Using gRPC

This guide provides practical examples for querying the Sui network using gRPC. For core concepts, see the corresponding concepts page.

Before you begin, ensure you have access to a gRPC-enabled Sui full node. Check the list of RPC and data providers that support gRPC on their full nodes, and contact a provider directly to request access.

If your provider doesn't yet support gRPC, you can:

• Ask them to enable it.

• Reach out to the Sui Foundation team on Discord or Telegram for help.

Field masks

A FieldMask in protocol buffers is a mechanism used to specify a subset of fields within a message that should be read, updated, or returned. Instead of retrieving the entire object, a client can request only the specific fields they need by providing a list of field paths. This improves performance and reduces unnecessary data transfer.

In the Sui gRPC API, FieldMasks are used in requests like GetTransaction, GetObject, and so on to control which parts are included in the response (like the effects and events of a transaction).

Field masks are defined using google.protobuf.FieldMask and typically appear in the request message as read_mask. You can pass an explicit value of * to request all fields.

If you omit read_mask, it defaults to * (all fields), unless documented otherwise.

• Each field path in the mask must match the field structure of the response proto message. Nested fields are supported using dot notation.

• In batch APIs, only the top-level read_mask is respected. The API ignores any masks inside sub-requests.

• In some cases, non-terminal repeated fields might be supported in the mask, even if this is atypical per standard FieldMask behavior.

Field presence

When using gRPC with Sui, it's important to understand how field presence works, especially when dealing with proto3 syntax. In proto3, primitive fields like numbers, booleans, and strings are always initialized to a default value if not present in the message. This means you cannot tell whether a value is explicitly set or just left out. To give you that distinction, Sui marks all fields as optional, even if they are required by the API.

As a user of the API, this lets you detect whether a field value is actually provided or just defaulted and write clients that can perform partial updates or simulate intent like distinguishing between an explicitly empty input versus a missing one.

:::info

If a field is marked optional in the proto, it might still be required for the request to be valid. This is a protobuf quirk, not an indication of actual business logic.

:::

Encoding

In the Sui gRPC API, identifiers with standard human-readable formats are represented as strings in the proto schema:

• Address and ObjectId: Represented as 64 hexadecimal characters with a leading 0x.

• Digests: Represented as Base58.

• TypeTag and StructTag: Represented in their canonical string format (such as 0x0000000000000000000000000000000000000000000000000000000000000002::coin::Coin<0x0000000000000000000000000000000000000000000000000000000000000002::sui::SUI>)

Pagination

When using gRPC APIs that return lists of data like account balances, owned objects, and so on, you typically need to handle pagination. These APIs return results in chunks and include tokens to help you request the next batch.

In the request, provide a page_size to control how many items you want returned. If you leave this unset or set it to 0, the API uses a sensible default. You can also include a page_token in the request, which tells the server where to continue from. You get this token from the previous response.

The response includes a list of results and a next_page_token value, which you can pass into your next request to get the next page. When the server returns an empty next_page_token, you have reached the end of the list.

Make sure to keep all other parameters in your request the same between paginated calls. Otherwise, the server might reject the request with an INVALID_ARGUMENT error.

Errors

The Sui gRPC services follow the richer error model defined in AIP-193. When an RPC returns a non-OK status code, detailed error information is typically included in the grpc-status-details-bin header. This header contains a google.rpc.Status message encoded in Base64.

You can decode this message to access structured error details, which might include specific causes, context, or metadata. This makes it easier to understand and handle errors programmatically in your client applications.

HTTP headers

In many gRPC responses, the Sui API includes additional metadata in the form of HTTP headers. These headers provide contextual information about the current network state and might be useful for debugging, telemetry, or understanding the data's freshness.

Here are the headers you might encounter:

• x-sui-chain-id: The chain ID of the current network.

• x-sui-chain: A human-readable name for the current network (mainnet, testnet, or devnet).

• x-sui-checkpoint-height: The height of the latest checkpoint at the time of the response.

• x-sui-lowest-available-checkpoint: The earliest checkpoint for which transaction and checkpoint data can still be queried.

• x-sui-lowest-available-checkpoint-objects: The earliest checkpoint from which object data (input and output) is available.

• x-sui-epoch: The current epoch of the network.

• x-sui-timestamp-ms: The network timestamp in milliseconds since the Unix epoch.

