Import File into Table

Reads a CSV or Excel file into a named in-memory table on the runner. Only metadata (tableName, columns, totalRows) is returned to the workflow — rows stay on the runner and don't cross the Temporal payload boundary. Call DropTable when done to free runner memory. Preferred over ExtractDataExcel (deprecated) for non-trivial files, and over ExcelToTable + CreateTable + TablePushMany for round-trip workflows.

Category: Named Tables & Rules Engine · Group: Tables · Activity name: TableImport

Example

outputVariable = TableImport("<tableName>", "<filePath>", keyColumns={}, format="auto", sheetName="Sheet1", sheetIndex=0, rowToSkip=1, delimiter=",")

Inputs

Name	Type	Required	Default	Description
Table Name	string	Yes		Name for the imported table. If a table with this name already exists in the session, it is overwritten.
File Path	string	Yes		Path to the source CSV or .xlsx file.
Key Columns	json			Key column names as an array of strings, e.g. ["OrderId"]. Required if the table will be used for upserts; can be set later via SetTableKeyColumns.
Format	string		auto	Input format. 'auto' infers from the file extension. One of: auto, csv, xlsx.
Sheet Name	string		Sheet1	Excel sheet to read. Takes precedence over sheetIndex when both are set. Ignored for CSV.
Sheet Index	integer			Excel sheet index (0-based). Used only when sheetName is not provided. Ignored for CSV.
Header Row	integer		1	Row number (1-based) to treat as the header row. Data rows start immediately after.
CSV Delimiter	string		,	Field delimiter for CSV input. Ignored for Excel.

Outputs

Name	Type	Description
outputVariable	json	Stores import result (tableName, columns, totalRows).