How to set DecimalSeparator for a float in C++ Builder - c++builder

Question

I'm using C++ Builder and my locale is European, so I have a comma as a decimal separator.
I need to convert a double to a decimal point value with a DOT as separator.
I can't find an answer anywhere.

Answer 1

DecimalSeparator is a global variable, simply set it to the desired character before formatting the double, eg:
#include <SysUtils.hpp>
System::String FormatWithDot(double value)
{
System::Char old = Sysutils::DecimalSeparator;
Sysutils::DecimalSeparator = _D('.');
System::String s = Sysutils::FloatToStr(value);
Sysutils::DecimalSeparator = old;
return s;
}
System::String s = FormatWithDot(123.45);
Or, if you need to do this in multiple threads, use the thread-safe version, TFormatSettings::DecimalSeparator:
#include <SysUtils.hpp>
System::String FormatWithDot(double value)
{
Sysutils::TFormatSettings fmt = Sysutils::TFormatSettings::Create();
fmt.DecimalSeparator = _D('.');
return Sysutils::FloatToStr(value, fmt);
}
System::String s = FormatWithDot(123.45);
Just note that DecimalSeparator only applies to Delphi-based RTL functions like FloatToStr(), Format(), etc. It does not apply to C++-based functions like std::(s)printf(), std::to_string(), std::ostream::operator<<, etc. For those, you need to use C++ locales instead.

Answer 2

I solved it in the wee hours of the morning myself. You can use:
TFormatSettings fmt = TFormatSettings::Create();
fmt.DecimalSeparator = '.';
and then format a double like that:
FloatToStr(price, fmt);
I hope it helps someone. I was going crazy.

Answer 3

I know this this is a very old thread, but I was facing this problem today. In my case, my solution is this:
String DotFormatted(long double value, int numberOfDecimals=2) {
String result = Format("%." + IntToStr(numberOfDecimals) + "f", ARRAYOFCONST((value)));
result = StringReplace( result, ",", ".", TReplaceFlags() << rfReplaceAll);
return result;
}
It's equivalent to another one I had in a different language. For me, it's useful because of the numberOfDecimals argument.