• x-sui-timestamp: The network timestamp in milliseconds since the Unix epoch in human-readable RFC 3339 format.

Not all headers are guaranteed to be present in every API response. They are only included when applicable to the given RPC.

Access data using grpcurl

Interact with gRPC by using grpcurl.

List available gRPC services

$ grpcurl <full node URL:port> list

The port on Sui Foundation managed full nodes is 443.

List available APIs in the LedgerService

$ grpcurl <full node URL:port> list sui.rpc.v2.LedgerService

Get the events and effects details of a particular transaction

$ grpcurl -d '{ "digest": "J4NvV5iQZQFm1xKPYv9ffDCCPW6cZ4yFKsCqFUiDX5L4" }' <full node URL:port> sui.rpc.v2.LedgerService/GetTransaction

Get the transactions in a particular checkpoint

$ grpcurl -d '{ "sequence_number": "164329987", "read_mask": { "paths": ["transactions"]} }' <full node URL:port> sui.rpc.v2.LedgerService/GetCheckpoint

Get the latest information for a coin type

$ grpcurl -d '{ "coin_type": "0x2::sui::SUI" }' <full node URL:port> sui.rpc.v2.StateService/GetCoinInfo

List the objects owned by a particular address

$ grpcurl -d '{ "owner": "0x94096a6a54129234237759c66e6ef1037224fb3102a0ae29d33b490281c8e4d5" }' <full node URL:port> sui.rpc.v2.StateService/ListOwnedObjects

List the dynamic fields in a particular object

$ grpcurl -d '{ "parent": "0xb57fba584a700a5bcb40991e1b2e6bf68b0f3896d767a0da92e69de73de226ac" }' <full node URL:port> sui.rpc.v2.StateService/ListDynamicFields

Access streaming data with Buf

grpcurl does not support server-side streaming RPCs. To test or experiment with SubscriptionService, use the Buf CLI instead.

$ buf curl --protocol grpc https://<full node URL>/sui.rpc.v2.SubscriptionService/SubscribeCheckpoints -d '{ "readMask": "sequenceNumber,digest,summary.timestamp" }'  --timeout 1m

Build gRPC clients

:::info

It is recommended to use one of the supported SDKs for building gRPC clients. If those SDKs do not satisfy your use case, refer to the examples below. :::

This example shows how to build a TypeScript client for the Sui gRPC API.

Step 1: Install dependencies

npm init -y
npm install @grpc/grpc-js @grpc/proto-loader
npm i -D tsx

The default project has this structure:

.
├── protos/
│   └── sui/
│       └── node/
│           └── v2/
│               ├── ledger_service.proto
│               └── *.proto
├── client.ts
├── package.json

Download all of the sui/rpc/v2 proto files from Github in the same folder.

Step 2: Edit client.ts to get events and effects details of a particular transaction

import * as grpc from '@grpc/grpc-js';
import * as protoLoader from '@grpc/proto-loader';
import * as path from 'path';

const PROTO_PATH = path.join(__dirname, 'protos/sui/rpc/v2/ledger_service.proto');

// Load proto definitions
const packageDefinition = protoLoader.loadSync(PROTO_PATH, {
  keepCase: true,
  longs: String,
  enums: String,
  defaults: true,
  oneofs: true,
  includeDirs: [path.join(__dirname, 'protos')],
});

const suiProto = grpc.loadPackageDefinition(packageDefinition) as any;
const LedgerService = suiProto.sui.rpc.v2.LedgerService;

// Create gRPC client
const client = new LedgerService(
  '<full node URL>:443',
  grpc.credentials.createSsl()
);

// Sample transaction digest in Base58 format
const base58Digest = '3ByWphQ5sAVojiTrTrGXGM5FmCVzpzYmhsjbhYESJtxp';

// Construct the request
const request = {
  digest: base58Digest,
  read_mask: {
    paths: ['events', 'effects'],
  },
};

// Make gRPC call
client.GetTransaction(request, (err: any, response: any) => {
  if (err) {
    console.error('Error:', err);
  } else {
    console.log('Response:', JSON.stringify(response, null, 2));
  }
});

Step 3: Run the sample client

npx tsx c

• proto-loader handles any nested .proto files. Just make sure paths and imports are correct.

• The example assumes that gRPC is available on port 443 which requires SSL.

• Digest in the request is directly provided in the Base58 format, but check if you need to decode from your source format.

