Skip to content

Chart TCP API

Overview

The Chart section describes how historical candlestick (OHLC) data is managed via the TCP API of the trading server.

The Chart API provides a complete set of operations for managing price history:

  • Retrieve historical candles
  • Import candle history
  • Update a single candle
  • Delete a candle
  • Fix spike candles
  • Synchronize chart history

All operations are handled internally by the QuoteManager module and use the unified CandleStore structure.

Candle Data Model

All candlestick data is transferred in a compact array format:

[open, high, low, close, volume, time]

Standard Response Structure

{
  "structure": ["open", "high", "low", "close", "volume", "time"],
  "data": [
    [1.1234, 1.1250, 1.1200, 1.1240, 1500, 1700000000]
  ]
}

Field Description

Field Type Description

open double Opening price high double Highest price during period low double Lowest price during period close double Closing price volume double Trading volume time int64 Unix timestamp

Validation Rules

Most methods require:

  • symbol --- required parameter
  • from, to --- time range (Unix time)
  • properly formatted candles array
  • strict OHLC structure compliance

If validation fails, the API returns:

{
  "error": "INVALID_DATA",
  "message": "validation error description"
}

Navigation

Method Name Description
Get Candles Data Retrieve historical OHLC candles within a specified time range
Import Candles Data Import historical candle data with optional full history replacement (flushData)
Update Candle Data Update a specific existing candle in history
Delete Candle Data Delete a specific candle from historical data
Fix Spike Candle Automatically detect and correct abnormal spike candles
Sync History Chart Data Synchronize historical chart data between data sources or servers

Data Flow

flowchart LR
    Client --> TCP_API
    TCP_API --> QuoteManager
    QuoteManager --> HistoryStorage

Performance Recommendations

  • Limit the from/to time range when possible
  • Import data in batches
  • Use flushData carefully in production environments