Cloud Platform fully Implemented to run Trading Bots
This commit is contained in:
1 parent
b3bb241186
commit
93bc2ebed6
3 files changed
+52
-25
No files matched your search
@@ -0,0 +1,296 @@
|
||||
exports.newUserBot = function newUserBot(BOT, DEBUG_MODULE) {
|
||||
|
||||
let bot = BOT;
|
||||
|
||||
const MODULE_NAME = "This.Bot";
|
||||
const LOG_INFO = true;
|
||||
|
||||
logger.fileName = MODULE_NAME;
|
||||
logger.bot = bot;
|
||||
|
||||
thisObject = {
|
||||
initialize: initialize,
|
||||
start: start
|
||||
};
|
||||
|
||||
let platform; // You will receive a reference to the platform at your initialize function.
|
||||
|
||||
/*
|
||||
|
||||
The Platform object represents what the AA Platform provides you to help you with your bot. Inside it you can access to these inner objects:
|
||||
|
||||
platform = {
|
||||
context: context, // This one will allow you to get historical information of what your bot did on previous executions and learn about which is your current status every time your bot runs.
|
||||
datasource: datasource, // This one will provide you with pre-loaded data ready for you to consume. In this version candlesticks and stari patterns.
|
||||
assistant: assistant, // This one will help you to to create, and move positions at the exchange.
|
||||
processDatetime: processDatetime // This is the current oficial execution time that will be recorded in logs, and used to retrieve stored data.
|
||||
};
|
||||
|
||||
More details about these objects:
|
||||
|
||||
context = {
|
||||
statusReport: undefined, // Here is the information that defines which was the last sucessfull execution and some other details.
|
||||
executionHistory: [], // This is the record of bot execution.
|
||||
executionContext: undefined, // Here is the business information of the last execution of this bot process.
|
||||
};
|
||||
|
||||
datasource = {
|
||||
candlesFiles: new Map, // Complete sets of candles for different Time Periods. For Time Periods < 1hs sets are of current day only, otherwise whole market.
|
||||
candlesMap: new Map, // The last 10 candles for each Time Period will be stored here.
|
||||
stairsFiles: new Map, // Complete sets of patterns for different Time Periods. For Time Periods < 1hs sets are of current day only, otherwise whole market.
|
||||
stairsMap: new Map // The patterns we are currently in will be stored here.
|
||||
};
|
||||
|
||||
assistant = {
|
||||
putPositionAtExchange: putPositionAtExchange,
|
||||
movePositionAtExchange: movePositionAtExchange
|
||||
};
|
||||
|
||||
*/
|
||||
|
||||
return thisObject;
|
||||
|
||||
function initialize(pProcessDatetime, pPlatform, callBackFunction) {
|
||||
|
||||
try {
|
||||
|
||||
logger.fileName = MODULE_NAME;
|
||||
|
||||
/* Store local values. */
|
||||
platform = pPlatform;
|
||||
|
||||
logger.write("[INFO] initialize -> Entering function 'initialize' ");
|
||||
|
||||
callBackFunction(DEFAULT_OK_RESPONSE);
|
||||
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] initialize -> onDone -> err = " + err.message);
|
||||
callBackFunction(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
|
||||
function start(callBackFunction) {
|
||||
|
||||
try {
|
||||
|
||||
if (LOG_INFO === true) {
|
||||
logger.write("[INFO] Entering function 'start'");
|
||||
}
|
||||
|
||||
/*
|
||||
|
||||
This is an example. This bot will trade with a pseudo strategy based on candle and volumes stairs patterns.
|
||||
Essentially it will look at the patterns it is in at different time periods and try to make a guess if it is a good time to buy,
|
||||
sell, or do nothing.
|
||||
|
||||
*/
|
||||
|
||||
businessLogic(onDone);
|
||||
|
||||
function onDone(err) {
|
||||
try {
|
||||
|
||||
switch (err.result) {
|
||||
case DEFAULT_OK_RESPONSE.result: {
|
||||
callBackFunction(DEFAULT_OK_RESPONSE);
|
||||
return;
|
||||
}
|
||||
break;
|
||||
case DEFAULT_RETRY_RESPONSE.result: { // Something bad happened, but if we retry in a while it might go through the next time.
