Table Join

Description

The Table Join activity joins two previous activity output tables using DuckDB. It reads the selected left and right activity tables, applies the configured join type and conditions, and writes the joined result as transformed data.

Supported Features

• Two-table joins: Join a left activity table with a right activity table.
• Multiple join types: Supports join type values supplied by the table join editor, including inner, left, right, full outer, and cross join forms.
• Multiple conditions: Combines join conditions with AND.
• Column-to-column conditions: Compare a left table column to a right table column.
• Column-to-value conditions: Compare a left table column to a literal value when Selected is enabled.
• Text operators: Supports operators such as contains, icontains, startswith, endswith, like, ilike, notlike, notilike, and rlike.
• Cross join support: For CROSS JOIN, no ON clause is generated.

---

Input

Type	Required	Description
Data	Yes	Data tables produced by previous activities. The configured LeftTable and RightTable identify which previous activity outputs are joined.

Input Scenarios

1. Column-To-Column Join

Left table:

CustomerId	Name
1	Alice
2	Bob

Right table:

CustomerId	OrderId	Amount
1	O-001	120
2	O-002	90

2. Column-To-Value Join

Use when filtering the left table during the join condition, for example Status = Active.

3. Cross Join

Use when every row from the left table should be paired with every row from the right table. Conditions are ignored.

---

Output

Field	Type	Description
Data	Array	Joined rows containing columns from both left and right tables.

Example Output

{
  "Data": [
    {
      "CustomerId": 1,
      "Name": "Alice",
      "OrderId": "O-001",
      "Amount": 120
    },
    {
      "CustomerId": 2,
      "Name": "Bob",
      "OrderId": "O-002",
      "Amount": 90
    }
  ]
}

---

Configuration Fields

Field Name	Type	Required	Description
JoinValues	Table Join Editor	Yes	Join configuration containing the selected left table, right table, join type, and join conditions.
LeftTable	Activity Table	Yes	Previous activity output used as the left side of the join.
RightTable	Activity Table	Yes	Previous activity output used as the right side of the join.
JoinType	Dropdown	Yes	SQL join type inserted between the left and right tables. CROSS JOIN is treated specially and does not use conditions.
JoinConditions	Object Array	Conditional	Conditions used in the join ON clause. Required for non-cross joins. Multiple conditions are joined using AND.
LeftColumn	Column	Yes	Column from the left table used in the condition.
Operator	Text/Dropdown	Yes	Comparison operator. Standard SQL operators and supported text operators can be used.
RightColumn	Column or Value	Yes	Right table column when Selected is false; literal comparison value when Selected is true.
Selected	Boolean	No	When false, builds a column-to-column condition. When true, builds a column-to-value condition using RightColumn as a literal value.

Conditional Field Rendering Rules

The rendering behavior is controlled by the table join editor. In execution:

• JoinConditions are not used when JoinType is CROSS JOIN.
• RightColumn is treated as a literal value when Selected is true.
• RightColumn is treated as a right-table column when Selected is false.

---

Sample Configurations

Scenario 1: Inner Join On CustomerId

{
  "JoinValues": {
    "leftTable": "Customers",
    "rightTable": "Orders",
    "joinType": "INNER JOIN",
    "joinConditions": [
      {
        "LeftColumn": "CustomerId",
        "Operator": "=",
        "RightColumn": "CustomerId",
        "Selected": false
      }
    ]
  }
}

Scenario 2: Join With Literal Value

{
  "JoinValues": {
    "leftTable": "Customers",
    "rightTable": "Orders",
    "joinType": "INNER JOIN",
    "joinConditions": [
      {
        "LeftColumn": "Status",
        "Operator": "=",
        "RightColumn": "Active",
        "Selected": true
      }
    ]
  }
}

---

Sample Output

{
  "Data": [
    {
      "CustomerId": 1,
      "Name": "Alice",
      "Status": "Active",
      "OrderId": "O-001",
      "Amount": 120
    }
  ]
}