Function float_bits_magnitude_exceeds_2_24

src/fl/stl/json.cpp.hpp:56–61

Returns true iff `bits` represents a finite IEEE-754 single-precision value whose magnitude exceeds 2**24 (the float-integer-precision boundary). Used to decide whether an array of numbers can survive packing into `fl::vector ` without integer-precision loss. Integer-only -- operates on the bit pattern. Math: |value| > 2**24 iff biased_exp > 151 (exponent >= 25, m

// is taken -- matches the previous `canBeRepresentedAsFloat(double)` semantics
// before this code was bit-twiddled (FastLED #3022 phase 2).
static inline bool float_bits_magnitude_exceeds_2_24(u32 bits) FL_NOEXCEPT {
    const u32 biased_exp = (bits >> 23) & 0xFFu;
    const u32 mantissa = bits & 0x7FFFFFu;
    return biased_exp == 0xFFu || biased_exp > 151u ||
           (biased_exp == 151u && mantissa != 0u);
}


json_value& get_null_json_value() {