Skip to content
/ tradingview-websocket Public

A microservice for real-time TradingView OHLCV data streaming via WebSocket, with dynamic subscriptions, Prometheus metrics, and proxy support. Docker-ready.

39 stars 2 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

dovudo/tradingview-websocket

BranchesTags

Repository files navigation

TradingView Fetcher

A microservice for fetching real-time OHLCV data from TradingView and broadcasting it via a WebSocket API. Supports dynamic subscription management, resilience, and Prometheus metrics.

All code comments and documentation are in English.

Features

  • Real-time data from TradingView
  • WebSocket API for dynamic symbol subscription management
  • Broadcasts price updates via WebSocket
  • HTTP API integration for pushing data to external systems
  • Prometheus metrics monitoring
  • Configurable logging
  • Automatic reconnection on failure
  • Proxy support (optional)
  • Docker-ready for easy deployment
  • Detailed price logging for diagnostics (see below)

Installation

From Source

# Clone the repository
git clone <repository-url>
cd tv-fetcher

# Install dependencies
npm install

# Build the project
npm run build

# Start the service
npm start

Using Docker

# Build the image
docker build -t tv-fetcher .

# Run the container
docker run -p 8081:8081 -p 9100:9100 --env-file .env tv-fetcher

Using Docker Compose

docker-compose up -d

Configuration

Copy .env.example to .env and set your parameters:

cp .env.example .env

Configuration Parameters

Variable Description Default
TV_API_PROXY Proxy for TradingView API (optional) (empty)
TV_API_TIMEOUT_MS Timeout for TradingView API requests (ms) 10000
SUBSCRIPTIONS JSON array of initial subscriptions [{"symbol":"BINANCE:BTCUSDT","timeframe":"1"}]
BACKEND_ENDPOINT HTTP endpoint for pushing data (empty)
BACKEND_API_KEY API key for pushing data (empty)
WEBSOCKET_PORT WebSocket server port 8081
WEBSOCKET_ENABLED Enable WebSocket API true
METRICS_PORT Prometheus metrics port 9100
LOG_LEVEL Logging level (debug, info, warn, error) info
LOG_FILE Log file path ./logs/tv-fetcher.log
DEBUG_PRICES Enable detailed price logging (true/false) false
PRICES_LOG_FILE File to log all received price bars if DEBUG_PRICES is true ./logs/prices.log
HEALTH_CHECK_INTERVAL_MS How often to check for stale subscriptions (ms) 60000
HEALTH_STALE_THRESHOLD_MULTIPLIER Multiplier for stale detection (timeframe × multiplier) 3
HEALTH_AUTO_RECOVERY_ENABLED Enable automatic recovery of stale subscriptions true
HEALTH_MAX_RECOVERY_ATTEMPTS Maximum recovery attempts per subscription 3
HEALTH_FULL_RECONNECT_THRESHOLD Stale subscriptions count to trigger full reconnect 3
HEALTH_FULL_RECONNECT_COOLDOWN_MS Cooldown between full reconnects (ms) 600000

Detailed Price Logging

If you set DEBUG_PRICES=true in your .env, every price bar received from TradingView will be logged in detail to the file specified by PRICES_LOG_FILE (default: ./logs/prices.log).

Each log entry will include the symbol, timeframe, timestamp, open, high, low, close, and volume, e.g.:

2024-05-16T14:00:00.000Z [PRICE] BINANCE:BTCUSDT/1 2024-05-16T14:00:00.000Z O:65000.00 H:65100.00 L:64900.00 C:65050.00 V:12.34

This is useful for diagnostics and for verifying exactly what data is being received from TradingView, especially if you suspect issues with data delivery or backend integration.

Timeframes

TradingView API uses the following timeframe formats:

Human format TradingView API format
1 minute "1"
5 minutes "5"
15 minutes "15"
30 minutes "30"
1 hour "60"
4 hours "240"
1 day "D"
1 week "W"
1 month "M"

Health Monitoring System

This service includes a comprehensive health monitoring system for TradingView data flow, ensuring reliable data delivery under all conditions.

How It Works

  1. Data Flow Monitoring: The system tracks the timestamp of the last bar received for each subscription.
  2. Stale Detection: A subscription is considered "stale" if no data has been received for longer than expected (timeframe duration × multiplier).
  3. Auto-Recovery: When stale subscriptions are detected, the system automatically attempts recovery through targeted resubscription.
  4. Progressive Recovery: Multiple recovery strategies are employed based on the severity of the issue:
    • Individual Recovery: First attempts to unsubscribe and resubscribe to the affected symbol/timeframe.
    • Full Reconnection: If multiple subscriptions become stale, performs a complete TradingView reconnection.

