how to convert double with zeros in front to integer?

[BluePipsOnly]

I am trying to figure out how to convert a double that has zeros in front (eg, 0.0341 or 0.00020) to an integer  (eg, 0.0341 to 341 or 0.00020 to 20). All zeros before the first non-zero digit should be dropped and the first non-zero digit to the last digit should form an integer.

 

[Fernando Carreiro]

BluePipsOnly: I am trying to figure out how to convert a double that has zeros in front (eg, 0.0341 or 0.00020) to an integer  (eg, 0.0341 to 341 or 0.00020 to 20). All zeros before the first non-zero digit should be dropped and the first non-zero digit to the last digit should form an integer. 

Take some time to learn how floating point number work and are stored:

• IEEE-754 Floating Point Converter

• Double (IEEE754 Double precision 64-bit) Converter

To help understand (debugging purposes) the format of real number also look at the follow printing formats — PrintFormat

a	double	A real number in format [−]0xh.hhhh p±dd, where h.hhhh – mantissa in the form of hexadecimal digits, using "abcdef", dd - One or more digits of exponent. Number of decimal places is determined by the accuracy specification
A	double	A real number in format [−]0xh.hhhh P±dd, where h.hhhh – mantissa in the form of hexadecimal digits, using "ABCDEF", dd - One or more digits of exponent. Number of decimal places is determined by the accuracy specification

[William Roeder]

BluePipsOnly: (eg, 0.0341 to 341 or 0.00020 to 20).

Divide by _Point (0.00020 / 0.00001 = 20).

[amrali]

BluePipsOnly:

I am trying to figure out how to convert a double that has zeros in front (eg, 0.0341 or 0.00020) to an integer  (eg, 0.0341 to 341 or 0.00020 to 20). All zeros before the first non-zero digit should be dropped and the first non-zero digit to the last digit should form an integer.

 

The idea is to get number of decimal digits after the point, then multiply by 10^digits

long RemoveDecimalPoint(double number)
  {
   double pwr=1.0;
   while(MathRound(number*pwr)/pwr != number)
      pwr*=10;
   return long(number*pwr);
  }

void OnStart()
  {
   RemoveDecimalPoint(0.0341);   // 341
   RemoveDecimalPoint(0.00020);  // 2
   RemoveDecimalPoint(-0.0341);  // -341
  }

[William Roeder]

amrali #: then multiply by 10^digits

Same as dividing by _Point. _Point = 1 / 10^_digits.

[amrali]

William Roeder #:

Same as dividing by _Point. _Point = 1 / 10^_digits.

yes, only if his values are ATR.