Precision for long double

[sukolenvo]

Hi, here is sample code to reproduce the problem I struggle with:

#include <fmt/core.h>

int main(int argc, char ** argv) {    
    fmt::print("{:.2f}\n",  0.099f);
    fmt::print("{:.2f}\n",  0.099L);
    return 0;
}

Output is :

0.10
0.1

https://godbolt.org/z/rxMb5nEGP

Is there a reason why precision 2 is working differently for long double compared to float?

PS Output also vary by compiler:
clang 14, gcc 12 output is 0.1
vs 2022 output is 0.10

[ilyapopov]

In MSVC, long double is the same as double. See https://learn.microsoft.com/en-us/cpp/cpp/fundamental-types-cpp?view=msvc-170

Microsoft-specific: The representation of long double and double is identical. However, long double and double are treated as distinct types by the compiler. The Microsoft C++ compiler uses the 4- and 8-byte IEEE-754 floating-point representations.

This may explain why MSVC behaves differently.

[vitaut]

Fixed in d424862. Thanks for reporting.