Health Monitoring Configuration

The health monitoring system is configurable through the following environment variables:

  • HEALTH_CHECK_INTERVAL_MS: How often to check for stale subscriptions (default: 60000 ms)
  • HEALTH_STALE_THRESHOLD_MULTIPLIER: How many timeframe intervals to wait before considering a subscription stale (default: 3)
  • HEALTH_AUTO_RECOVERY_ENABLED: Enable/disable automatic recovery attempts (default: true)
  • HEALTH_MAX_RECOVERY_ATTEMPTS: Maximum recovery attempts per subscription before giving up (default: 3)
  • HEALTH_FULL_RECONNECT_THRESHOLD: Number of stale subscriptions that triggers a full reconnect (default: 3)
  • HEALTH_FULL_RECONNECT_COOLDOWN_MS: Minimum time between full reconnects (default: 600000 ms)

Health Monitoring Logs

The health monitoring system logs detailed information about its operation with the [HEALTH] prefix:

[HEALTH] Stale subscription detected for BINANCE:BTCUSDT/1 - no data for 3m 45s (threshold: 180s)
[HEALTH] Attempting recovery for BINANCE:BTCUSDT/1 (attempt 1/3)
[HEALTH] Successfully resubscribed to BINANCE:BTCUSDT/1

For severe issues, more aggressive recovery actions are logged:

[HEALTH] 4 stale subscriptions exceeds threshold (3), performing full reconnect
[HEALTH] Performing full TradingView reconnect due to multiple stale subscriptions
[HEALTH] Full TradingView reconnect successful

Health Monitoring Metrics

The following Prometheus metrics are exposed for monitoring the health of TradingView data flow:

  • stale_subscriptions: Gauge of currently stale subscriptions
  • recovery_attempts_total: Counter of recovery attempts
  • successful_recoveries_total: Counter of successful recovery attempts
  • failed_recoveries_total: Counter of failed recovery attempts
  • full_reconnects_total: Counter of full TradingView reconnections
  • last_data_received_seconds: Gauge of seconds since last data per subscription (labeled by symbol and timeframe)

These metrics can be used to set up alerts for persistent data flow issues.

WebSocket API

The service provides a WebSocket API for managing subscriptions and receiving real-time data.

Message Format

Requests (client → server)

{
  "action": "subscribe", // or unsubscribe, list, subscribe_many, unsubscribe_many
  "symbol": "BINANCE:BTCUSDT", // for subscribe/unsubscribe
  "timeframe": "1",           // for subscribe/unsubscribe
  "pairs": [                   // for subscribe_many/unsubscribe_many
    { "symbol": "BINANCE:BTCUSDT", "timeframe": "1" },
    { "symbol": "BINANCE:ETHUSDT", "timeframe": "5" }
  ],
  "requestId": "optional-string-id"
}

Responses (server → client)

{
  "type": "subscribe", // or unsubscribe, list, bar, error, info, subscribe_many, unsubscribe_many
  "success": true,
  "message": "Subscription created",
  "requestId": "optional-string-id",
  "symbol": "BINANCE:BTCUSDT",
  "timeframe": "1",
  "subscriptions": [ { "symbol": "BINANCE:BTCUSDT", "timeframe": "1" } ], // for list and bulk
  "bar": { /* ... */ }, // for type: bar
  "results": [ /* ... */ ] // for bulk operations
}

Example Requests

  • Subscribe to a single instrument:
{
  "action": "subscribe",
  "symbol": "BINANCE:BTCUSDT",
  "timeframe": "1",
  "requestId": "sub-1"
}
  • Unsubscribe:
{
  "action": "unsubscribe",
  "symbol": "BINANCE:BTCUSDT",
  "timeframe": "1",
  "requestId": "unsub-1"
}
  • Get subscription list:
{
  "action": "list",
  "requestId": "list-1"
}
  • Bulk subscribe:
{
  "action": "subscribe_many",
  "pairs": [
    { "symbol": "BINANCE:BTCUSDT", "timeframe": "1" },
    { "symbol": "BINANCE:ETHUSDT", "timeframe": "5" }
  ],
  "requestId": "bulk-sub-1"
}
  • Bulk unsubscribe:
{
  "action": "unsubscribe_many",
  "pairs": [
    { "symbol": "BINANCE:BTCUSDT", "timeframe": "1" },
    { "symbol": "BINANCE:ETHUSDT", "timeframe": "5" }
  ],
  "requestId": "bulk-unsub-1"
}

Example Responses

  • Successful subscription:
{
  "type": "subscribe",
  "success": true,
  "message": "Subscription created",
  "symbol": "BINANCE:BTCUSDT",
  "timeframe": "1",
  "requestId": "sub-1"
}
  • Bulk subscribe:
{
  "type": "subscribe_many",
  "success": true,
  "message": "Bulk subscribe processed",
  "subscriptions": [
    { "symbol": "BINANCE:BTCUSDT", "timeframe": "1" },
    { "symbol": "BINANCE:ETHUSDT", "timeframe": "5" }
  ],
  "results": [
    { "symbol": "BINANCE:BTCUSDT", "timeframe": "1", "success": true, "message": "Subscription created" },
    { "symbol": "BINANCE:ETHUSDT", "timeframe": "5", "success": true, "message": "Subscription created" }
  ],
  "requestId": "bulk-sub-1"
}
  • Error:
{
  "type": "error",
  "success": false,
  "message": "Symbol and timeframe are required for subscription",
  "requestId": "sub-err-1"
}

About

A microservice for real-time TradingView OHLCV data streaming via WebSocket, with dynamic subscriptions, Prometheus metrics, and proxy support. Docker-ready.

Resources

Readme
Activity

Stars

39 stars

Watchers

1 watching

Forks

2 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages