Module lossless_float 

Available on crate feature lossless-float only.

A float that refuses to silently lose information (requires lossless-float feature) — the floating-point sibling of lossless-u64. A float that refuses to silently lose information.

LosslessFloat is the floating-point sibling of the lossless-u64 feature’s integer guarantee: an opt-in “refuse to silently lose precision” number type. Deserialization rejects infinities, NaN, and any value that does not survive a format-and-reparse round trip, so a config field declared LosslessFloat can never quietly hold a distorted number.

Formerly noyalib::robotics::StrictFloat — the mechanism was never robotics-specific, only the label; the robotics module carried deprecated aliases for one release (v0.0.29) and was removed in v0.0.30.

Examples

use noyalib::lossless_float::LosslessFloat;

let f: LosslessFloat = noyalib::from_str("3.14159").unwrap();
assert!((f.get() - 3.14159).abs() < 1e-10);

// Infinity is rejected.
let result: Result<LosslessFloat, _> = noyalib::from_str(".inf");
assert!(result.is_err());

Structs

LosslessFloat

A float that rejects values outside f64’s precise representation range.

LosslessFloatError

Error returned when a float value fails the precision check.