This example shows how to build a Go client for the Sui gRPC API.

Step 1: Install dependencies

First make sure you have go and protoc installed in your environment, and then install:

go install google.golang.org/protobuf/cmd/protoc-gen-go@latest
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@latest

Step 2: Edit go.mod

Add the following content, making sure to update the version numbers to the latest versions:

require (
  google.golang.org/grpc v1.60.0
  google.golang.org/protobuf v1.33.0
)

Step 3: Generate Go code from proto files

Assuming you have the v2 proto files from Github, run:

protoc --proto_path=./protos --go_out=. --go-grpc_out=. protos/sui/rpc/v2/ledger_service.proto

Step 4: Edit main.go to get events and effects details of a particular transaction

package main

import (
    "context"
    "crypto/tls"
    "fmt"
    "log"
    "time"

    "google.golang.org/grpc"
    "google.golang.org/grpc/credentials"
    
    pb "your_project/sui/rpc/v2" // adjust path based on where your generated .pb.go files are
)

func main() {
    // Set up gRPC connection with TLS (port 443)
		creds := credentials.NewTLS(&tls.Config{})
		conn, err := grpc.Dial("<full node URL>:443", grpc.WithTransportCredentials(creds))
		if err != nil {
			log.Fatalf("failed to connect: %v", err)
		}
		defer conn.Close()

    client := pb.NewLedgerServiceClient(conn)
    
    // Sample transaction digest in Base58 format
    base58Digest := "3ByWphQ5sAVojiTrTrGXGM5FmCVzpzYmhsjbhYESJtxp"

    // Build request
		req := &pb.GetTransactionRequest{
			Digest: base58Digest,
			ReadMask: &pb.TransactionReadMask{
				Paths: []string{"events", "effects"},
			},
		}

    // Make the request
    ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
    defer cancel()

    resp, err := client.GetTransaction(ctx, req)
    if err != nil {
        log.Fatalf("GetTransaction failed: %v", err)
    }

    // Print the response
    fmt.Printf("Response:\n%+v\n", resp)
}

Step 5: Run the sample client

If your go.mod is properly set up, and you've already generated the gRPC code (*.pb.go, *_grpc.pb.go), simply run:

go run main.go

If you see errors like cannot find package, ensure you've generated the proto files and imported them correctly. If your generated files are in a nested directory like sui/rpc/v2, your import in main.go should match:

import pb "your_project/sui/rpc/v2"

You can replace your_project with a module name or relative import path depending on how your go.mod is defined.

The example assumes that gRPC is available on port 443 which requires SSL. The digest in the request is directly provided in the Base58 format, but check if you need to decode from your source format.

This example shows how to build a Python client for the Sui gRPC API. Optionally, refer to pysui, a community contributed SDK.

Step 1: Install dependencies

First make sure you have python and protoc installed in your environment, and then install:

pip install grpcio grpcio-tools protobuf

Step 2: Generate Python code from proto files

Assuming you have the v2 proto files from Github, run:

python -m grpc_tools.protoc -I./protos --python_out=. --grpc_python_out=. ./protos/sui/rpc/v2/ledger_service.proto

Step 3: Edit client.py to get events and effects details of a particular transaction

import grpc
from sui.rpc.v2 import ledger_service_pb2, ledger_service_pb2_grpc

def main():
    # Create secure channel to port 443
    channel = grpc.secure_channel("<full node URL>:443", grpc.ssl_channel_credentials())
    stub = ledger_service_pb2_grpc.LedgerServiceStub(channel)

    # Sample transaction digest in Base58 format
    base58_digest = "3ByWphQ5sAVojiTrTrGXGM5FmCVzpzYmhsjbhYESJtxp"

    # Build the request
    request = ledger_service_pb2.GetTransactionRequest(
        digest=base58_digest,
        read_mask=ledger_service_pb2.TransactionReadMask(paths=["events", "effects"])
    )

    # Make the RPC call
    response = stub.GetTransaction(request)

    # Print response
    print(response)

if __name__ == "__main__":
    main()

Step 4: Run the sample client

python client.py

The import paths like sui.rpc.v2.ledger_service_pb2 depend on your proto structure. You might need to adjust __init__.py files or $PYTHONPATH to ensure proper module resolution.

The example assumes that gRPC is available on port 443 which requires SSL. Digest in the request is directly provided in the Base58 format, but check if you need to decode from your source format.