NATSBridge/docs/implementation.md

Implementation Guide: Bi-Directional Data Bridge

Overview

This document describes the implementation of the high-performance, bi-directional data bridge between Julia and JavaScript services using NATS (Core & JetStream), implementing the Claim-Check pattern for large payloads.

Multi-Payload Support

The implementation uses a standardized list-of-tuples format for all payload operations. Even when sending a single payload, the user must wrap it in a list.

API Standard:

# Input format for smartsend (always a list of tuples with type info)
[(dataname1, data1, type1), (dataname2, data2, type2), ...]

# Output format for smartreceive (returns envelope dictionary with payloads field)
# Returns: Dict with envelope metadata and payloads field containing list of tuples
# {
#   "correlationId": "...",
#   "msgId": "...",
#   "timestamp": "...",
#   "sendTo": "...",
#   "msgPurpose": "...",
#   "senderName": "...",
#   "senderId": "...",
#   "receiverName": "...",
#   "receiverId": "...",
#   "replyTo": "...",
#   "replyToMsgId": "...",
#   "brokerURL": "...",
#   "metadata": {...},
#   "payloads": [(dataname1, data1, type1), (dataname2, data2, type2), ...]
# }

Where type can be: "text", "dictionary", "table", "image", "audio", "video", "binary"

Examples:

# Single payload - still wrapped in a list (type is required as third element)
smartsend("/test", [(dataname1, data1, "text")], ...)

# Multiple payloads in one message (each payload has its own type)
smartsend("/test", [(dataname1, data1, "dictionary"), (dataname2, data2, "table")], ...)

# Receive returns a dictionary envelope with all metadata and deserialized payloads
envelope = smartreceive(msg, ...)
# envelope["payloads"] = [(dataname1, data1, "text"), (dataname2, data2, "table"), ...]
# envelope["correlationId"], envelope["msgId"], etc.

Architecture

The implementation follows the Claim-Check pattern:

┌─────────────────────────────────────────────────────────────────────────┐
│                         SmartSend Function                              │
└─────────────────────────────────────────────────────────────────────────┘
                             │
                             ▼
┌─────────────────────────────────────────────────────────────────────────┐
│  Is payload size < 1MB?                                                 │
└─────────────────────────────────────────────────────────────────────────┘
                             │
           ┌─────────────────┴─────────────────┐
           ▼                                   ▼
     ┌─────────────────┐                 ┌─────────────────┐
     │  Direct Path    │                 │  Link Path      │
     │  (< 1MB)        │                 │  (> 1MB)        │
     │                 │                 │                 │
     │ • Serialize to  │                 │ • Serialize to  │
     │   IOBuffer      │                 │   IOBuffer      │
     │ • Base64 encode │                 │ • Upload to     │
     │ • Publish to    │                 │   HTTP Server   │
     │   NATS          │                 │ • Publish to    │
     │                 │                 │   NATS with URL │
     └─────────────────┘                 └─────────────────┘

Files

Julia Module: src/julia_bridge.jl

The Julia implementation provides:

• MessageEnvelope: Struct for the unified JSON envelope
• SmartSend(): Handles transport selection based on payload size
• SmartReceive(): Handles both direct and link transport

JavaScript Module: src/NATSBridge.js

The JavaScript implementation provides:

• MessageEnvelope class: For the unified JSON envelope
• MessagePayload class: For individual payload representation
• smartsend(): Handles transport selection based on payload size
• smartreceive(): Handles both direct and link transport

Installation

Julia Dependencies

using Pkg
Pkg.add("NATS")
Pkg.add("Arrow")
Pkg.add("JSON3")
Pkg.add("HTTP")
Pkg.add("UUIDs")
Pkg.add("Dates")

JavaScript Dependencies

npm install nats.js apache-arrow uuid base64-url

Usage Tutorial

Step 1: Start NATS Server

docker run -p 4222:4222 nats:latest

Step 2: Start HTTP File Server (optional)

# Create a directory for file uploads
mkdir -p /tmp/fileserver

# Use any HTTP server that supports POST for file uploads
# Example: Python's built-in server
python3 -m http.server 8080 --directory /tmp/fileserver