|
||||
logger.write("[ERROR] start -> onDone -> Retry Later. Requesting Execution Retry.");
|
||||
callBackFunction(err.message);
|
||||
return;
|
||||
}
|
||||
break;
|
||||
case DEFAULT_FAIL_RESPONSE.result: { // This is an unexpected exception that we do not know how to handle.
|
||||
logger.write("[ERROR] start -> onDone -> Operation Failed. Aborting the process.");
|
||||
callBackFunction(err.message);
|
||||
return;
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] start -> onDone -> err = " + err.message);
|
||||
callBackFunction(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
|
||||
function businessLogic(callBack) {
|
||||
|
||||
try {
|
||||
/*
|
||||
|
||||
First thing we need to know is to see where we are:
|
||||
|
||||
Do we have open positions?
|
||||
|
||||
If not, shall we create one?
|
||||
If yes, shall we move them?
|
||||
|
||||
As this is an example we can assume that we will have only one position, since we will be trading the whole allowed capital all at once.
|
||||
You dont need to do this, you can have as many positions as you wish, and you will find them all at the positions array used below.
|
||||
|
||||
*/
|
||||
|
||||
if (platform.context.executionContext.positions.length > 0) {
|
||||
|
||||
if (platform.context.executionContext.positions[0].type === "buy") {
|
||||
decideAboutBuyPosition(platform.context.executionContext.positions[0], callBack);
|
||||
} else {
|
||||
decideAboutSellPosition(platform.context.executionContext.positions[0], callBack);
|
||||
}
|
||||
|
||||
} else {
|
||||
|
||||
/*
|
||||
|
||||
Because this is an example, this bot is expected to always have an open position, either buy or sell.
|
||||
If it does not have one, that means that it is running for the first time. In which case, we will create one
|
||||
sell position at a very high price. Later, once the bot executes again, it will take it and move it to a reasonable
|
||||
place and monitor it during each execution round.
|
||||
|
||||
Lets see first which is the current price.
|
||||
|
||||
*/
|
||||
|
||||
let candleArray = platform.datasource.candlesMap.get("01-min"); // In this version of the platform, this array will contain the las 10 candles.
|
||||
let candle = candleArray[candleArray.length - 1]; // The last candle of the 10 candles array for the 1 min Time Period.
|
||||
|
||||
let currentRate = candle.close;
|
||||
|
||||
/*
|
||||
Now we verify that this candle is not too old. Lets say no more than 5 minutes old. This could happen if the datasets for
|
||||
any reason stops being updated.
|
||||
*/
|
||||
|
||||
if (candle.begin < platform.processDatetime.valueOf() - 5 * 60 * 1000) {
|
||||
|
||||
const logText = ;
|
||||
logger.write("[WARN] start -> businessLogic -> Last one min candle more than 5 minutes old. Bot cannot operate with this delay. Retrying later.");
|
||||
callBack('Retry Later');
|
||||
return;
|
||||
}
|
||||
|
||||
/*
|
||||
As we just want to create the first order now and we do not want this order to get executed, we will put it at
|
||||
the +50% of current exchange rate. Next Bot execution will move it strategically.
|
||||
*/
|
||||
|
||||
let rate = candle.close * 1.50;
|
||||
|
||||
/*
|
||||
The rules of the this first competition states that the bot will have the following initial balance in USDT and BTC to trade with.
|
||||
*/
|
||||
|
||||
const INITIAL_BALANCE_A = 0.0000;
|
||||
const INITIAL_BALANCE_B = 0.0001;
|
||||
|
||||
let AmountA = platform.context.statusReport.initialBalance.amountA;
|
||||
let AmountB = platform.context.statusReport.initialBalance.amountB;
|
||||
|
||||
/*
|
||||
Here is this bot example, we are going to sell all AmountB at once. You can do this or whatever you think is better.
