Candles gives you the ability to create candlesticks from your tick data. The candlesticks will be created from live tick data and not historical tick data. For example, via websocket clients.
This project started as an fork of gdax-candles. During use, it turned out that there was a need to use it on other exchanges. So it changed to candles.
To install the most recent release from npm, run:
npm i candles --save
Here is a basic example using gdax as data provider:
const candles = require('candles');
const Gdax = require('gdax');
const product = ['ETH-USD', 'BTC-USD'];
const timeframe = ['30s', '1m', '2m', '5m', '15m', '30m', '1h', '3h', '6h', '12h', '24h'];
const options = {
timediff: {
enabled: true,
fixed: false,
value: 50,
samples: 1000
}
}
const Candlecollection = new candles(options);
Candlecollection.addProduct(product, timeframe);
const websocket = new Gdax.WebsocketClient(product);
websocket.on('message', data => {
if (data.time) {
Candlecollection.adjustClock(data.time);
}
if (data.type === 'match') {
Candlecollection.SetSeriesPrice(data.product_id, data.price, data.size);
}
});
Candlecollection.on('close', (candle, candles) => {
console.log(candle);
});
/*
Candlestick {
timestamp: 2018-11-08T15:20:00.000Z,
timeframe: '30s',
product: 'BTC-USD',
price: '6479.01000000',
open: '6479.01000000',
high: '6479.01000000',
low: '6479.00000000',
close: '6479.01000000',
volume: '0.39818029',
size: '0.01000000',
wick:
{ top: '0.00000000',
bottom: '0.01000000',
size: '0.01000000',
ratio: '1.00000000' },
closed: true,
market: 'bullish' }
*/
When a new instance is created, some options for fixing timelag between provider and your local system can be specified. By default lagtime correction is disabled.
const options = {
timediff: {
enabled: true,
fixed: false,
value: 50,
samples: 1000
}
}
const Candlecollection = new candles(options);
enabled
Enabling the correction of timelag.fixed
For a fixed lag between your provider and you local system set fixed to true. Otherwise the lagtime will be calculated by the given message timestamp, you have to pass.value
Only when fixed lag selected, the lag in ms has to be specified.samples
For the calculated lagtime an average of x samples will be calculated.
After creating an instance of candles, you have to add products and timeframes.
const product = ['ETH-USD', 'BTC-USD'];
const timeframe = ['30s', '1m', '2m', '5m', '15m', '30m', '1h', '3h', '6h', '12h', '24h'];
Candlecollection.addProduct(product, timeframe, serieslength);
product
You can set a name for your own, but it make sense to put the name of the currencypair in here. You have to tell candle this name, every time you update the current candle.timeframe
Setup the timeframe of your choice. The timeframes have to be writen in any number of s=seconds, m=minutes, h=hours and d=days. Feel free to use 13m or 7s as well as 21h. Product and Timeframe can both be arrays. Then for every product the given timeframe will be created. You can also seperate it.serieslength
[optional] If you like to collect a amount of last x candles, set the amount here. If not passed, only the last candle will be hold in memmory.
Candlecollection.addProduct('ETH-USD', '30s');
Candlecollection.addProduct('BTC-USD', '1m', 60);
Removing products is as easy as adding, just provide the product(s) and timeframe(s).
// removing these timeframes in these products
const product = ['ETH-USD', 'BTC-USD'];
const timeframe = ['30s', '1m', '2m', '5m', '15m', '30m', '1h', '3h', '6h', '12h', '24h'];
Candlecollection.removeProduct(product, timeframe);
product
You have to set the same name(s), you use for adding your product(s).timeframe
same as above.
// removing one timeframe in a product
Candlecollection.removeProduct('ETH-USD', '30s');
Candlecollection.removeProduct('BTC-USD', '1m');
Candles need to know when the price of the product has been changed.
const product = 'BTC-USD';
const price = 6479.01;
const size = 0.39818029;
Candlecollection.SetSeriesPrice(product, price, size)
product
The name you set in addProduct() before.price
The actual price you got from provider.size
This is an optional parameter. If your provider don't give you any volume information, just leave it. Then your candlesticks volume will be 0.
If you setup timelag correction, by enabling and setting samples count, you can pass your message timestamps to candles. Candles will build an average of these and your system clock, so the timestamp of the created candle will be more accorate.
Candlecollection.adjustClock(timestamp);
timestamp
Will be the timestamp from your providers message.
Gives you the average difference between exchange and local time. Depends on the timediff options you set. The value is in ms.
Candlecollection.getTimeDrift();
Events emitted from candles. For all events the current candle and the candle collection will be passed.
open
when a new candle createdclose
when a candle is closedcurrent
when the current candle where updated
These events can be more specific by adding the timeframe and/or the productname.
open BTC-USD
when a new candle of product 'BTC-USD' createdopen BTC-USD 1m
when a new candle of product 'BTC-USD' with timeframe '1m' createdopen 1m
when a new candle with timeframe '1m' created
This works also for close and current events to.
like:
Candlecollection.on('close BTC-USD 30s', (currentCandle, candles) => {
console.log(currentCandle);
});
The candles instance provides you following properties.
Candles collects all candlesticks for you in a series. You can get it by using the product and timeframe as property name.
Candlecollection.series['BTC-USD'].timeframe['30s'].candles
This is an overview of the candlestick properties.
timestamp
Time when this candle started. Opening time.timeframe
How much time this candle burned.price
The actual price.open
Opening price.high
Highest price in timeframe.low
Lowest price in timeframe.close
Closing price.volume
Volume traded in this timeframe.size
The Candlesize.wick.top
Top of the wick.wick.bottom
Bottom of the wick.wick.size
Size of the wick.wick.ratio
Ration body to wick.closed
Is this candle closed?market
Market situation of this candle.