Step 3: Run Test Scenarios

# Scenario 1: Command & Control (JavaScript sender)
node test/scenario1_command_control.js

# Scenario 2: Large Arrow Table (JavaScript sender)
node test/scenario2_large_table.js

# Scenario 3: Julia-to-Julia communication
# Run both Julia and JavaScript versions
julia test/scenario3_julia_to_julia.jl
node test/scenario3_julia_to_julia.js

Usage

Scenario 0: Basic Multi-Payload Example

Julia (Sender)

using NATSBridge

# Send multiple payloads in one message (type is required per payload)
smartsend(
    "/test",
    [("dataname1", data1, "dictionary"), ("dataname2", data2, "table")],
    nats_url="nats://localhost:4222",
    fileserver_url="http://localhost:8080",
    metadata=Dict("custom_key" => "custom_value")
)

# Even single payload must be wrapped in a list with type
smartsend("/test", [("single_data", mydata, "dictionary")])

Julia (Receiver)

using NATSBridge

# Receive returns a dictionary envelope with all metadata and deserialized payloads
envelope = smartreceive(msg, "http://localhost:8080")
# envelope["payloads"] = [(dataname1, data1, "dictionary"), (dataname2, data2, "table"), ...]
# envelope["correlationId"], envelope["msgId"], etc.

Scenario 1: Command & Control (Small JSON)

JavaScript (Sender)

const { smartsend } = require('./src/NATSBridge');

// Single payload wrapped in a list
const config = [{
    dataname: "config",
    data: { step_size: 0.01, iterations: 1000 },
    type: "dictionary"
}];

await smartsend("control", config, {
    correlationId: "unique-id"
});

// Multiple payloads
const configs = [
    {
        dataname: "config1",
        data: { step_size: 0.01 },
        type: "dictionary"
    },
    {
        dataname: "config2", 
        data: { iterations: 1000 },
        type: "dictionary"
    }
];

await smartsend("control", configs);

Julia (Receiver)

using NATS
using JSON3

# Subscribe to control subject
subscribe(nats, "control") do msg
    env = MessageEnvelope(String(msg.data))
    config = JSON3.read(env.payload)
    
    # Execute simulation with parameters
    step_size = config.step_size
    iterations = config.iterations
    
    # Send acknowledgment
    response = Dict("status" => "Running", "correlation_id" => env.correlation_id)
    publish(nats, "control_response", JSON3.stringify(response))
end

JavaScript (Receiver)

const { smartreceive } = require('./src/NATSBridge');

// Subscribe to messages
const nc = await connect({ servers: ['nats://localhost:4222'] });
const sub = nc.subscribe("control");

for await (const msg of sub) {
    const envelope = await smartreceive(msg);
    
    // Process the payloads from the envelope
    for (const payload of envelope.payloads) {
        const { dataname, data, type } = payload;
        console.log(`Received ${dataname} of type ${type}`);
        console.log(`Data: ${JSON.stringify(data)}`);
    }
    
    // Also access envelope metadata
    console.log(`Correlation ID: ${envelope.correlationId}`);
    console.log(`Message ID: ${envelope.msgId}`);
}

Scenario 2: Deep Dive Analysis (Large Arrow Table)

Julia (Sender)

using Arrow
using DataFrames

# Create large DataFrame
df = DataFrame(
    id = 1:10_000_000,
    value = rand(10_000_000),
    category = rand(["A", "B", "C"], 10_000_000)
)

# Send via SmartSend - wrapped in a list (type is part of each tuple)
await SmartSend("analysis_results", [("table_data", df, "table")]);

JavaScript (Receiver)

const { smartreceive } = require('./src/NATSBridge');

const envelope = await smartreceive(msg);

// Use table data from the payloads field
// Note: Tables are sent as arrays of objects in JavaScript
const table = envelope.payloads;

Scenario 3: Live Binary Processing

JavaScript (Sender)

const { smartsend } = require('./src/NATSBridge');

// Binary data wrapped in a list
const binaryData = [{
    dataname: "audio_chunk",
    data: binaryBuffer,  // ArrayBuffer or Uint8Array
    type: "binary"
}];

await smartsend("binary_input", binaryData, {
    metadata: {
        sample_rate: 44100,
        channels: 1
    }
});

Julia (Receiver)

using WAV
using DSP

# Receive binary data
function process_binary(data)
    # Perform FFT or AI transcription
    spectrum = fft(data)
    
    # Send results back (JSON + Arrow table)
    results = Dict("transcription" => "sample text", "spectrum" => spectrum)
    await SmartSend("binary_output", results, "json")
end

