.xlsx#sheet=Forecast,skip=4,date_cols=Date|Month"
  destination: duckdb

Multiple sheets from one workbook:

name: raw.sharepoint_regional_data
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/regional data.xlsx#sheets=North|South|East|West"
  destination: duckdb

Multiple sheets from every workbook matched by a glob:

name: raw.sharepoint_monthly
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/monthly/*.xlsx#sheets=Jan|Feb|Mar"
  destination: duckdb

CSV with explicit encoding and tab separator:

name: raw.sharepoint_export
type: ingestr
connection: duckdb

parameters:
  source_connection: my-sharepoint
  source_table: "Reports/export.csv#csv,encoding=utf-16le,sep=tab"
  destination: duckdb

Available source_table formats

SharePoint does not expose fixed table names. The source_table value is a file path, or a glob of file paths, relative to the configured document library root.

Format	Example	Description
<path/to/file.xlsx>	Reports/products.xlsx	Reads the first sheet from an Excel workbook.
<path/to/file.xlsx>#sheet=<sheet_name>	Reports/products.xlsx#sheet=Sheet1	Reads a single named Excel sheet.
<path/to/file.xlsx>#sheets=<a>|<b>	Reports/budget.xlsx#sheets=North|South	Reads and stacks multiple Excel sheets, unioning columns by name.
<path/to/files/*.xlsx>#sheets=<a>|<b>	Reports/monthly/*.xlsx#sheets=Jan|Feb|Mar	Reads every matching workbook and stacks the requested sheets.
<path/to/file.csv>	Exports/customers.csv	Reads a CSV file using default CSV parsing.
<path/to/file.csv>#csv,encoding=<enc>,sep=<sep>	Exports/customers.csv#csv,encoding=utf-16le,sep=tab	Reads a CSV file with explicit format, encoding, or separator hints.
<path/to/files/**>	Reports/**	Reads all matching files recursively, detecting .xlsx or .csv per file.

Paths can include spaces, &, and glob wildcards: *, **, ?, [...], and {a,b}. If a file path contains a literal #, encode it as %23.

Source table hints

Hints are appended after # and separated with commas. A bare token is a format or behavior flag; key=value sets a named option.

Hint	Applies to	Description
xlsx or csv	all	Overrides the file format when it cannot be inferred from the extension.
sheet=<name>	Excel	Reads a single named sheet.
sheets=<a>|<b>	Excel	Reads and stacks multiple sheets separated by |.
skip=<n>	all	Skips n rows before reading the header row.
drop_empty	all	Skips rows where every data column is empty.
date_cols=<a>|<b>	Excel	Converts the named Excel serial date columns to ISO date strings.
raw	Excel	Reads raw values. This is the default.
formatted	Excel	Reads displayed/formatted cell text instead of raw values.
encoding=<enc>	CSV	Sets input encoding, such as utf-16le.
sep=<sep>	CSV	Sets field separator. Use tab or \t for tab-delimited files.

Every row includes metadata columns:

Column	Description
_source_file	The document-library path of the file the row came from.
_sheet_name	The Excel sheet name. This is null for CSV files.
_row_idx	Zero-based row position in the source sheet/file after skip and the header row.

SharePoint extraction is not incremental. The source defaults to replace; append, merge, and delete+insert can be selected, but each run still reads the full file or glob.

Step 3: Run the asset

bruin run assets/sharepoint_ingestion.yml