Data Table

Read and write a project's Data Table as a workflow step — the register that lets a run remember.

The Data Table node (data_table) reads and writes one of your project's Data Tables: create and change tables, and insert, get, update, upsert and delete rows. It is documented in full in the Data Tables section — The Data Table node covers every operation, mapping, condition and flag, with examples checked against the shipped schema. This page is its entry in the node reference.

Configuration

Field	Type	Description
`resource`	`table` \| `row`	What the operation acts on. See Operations at a glance.
`operation`	string	The operation, for example `insert`, `get`, `update`, `upsert`, `bulkUpsert`, `delete`, `exists`, `create`, `list`, `rename`, `addColumn`, `removeColumn`.
`tableId`	string	The table, from this workflow's project. See Choosing the table.
`tableName`	string	The table's name — kept for display, and the name for Create table.
`columns`	list	Create table: the columns, each a name and a type.
`returnExistingIfExists`	boolean	Create table: return the table if one by that name exists, instead of failing.
`newTableName`	string	Rename table: the new name.
`columnName`	string	Add / Remove column: the column.
`columnType`	string	Add column: its type.
`mapping`	`manual` \| `automatic`	How a write's cells are set. See Writing cells.
`mappingRows`	list	Manual mapping: one row per column.
`automaticSource`	object	Automatic mapping: where the values come from.
`conditions`	list	Which rows to match. See Conditions.
`conditionMatch`	`all` \| `any`	Whether every condition or any one must hold.
`orderBy`	list	Sort order for Get and List tables.
`limit`	integer	How many rows or tables to return.
`returnAll`	boolean	Get: return every matching row.
`keyColumns`	list	Bulk upsert: the column names a row is matched on. See Bulk upsert by key.
`rowsSourceMode`	`array` \| `file`	Bulk upsert and an automatic insert: whether the rows come from an array expression or a file.
`rowsFileExpression`	string	The file holding the rows (JSON array or JSONL), when `rowsSourceMode` is `file`.
`cursor`	string	Get: the `nextCursor` of a previous page. See Paging with a cursor.
`readOutput`	`rows` \| `file`	Get: a page of rows, or every matching row as one JSONL file.
`dryRun`	boolean	Report what a destructive operation would do, without doing it.
`optimizeBulk`	boolean	Insert: write many rows in one statement.
`applyToAllRows`	boolean	Update / Delete with no conditions: confirm it is meant for every row.
`allowTableDelete`	boolean	Delete table: confirm it.

See Flags and guards for the last five.

Output

Field	Type	Description
`resource`	string	`table` or `row`.
`operation`	string	The operation that ran.
`table.id`	string	The table's id.
`table.name`	string	The table's name.
`affected`	number	Rows (or tables) affected.
`dryRun`	boolean	Whether this was a dry run.

Each operation's own result is described in What the node outputs.

Limits and errors

See Limits and failures.

Example

workflows/register-contract.yaml

name: Register contract
description: Adds each new contract to the contract register.
graph:
  nodes:
    - id: new-contract
      type: trigger_form
      position: { x: 0, y: 0 }
      data:
        authMode: none
        fields:
          - { name: title, label: Contract title, type: text, required: true }
    - id: register
      type: data_table
      position: { x: 280, y: 0 }
      data:
        resource: row
        operation: insert
        table: dataTables.contract-register
        tableName: Contract register
        mapping: manual
        mappingRows:
          - {
              id: map-1,
              column: Contract,
              value: '{{ trigger.body.title }}',
              valueMode: expression,
            }
  edges:
    - { id: contract-to-register, source: new-contract, target: register }

In a Package

Declare the table under dataTables: in the manifest; the node references it as dataTables.<name>. See Data Tables in a Package.

Travels: resource, operation, tableName, columns, returnExistingIfExists, newTableName, columnName, columnType, mapping, mappingRows, automaticSource, conditions, conditionMatch, orderBy, limit, returnAll, keyColumns, rowsSourceMode, rowsFileExpression, cursor, readOutput, dryRun, optimizeBulk, applyToAllRows, allowTableDelete
Converted: tableId → table (a dataTables.<name> reference)
Never carried: none