JavaScript (Receiver)

const { smartreceive } = require('./src/NATSBridge');

// Receive binary data
function process_binary(msg) {
    const envelope = await smartreceive(msg);
    
    // Process the binary data from envelope.payloads
    for (const payload of envelope.payloads) {
        if (payload.type === "binary") {
            // data is an ArrayBuffer or Uint8Array
            console.log(`Received binary data: ${payload.dataname}, size: ${payload.data.length}`);
            // Perform FFT or AI transcription here
        }
    }
}

Scenario 4: Catch-Up (JetStream)

Julia (Producer)

using NATSBridge

function publish_health_status(nats_url)
    # Send status wrapped in a list (type is part of each tuple)
    status = Dict("cpu" => rand(), "memory" => rand())
    smartsend("health", [("status", status, "dictionary")], nats_url=nats_url)
    sleep(5)  # Every 5 seconds
end

JavaScript (Consumer)

const { connect } = require('nats');
const { smartreceive } = require('./src/NATSBridge');

const nc = await connect({ servers: ['nats://localhost:4222'] });
const js = nc.jetstream();

// Request replay from last 10 minutes
const consumer = await js.pullSubscribe("health", { 
    durable_name: "catchup",
    max_batch: 100,
    max_ack_wait: 30000
});

// Process historical and real-time messages
for await (const msg of consumer) {
    const envelope = await smartreceive(msg);
    // envelope.payloads contains the list of payloads
    // Each payload has: dataname, data, type
    msg.ack();
}

Scenario 5: Selection (Low Bandwidth)

Focus: Small Arrow tables, Julia to JavaScript. The Action: Julia wants to send a small DataFrame to show on a JavaScript dashboard for the user to choose.

Julia (Sender):

using NATSBridge
using DataFrames

# Create small DataFrame (e.g., 50KB - 500KB)
options_df = DataFrame(
    id = 1:10,
    name = ["Option A", "Option B", "Option C", "Option D", "Option E",
            "Option F", "Option G", "Option H", "Option I", "Option J"],
    description = ["Description A", "Description B", "Description C", "Description D", "Description E",
                   "Description F", "Description G", "Description H", "Description I", "Description J"]
)

# Convert to Arrow IPC stream
# Check payload size (< 1MB threshold)
# Publish directly to NATS with Base64-encoded payload
# Include metadata for dashboard selection context
smartsend(
    "dashboard.selection",
    [("options_table", options_df, "table")],
    nats_url="nats://localhost:4222",
    metadata=Dict("context" => "user_selection")
)

JavaScript (Receiver):

const { smartreceive, smartsend } = require('./src/NATSBridge');

// Receive NATS message with direct transport
const envelope = await smartreceive(msg);

// Decode Base64 payload (for direct transport)
// For tables, data is in envelope.payloads
const table = envelope.payloads;  // Array of objects

// User makes selection
const selection = uiComponent.getSelectedOption();

// Send selection back to Julia
await smartsend("dashboard.response", [
    { dataname: "selected_option", data: selection, type: "dictionary" }
]);

Use Case: Julia server generates a list of available options (e.g., file selections, configuration presets) as a small DataFrame and sends to JavaScript dashboard for user selection. The selection is then sent back to Julia for processing.

Scenario 6: Chat System

Focus: Every conversational message is composed of any number and any combination of components, spanning the full spectrum from small to large. This includes text, images, audio, video, tables, and files—specifically accommodating everything from brief snippets to high-resolution images, large audio files, extensive tables, and massive documents. Support for claim-check delivery and full bi-directional messaging.

Multi-Payload Support: The system supports mixed-payload messages where a single message can contain multiple payloads with different transport strategies. The smartreceive function iterates through all payloads in the envelope and processes each according to its transport type.

