Skip to main content

Data Model: MongoDB Persistence

NetPad uses MongoDB Atlas for cloud persistence of diagrams, shapes, and user data. All MongoDB access is handled via connectDB from @/lib/mongodb.

Diagram Document Structure​

A typical diagram document in MongoDB includes:

{
  "_id": ObjectId,
  "userId": "user@example.com" | null, // null for anonymous
  "title": "Network Diagram",
  "createdAt": ISODate,
  "updatedAt": ISODate,
  "shapes": [ ... ], // Array of shape objects
  "connections": [ ... ], // Array of connector objects
  // ...other metadata
}

Shape Object Structure​

Each shape in the shapes array should have:

  • id: Unique identifier
  • type: Shape type (e.g., rectangle, circle, hexagon, etc.)
  • x, y: Position
  • width, height: Dimensions
  • label: Text label
  • fill: Fill color
  • stroke: Stroke color
  • strokeWidth: Stroke width
  • Shape-specific properties (e.g., points for polygons)

Connector Object Structure​

Each connector in the connections array should have:

  • id: Unique identifier
  • from: Shape id and connector point
  • to: Shape id and connector point
  • type: Connector type (e.g., straight, elbow)
  • label: Optional label

Best Practices​

  • All shape and diagram properties must be serializable to MongoDB.
  • Use Mongoose models for validation and schema enforcement.
  • Support both anonymous and authenticated users (use userId or sessionId).
  • Store timestamps for auditing and sorting.

For more details, see the code in src/models/Shape.js and related files.