Tuesday, August 5, 2025

Monitoring a number of crypto exchanges on-line through MQL5 library for Node.js/CCXT – Different – 5 July 2025


CcxtAppServerLib is meant for working with all top-100 crypto exchanges from MQL5 through Node.js and CCXT Utility Server, and it’s evolving step-by-step. Newest beta-version contains optimized caching of alternate “markets” (instrument specs and different buying and selling situations), non-blocking studying of websocket messages, and extra.
One introductory instance of utilizing the library – the script CcxtAppSrvShowcase.mq5 – was posted within the weblog earlier.
As one other instance of customized software improvement utilizing the library, we offer a script for parallel monitoring of a number of exchanges for chosen sort of information. This function is applied by subscribing via CCXT on Node.js to particular websocket “watches”, similar to watchOrderBook, watchTicker, watchTrades, and so forth. For particulars, please, see widespread CCXT documentation and extra superior on CCXT PRO. Because the script is meant to make use of solely public APIs, consumer credentials aren’t concerned right here, however the library helps them in full diploma.

Right here is an important components of the brand new demo-script MultiExchangeWatch.mq5 (it is supplied together with the beta-version of the library).

NB: If the script is operating very first time, it can ask to unpack (manually) CCXT Utility Server (extracted as ccxtappsrvbundle.jsc from built-in useful resource), and run Node.js with it.

First, embrace the headers.

#embrace "ccxtjsmtlib.mqh" 
#embrace "ccxtutil.mqh"
#embrace 

Within the inputs, the Node server setup ought to be achieved.

enter group "Connection settings"
enter string NodeServer = "http://127.0.0.1:8124";
enter string NodeAuth = ""; 

Subsequent, specify various exchanges you wish to monitor, a ticker, and a sort of the watch. By default, the script watches for order books for BCH/USDT.

To fill in these inputs correctly with most well-liked values, you need to in all probability have to output the checklist of supported exchanges and their markets beforehand. It may be considered in one other instance script CcxtAppSrvShowcase.mq5, supplied with the lib.

enter string Exchanges = "ascendex,bitmart,binance"; 
enter string Ticker = "BCH/USDT";
enter string Watch = "watchOrderBook";
enter uint WatchingDuration = 10; 

Then OnStart occasion handler does its job. Inline feedback clarify the method. The imported features, courses and strategies from the library are highlighted in yellow.

string Trade[];

void OnStart()
{
   
   
   
   PrintFormat("CCXT AppSrvLibrary model: %.2f", AppSrvLibraryVersion());
   const static string standing[] = {"Cannot deploy",
      "App server ZIP is deployed, however not extracted",
      "App server recordsdata are deployed"};
   const int d = DeployCcxtAppServer();
   Print(standing[d + 1]);
   if(d <= 0)
   {
      return; 
   }
   
   
   
   
   SetNodeServer(NodeServer, NodeAuth);

   CcxtLink *hyperlink = GetLink();
   
   
   

   
   
   AutoPtr<CcxtJsExchangeProIntf> ccxt[];
   const int n = StringSplit(Exchanges, ',', Trade);
   ArrayResize(ccxt, n);
   
   for(int i = 0; i < n; i++)
   {
      ccxt[i] = CreateExchangePro(Trade[i]); 

      if(hyperlink.getLastHttpCode() != 200 || !ccxt[i][] || ccxt[i][][].t >= JS_NULL)
      {
         Print("Development failed for alternate: ", Trade[i]);
         return;
      }
      
      const bool isPro = !!*ccxt[i][]["pro"]; 
      
      if(!isPro)
      {
         PrintFormat("WARNING! %s is not PRO, there isn't any websocket help", Trade[i]);
      }
      
      if(!ccxt[i][]["has"][Watch].get<bool>())
      {
         PrintFormat("WARNING! %s doesn't help '%s' subscriptions", Trade[i], Watch);
      }
   }
   
   
   
   int lively = 0;
   
   for(int i = 0; i < n; i++)
   {
      
      if(ccxt[i][].improve())
      {
         if(!ccxt[i][].watchAnything(StringFormat("%s("%s")", Watch, Ticker)))
         {
            PrintFormat("Cannot begin %s for %s", Watch, Trade[i]);
            ccxt[i][].shut(); 
            ccxt[i] = NULL;
         }
         else
         {
            lively++;
         }
      }
      else
      {
         if(ccxt[i][].isConnected())
         {
            Print("Cannot improve to websockets");
            string headers[][2];
            if(ccxt[i][].ws().getHeaders(headers))
            {
               
            }
            ccxt[i][].ws().shut(); 
            ccxt[i][].shut();
            ccxt[i] = NULL;
         }
      }
   }
   
   if(!lively) return;

   
   PrintFormat("* Monitoring %d subscriptions", lively);

   const uint begin = GetTickCount();
   whereas(!IsStopped() && (!WatchingDuration || GetTickCount() - begin < WatchingDuration * 1000))
   {
      for(int i = 0; i < n; i++)
      {
         if(ccxt[i][] && ccxt[i][].isConnected())
         {
            AutoPtr j = ccxt[i][].readMessage(false); 
            if(j[])
            {
               ChronoComment(j[].stringify(0, 0)); 
            }
         }
      }
   }
  
   Print("* Unsubscribing...");
   for(int i = 0; i < n; i++)
   {
      if(ccxt[i][] && ccxt[i][].isConnected())
      {
         ccxt[i][].un().watchAnything(StringFormat("%s("%s")", Watch, Ticker));
      }
   }
   
   
   GracefullClose(ccxt); // not offered right here within the weblog

   
   
   GracefullClose(ccxt, 5, true);
   
   Remark("");
}

When the script is operating, the checklist of incoming order books (json-messages) is outputted and actively up to date on the chart.

On prime of such a dataflow it is simple to implement numerous arbitrage methods and calculate mixed statistics.

Related Articles

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Latest Articles