Julia (Sender/Receiver):

using NATSBridge
using DataFrames

# Build chat message with mixed payloads:
# - Text: direct transport (Base64)
# - Small images: direct transport (Base64)
# - Large images: link transport (HTTP URL)
# - Audio/video: link transport (HTTP URL)
# - Tables: direct or link depending on size
# - Files: link transport (HTTP URL)
# 
# Each payload uses appropriate transport strategy:
# - Size < 1MB → direct (NATS + Base64)
# - Size >= 1MB → link (HTTP upload + NATS URL)
# 
# Include claim-check metadata for delivery tracking
# Support bidirectional messaging with replyTo fields

# Example: Chat with text, small image, and large file
chat_message = [
    ("message_text", "Hello, this is a test message!", "text"),
    ("user_avatar", image_bytes, "image"),  # Small image, direct transport
    ("large_document", large_file_bytes, "binary")  # Large file, link transport
]

smartsend(
    "chat.room123",
    chat_message,
    nats_url="nats://localhost:4222",
    msg_purpose="chat",
    reply_to="chat.room123.responses"
)

JavaScript (Sender/Receiver):

const { smartsend, smartreceive } = require('./src/NATSBridge');

// Build chat message with mixed content:
// - User input text: direct transport
// - Selected image: check size, use appropriate transport
// - Audio recording: link transport for large files
// - File attachment: link transport
// 
// Parse received message:
// - Direct payloads: decode Base64
// - Link payloads: fetch from HTTP with exponential backoff
// - Deserialize all payloads appropriately
// 
// Render mixed content in chat interface
// Support bidirectional reply with claim-check delivery confirmation

// Example: Send chat with mixed content
const message = [
    {
        dataname: "text",
        data: "Hello from JavaScript!",
        type: "text"
    },
    {
        dataname: "image",
        data: selectedImageBuffer,  // Small image (ArrayBuffer or Uint8Array)
        type: "image"
    },
    {
        dataname: "audio",
        data: audioUrl,  // Large audio, link transport
        type: "audio"
    }
];

await smartsend("chat.room123", message);

Use Case: Full-featured chat system supporting rich media. User can send text, small images directly, or upload large files that get uploaded to HTTP server and referenced via URLs. Claim-check pattern ensures reliable delivery tracking for all message components.

Implementation Note: The smartreceive function iterates through all payloads in the envelope and processes each according to its transport type. See the standard API format in Section 1: msgEnvelope_v1 supports AbstractArray{msgPayload_v1} for multiple payloads.

Configuration

Environment Variables

Variable	Default	Description
NATS_URL	nats://localhost:4222	NATS server URL
FILESERVER_URL	http://localhost:8080	HTTP file server URL (base URL without /upload suffix)
SIZE_THRESHOLD	1_000_000	Size threshold in bytes (1MB)

Message Envelope Schema

{
  "correlationId": "uuid-v4-string",
  "msgId": "uuid-v4-string",
  "timestamp": "2024-01-15T10:30:00Z",
  
  "sendTo": "topic/subject",
  "msgPurpose": "ACK | NACK | updateStatus | shutdown | chat",
  "senderName": "agent-wine-web-frontend",
  "senderId": "uuid4",
  "receiverName": "agent-backend",
  "receiverId": "uuid4",
  "replyTo": "topic",
  "replyToMsgId": "uuid4",
  "BrokerURL": "nats://localhost:4222",
  
  "metadata": {
    "content_type": "application/octet-stream",
    "content_length": 123456
  },
  
  "payloads": [
    {
      "id": "uuid4",
      "dataname": "login_image",
      "type": "image",
      "transport": "direct",
      "encoding": "base64",
      "size": 15433,
      "data": "base64-encoded-string",
      "metadata": {
        "checksum": "sha256_hash"
      }
    }
  ]
}

Performance Considerations

Zero-Copy Reading

• Use Arrow's memory-mapped file reading
• Avoid unnecessary data copying during deserialization
• Use Apache Arrow's native IPC reader

Exponential Backoff

• Maximum retry count: 5
• Base delay: 100ms, max delay: 5000ms
• Implemented in both Julia and JavaScript implementations

Correlation ID Logging

• Log correlation_id at every stage
• Include: send, receive, serialize, deserialize
• Use structured logging format

Testing

Run the test scripts:

# Scenario 1: Command & Control (JavaScript sender)
node test/scenario1_command_control.js

# Scenario 2: Large Arrow Table (JavaScript sender)
node test/scenario2_large_table.js

Troubleshooting

Common Issues

1. NATS Connection Failed

   • Ensure NATS server is running
   • Check NATS_URL configuration

2. HTTP Upload Failed

   • Ensure file server is running
   • Check FILESERVER_URL configuration
   • Verify upload permissions

3. Arrow IPC Deserialization Error

   • Ensure data is properly serialized to Arrow format
   • Check Arrow version compatibility

License

MIT