If www.tickstory.com is not good enough for free sign up for www.esignal.com classic delayed for $50 and use www.mechtrading.com qcollector to get the complete intraday and tick data history out. Then cancel if you do not want the latest data appended to your csv files.
If that's the case don't bother with free data. Don't bother with eSignal. Buy high resolution tic data directly from the CME. It's clean and it's accurate. If you're going to put all of that time and effort into your project - it's insane not to spend a very modest amount of money on data that you know is good. Many professionals use Bloomberg data, but they get that because they pay $1900 per month for a terminal. https://www.cmegroup.com/market-data.html
this sorts an array at speed O(log(n!)) = O(nlog(n) ) which is a lower bound just for the record i saw this first. while(i<array_size) { j=i; while(j<array_size){ Y+=(X>=X[j]); Y[j]+=(X[j]>X); j+=1;} X_sorted[Y]=X; i+=1; }
correction. the Y array is the position of element X, X_sorted[] is the ordered array. Do both then you only have to start check from j =i to the end each time hence the n! you get the idea though its fast. while(i<array_size) { j=i; while(j<array_size) { Y+=(X>=X[j]); Y[j]+=(X[j]>X); j+=1; } X_sorted[Y]=X; i+=1; }
I don't if you are willing to pay, that's something that you should have elaborated more. You can try IEX.com, they offer free intraday data though in some time stamps their data can be very unclean just to warn you. if you wanna pay, the following vendors have a quality data, try tickdata.com, algoseek.com, polygon.io, quandl.com, quandl used to offer free API but they no longer do that. The famous bloomberg they offer free data sets, you'll get from them allthe data you ever need BUT you are limited in the amount of intraday data you can export from your Bloomberg and upto about 250 days historically.