Connection pattern (edges, nodes, pageInfo) in GraphQL - Step-by-Step Execution

Concept Flow - Connection pattern (edges, nodes, pageInfo)

Client requests connection

↓

Server fetches data list

↓

Wrap data in edges with cursor

↓

Extract nodes from edges

↓

Create pageInfo with cursors and hasNextPage

↓

Return connection object

↓

Client receives edges, nodes, pageInfo

The client asks for a list with pagination info; the server wraps items as edges with cursors, extracts nodes, builds pageInfo, and returns all.

Execution Sample

GraphQL

query {
  users(first: 2, after: "cursor1") {
    edges {
      cursor
      node {
        id
        name
      }
    }
    nodes {
      id
      name
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

This query asks for the first 2 users after a cursor, requesting edges with cursors and nodes, plus pageInfo.

Execution Table

Step	Action	Data State	Output Part
1	Receive query with first=2, after='cursor1'	No data fetched yet	Start processing
2	Fetch users after cursor1	Fetched users: [{id:3,name:'C'}, {id:4,name:'D'}, {id:5,name:'E'}]	Data ready for pagination
3	Select first 2 users	Selected users: [{id:3,name:'C'}, {id:4,name:'D'}]	Subset for response
4	Create edges with cursors	Edges: [{cursor:'cursor3', node:{id:3,name:'C'}}, {cursor:'cursor4', node:{id:4,name:'D'}}]	Edges array
5	Extract nodes from edges	Nodes: [{id:3,name:'C'}, {id:4,name:'D'}]	Nodes array
6	Build pageInfo	pageInfo: {hasNextPage:true, endCursor:'cursor4'}	Pagination info
7	Return connection object	Edges, Nodes, pageInfo ready	Complete response sent
8	Client receives response	Edges with cursors, nodes, pageInfo	Query result

💡 All requested data prepared and returned to client

Variable Tracker

Variable	Start	After Step 2	After Step 3	After Step 4	After Step 5	After Step 6	Final
fetchedUsers	[]	[{id:3,name:'C'},{id:4,name:'D'},{id:5,name:'E'}]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]
edges	[]	[]	[]	[{cursor:'cursor3',node:{id:3,name:'C'}},{cursor:'cursor4',node:{id:4,name:'D'}}]	[{cursor:'cursor3',node:{id:3,name:'C'}},{cursor:'cursor4',node:{id:4,name:'D'}}]	[{cursor:'cursor3',node:{id:3,name:'C'}},{cursor:'cursor4',node:{id:4,name:'D'}}]	[{cursor:'cursor3',node:{id:3,name:'C'}},{cursor:'cursor4',node:{id:4,name:'D'}}]
nodes	[]	[]	[]	[]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]	[{id:3,name:'C'},{id:4,name:'D'}]
pageInfo	{}	{}	{}	{}	{}	{hasNextPage:true,endCursor:'cursor4'}	{hasNextPage:true,endCursor:'cursor4'}

Key Moments - 3 Insights

Why do we have both edges and nodes in the response?

Edges include cursors for pagination and the node data; nodes are just the data items. This lets clients use cursors for paging (see execution_table step 4 and 5).

What does pageInfo's hasNextPage mean?

It tells if more data exists after the current page. In step 6, we set hasNextPage true because more users remain beyond the selected ones.

Why do we use cursors instead of page numbers?

Cursors point to exact positions in data, making pagination reliable even if data changes. This is shown in step 4 where each edge has a cursor.

Visual Quiz - 3 Questions

Look at the execution_table at step 4, what does each edge contain?

A Only the user data without cursor

B A cursor and the user data node

C Only the cursor string

D PageInfo information

💡 Hint

Check the 'Output Part' column at step 4 in execution_table

At which step does the server decide how many users to return?

A Step 3

B Step 2

C Step 5

D Step 6

💡 Hint

Look at 'Action' and 'Data State' columns in execution_table for step 3

If there were no more users after the selected ones, what would pageInfo.hasNextPage be?

A null

B true

C false

D undefined

💡 Hint

Refer to pageInfo meaning in key_moments and execution_table step 6

Concept Snapshot

Connection pattern returns paginated lists.
Data is wrapped in edges with cursors.
Nodes are extracted data items.
pageInfo shows pagination state (hasNextPage, endCursor).
Clients use cursors to fetch next pages reliably.

Full Transcript

The connection pattern in GraphQL helps clients get lists with pagination. The client sends a query asking for a number of items after a cursor. The server fetches the full list, then selects the requested slice. It wraps each item in an edge object that includes a cursor and the node data. Nodes are extracted as a simple list of data items. The server also creates pageInfo with information like whether more pages exist and the cursor of the last item. This connection object is returned to the client. The client uses edges and pageInfo to navigate pages smoothly. This pattern avoids problems with page numbers and changing data. The execution table shows each step from receiving the query, fetching data, slicing, wrapping edges, extracting nodes, building pageInfo, and returning the response. Variables track the data state at each step. Key moments clarify why edges and nodes both exist, what pageInfo means, and why cursors are used. The quiz tests understanding of edges content, when slicing happens, and pageInfo meaning. The snapshot summarizes the pattern in a few lines for quick recall.