|
||||
*/
|
||||
|
||||
AmountA = AmountB * rate;
|
||||
|
||||
platform.assistant.putPositionAtExchange("sell", rate, AmountA, AmountB, writeStatusAndContext, callBack);
|
||||
|
||||
}
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] start -> businessLogic -> err = " + err.message);
|
||||
callBack(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
|
||||
function decideAboutBuyPosition(pPosition, callBack) {
|
||||
|
||||
try {
|
||||
|
||||
/* For simplicity of this example bot, we will use here the same logic than when we are selling. */
|
||||
|
||||
decideAboutSellPosition(pPosition, callBack);
|
||||
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] start -> decideAboutBuyPosition -> err = " + err.message);
|
||||
callBack(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
|
||||
function decideAboutSellPosition(pPosition, callBack) {
|
||||
|
||||
try {
|
||||
/*
|
||||
|
||||
Here is where you decide what to do with your current sell position. Option are:
|
||||
|
||||
1. Do not touch it.
|
||||
2. Move it to another position by changing the rate.
|
||||
a. Up
|
||||
b. Down
|
||||
3. Cancell it. (not yet implemented at the platform.)
|
||||
|
||||
You can use here the information provided, analize it however you want and finally make a decition.
|
||||
|
||||
*/
|
||||
|
||||
let candleArray;
|
||||
let candle;
|
||||
let weight;
|
||||
|
||||
/*
|
||||
|
||||
Keeping in mind this is an example of traing bot, we are going to put some logic here that in the end will move the current position
|
||||
up or down. It will move it down if the bot feels it is time to sell, and up if it feels that selling is not a good idea.
|
||||
|
||||
To achieve a final rate to move the current position at the exchange, we are going to go through the available candles and patterns
|
||||
and each one is going to make a micro-move, and at the end we will have a final rate to send a move command to the exchange.
|
||||
|
||||
We will use a weight to give more or less importance to different Time Periods.
|
||||
|
||||
------
|
||||
NOTE: The code below is an example and you should replace it by your own logic. This is the key of your intervention here.
|
||||
------
|
||||
*/
|
||||
|
||||
let diff;
|
||||
let variationPercentage;
|
||||
let timePeriodName;
|
||||
|
||||
let targetRate = pPosition.rate;
|
||||
|
||||
let weightArray = [1 / (24 * 60), 1 / (12 * 60), 1 / (8 * 60), 1 / (6 * 60), 1 / (4 * 60), 1 / (3 * 60), 1 / (2 * 60), 1 / (1 * 60)];
|
||||
|
||||
for (i = 0; i < marketFilesPeriods.length; i++) {
|
||||
|
||||
weight = weightArray[i];
|
||||
|
||||
timePeriodName = marketFilesPeriods[i][1];
|
||||
|
||||
candleArray = platform.datasource.candlesMap.get(timePeriodName);
|
||||
candle = candleArray[candleArray.length - 1]; // The last candle of the 10 candles array.
|
||||
|
||||
diff = candle.close - candle.open;
|
||||
variationPercentage = diff * 100 / candle.open; // This is the % of how much the rate increased or decreced from open to close.
|
||||
|
||||
targetRate = targetRate + targetRate * variationPercentage / 100 * weight;
|
||||
|
||||
}
|
||||
|
||||
/* Finally we move the order position to where we have just estimated is a better place. */
|
||||
|
||||
platform.assistant.movePositionAtExchange(pPosition, targetRate, writeStatusAndContext, callBack);
|
||||
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] start -> decideAboutSellPosition -> err = " + err.message);
|
||||
callBack(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
|
||||
} catch (err) {
|
||||
logger.write("[ERROR] start -> err = " + err.message);
|
||||
callBackFunction(DEFAULT_FAIL_RESPONSE);
|
||||
}
|
||||
}
|
||||
};
|
||||
Reference in new issue
Block a user