Data Sources

The Data Sources panel (accessed via the database icon on the left sidebar) is where you define document schemas, manage preview records, and connect external data providers.

Creating & Connecting Collections

You can populate your workspace with data using two primary methods:

• Manual Collections (+ Add): Create standalone schema tables and populate mock rows directly inside the editor table view.
• Airtable Sync (Connect to Airtable): Link directly to an external Airtable base to import existing tables and field structures. For full configuration details, see the Airtable Integration Guide.

Root Collection & Document Generation

The first collection created or connected in your project automatically acts as the Root Collection. You can also manually set this.

Key Rules & Capabilities

• 1 Record = 1 Document: Each row in the root collection represents an individual output document instance (e.g., a single invoice, employee badge, or customized report).
• Preview Column: The root collection includes a dedicated Preview column featuring a Generate button for every row, allowing you to instantly test and render PDF outputs for specific records.
• Preview Limits: You can store up to 10 records per collection to test dynamic logic, layout tables, and variables directly on the canvas surface.

Field Types & Schema Configuration

Click the + icon in the table header to open the Create Field modal and add custom columns to a collection.

Standard Field Types

• Text: Standard string entries.
• Number: Numeric values (integers, floats, currency).
• Boolean: True or false states.
• Date: ISO timestamps and date strings.

JSON Fields (Primitive Arrays & Complex Objects)

When selecting the JSON field type, you can define whether the column stores complex objects or validated scalar arrays:

• Not Primitive: Accepts standard nested JSON objects or unconstrained complex structures.
• Primitive Arrays: Restricts and formats array elements to uniform scalar data types:
  • String (e.g., ["express", "priority", "fragile"])
  • Number (e.g., [19.99, 5.50, 42.00])
  • Boolean (e.g., [true, false, true])
  • Date (e.g., ["2026-01-01", "2026-06-01"])

Relation Fields (Linking Collections)

The Relation field type establishes relational foreign-key links between different collections or data sources.

Relation Settings

• Target Collection: Select the foreign collection to establish a connection with.
• Target Field: Choose the primary lookup field in the destination table.
• Cardinality: Determines how the linked data structure resolves during execution:
  • To-One: Resolves to a single Object ({ ... }).
  • To-Many: Resolves to an Array of Objects ([{ ... }, { ... }]).

Schema & Payload Examples

Example 1: To-One Relation (Resolves to Object)

Linking a user record to a single company profile.

{
  "id": "rec6607zl9YkWqUMI48",
  "name": "Sarah Connor",
  "company": {
    "company_name": "Cyberdyne Systems",
    "account_tier": "Enterprise"
  }
}

• Canvas Variable Usage: {{name}} and {{company.company_name}}

Example 2: To-Many Relation (Resolves to Array of Objects)

Linking an invoice root collection to multiple items.

{
  "invoice_number": "INV-2026-088",
  "client": "Acme Corp",
  "items": [
    { "sku": "ITEM-1", "description": "Consulting Hours", "amount": 150.00 },
    { "sku": "ITEM-2", "description": "Software License", "amount": 499.00 }
  ]
}

• Canvas Variable Usage: {{invoice_number}} at root level, and {{description}} / {{amount}} inside a Collection Block looping over items.

Example 3: Primitive Array JSON Field

A product item storing tag labels as a scalar string array.

{
  "product_name": "Ergonomic Desk Chair",
  "price": 299.99,
  "categories": ["Furniture", "Office", "In-Stock"]
}

• Canvas Variable Usage: Iterate over categories using a Collection Block to output tag pills.