protobuf-array-stream

Stream and decode Protocol Buffer arrays without memory issues. While other libraries need the entire message in memory, this one processes array elements directly from the source stream - making it perfect for large datasets.

Why Use This?

• 💾 Memory Efficient: No need to load entire message in memory
• 🌊 Streaming: Process array elements directly from source (file, network, etc.)
• 🎯 Specialized: Built for messages with a repeated field (array)
• ⚡ Scalable: Works with arrays of any size, memory usage stays constant

Installation

npm install protobuf-array-stream

Usage

Loading Protocol Definitions

First, you need to load your Protocol Buffer definitions. The library exposes these methods directly from protobufjs for convenience:

import { RepeatedFieldStream } from "protobuf-array-stream";
import { createReadStream } from "fs";

// 1. Load from a .proto file (async)
const root1 = await RepeatedFieldStream.loadRootFromFile(
  "path/to/messages.proto",
);

// 2. Load from a .proto file (sync)
const root2 = RepeatedFieldStream.loadRootFromFileSync(
  "path/to/messages.proto",
);

// 3. Load from a proto definition string
const root3 = RepeatedFieldStream.loadRootFromSource(`
  syntax = "proto3";
  
  message NumberList {
    repeated int32 numbers = 1;
  }
`);

// 4. Load from JSON schema
const root4 = RepeatedFieldStream.loadRootFromJSON({
  nested: {
    NumberList: {
      fields: {
        numbers: {
          id: 1,
          rule: "repeated",
          type: "int32",
        },
      },
    },
  },
});

Real-World Example

You're building an anti-cheat system for your massively multiplayer game. Players are generating TONS of data:

syntax = "proto3";

message GameLog {
  repeated PlayerAction actions = 1;
}

message PlayerAction {
  int64 timestamp = 1;
  string player_id = 2;
  string action = 3;
  Position position = 4;
}

message Position {
  float x = 1;
  float y = 2;
  float z = 3;
}

Process large game session logs efficiently:

const sessionStream = createReadStream("game_session.bin");

const decodeStream = new RepeatedFieldStream({
  root,
  messageName: "GameLog",
});

const analyzeStream = new Transform({
  objectMode: true,
  transform(action, _, callback) {
    const { timestamp, playerId, action: move, position } = action;

    if (move === "shoot" && position.y > 500) {
      console.log(
        `Detected: ${playerId} at height ${position.y} on ${new Date(timestamp.toNumber())}`,
      );
    }

    callback(null, action);
  },
});

sessionStream.pipe(decodeStream).pipe(analyzeStream);

Error Handling

The stream emits errors in these cases:

• Message type not found in proto definitions
• Message has no fields or multiple fields
• Field is not a repeated field
• Invalid protobuf encoding in input

stream.on("error", (error) => {
  console.error("Stream error:", error.message);
});

API

new RepeatedFieldStream(options)

Creates a new transform stream for processing Protocol Buffer repeated fields.

Options

• root: Root namespace from protobufjs containing message definitions
• messageName: Name of the message type to decode

Static Methods

loadRootFromFile(path: string): Promise<Root>

Asynchronously loads Protocol Buffer definitions from a .proto file.

• path: Path to the .proto file
• Returns: Promise that resolves to the root namespace

loadRootFromFileSync(path: string): Root

Synchronously loads Protocol Buffer definitions from a .proto file.

• path: Path to the .proto file
• Returns: Root namespace

loadRootFromSource(source: string): Root

Loads Protocol Buffer definitions from a string containing proto definitions.

• source: String containing the proto definitions
• Returns: Root namespace

loadRootFromJSON(json: object): Root

Loads Protocol Buffer definitions from a JSON object.

• json: JSON object containing the proto definitions
• Returns: Root namespace

Requirements

• Node.js >= 16.0.0
• Protocol Buffer messages must contain exactly one repeated field

License

MIT License - see the LICENSE file for details