Source code
Revision control
Copy as Markdown
Other Tools
# Release 0.2.15 (2022-05-02)
- [The new `Euclid` trait calculates Euclidean division][195], where the
remainder is always positive or zero.
- [The new `LowerBounded` and `UpperBounded` traits][210] separately describe
types with lower and upper bounds. These traits are automatically implemented
for all fully-`Bounded` types.
- [The new `Float::copysign` method copies the sign of the argument][207] to
to the magnitude of `self`.
- [The new `PrimInt::leading_ones` and `trailing_ones` methods][205] are the
complement of the existing methods that count zero bits.
- [The new `PrimInt::reverse_bits` method reverses the order of all bits][202]
of a primitive integer.
- [Improved `Num::from_str_radix` for floats][201], also [ignoring case][214].
- [`Float` and `FloatCore` use more from `libm`][196] when that is enabled.
**Contributors**: @alion02, @clarfonthey, @cuviper, @ElectronicRU,
@ibraheemdev, @SparrowLii, @sshilovsky, @tspiteri, @XAMPPRocky, @Xiretza
# Release 0.2.14 (2020-10-29)
- Clarify the license specification as "MIT OR Apache-2.0".
**Contributors**: @cuviper
# Release 0.2.13 (2020-10-29)
- [The new `OverflowingAdd`, `OverflowingSub`, and `OverflowingMul` traits][180]
return a tuple with the operation result and a `bool` indicating overflow.
- [The "i128" feature now overrides compiler probes for that support][185].
This may fix scenarios where `autocfg` probing doesn't work properly.
- [Casts from large `f64` values to `f32` now saturate to infinity][186]. They
previously returned `None` because that was once thought to be undefined
behavior, but [rust#15536] resolved that such casts are fine.
- [`Num::from_str_radix` documents requirements for radix support][192], which
are now more relaxed than previously implied. It is suggested to accept at
least `2..=36` without panicking, but `Err` may be returned otherwise.
**Contributors**: @cuviper, @Enet4, @KaczuH, @martin-t, @newpavlov
# Release 0.2.12 (2020-06-11)
- [The new `WrappingNeg` trait][153] will wrap the result if it exceeds the
boundary of the type, e.g. `i32::MIN.wrapping_neg() == i32::MIN`.
- [The new `SaturatingAdd`, `SaturatingSub`, and `SaturatingMul` traits][165]
will saturate at the numeric bounds if the operation would overflow. These
soft-deprecate the existing `Saturating` trait that only has addition and
subtraction methods.
- [Added new constants for logarithms, `FloatConst::{LOG10_2, LOG2_10}`][171].
**Contributors**: @cuviper, @ocstl, @trepetti, @vallentin
# Release 0.2.11 (2020-01-09)
- [Added the full circle constant τ as `FloatConst::TAU`][145].
- [Updated the `autocfg` build dependency to 1.0][148].
**Contributors**: @cuviper, @m-ou-se
# Release 0.2.10 (2019-11-22)
- [Updated the `libm` dependency to 0.2][144].
**Contributors**: @CryZe
# Release 0.2.9 (2019-11-12)
- [A new optional `libm` dependency][99] enables the `Float` and `Real` traits
in `no_std` builds.
- [The new `clamp_min` and `clamp_max`][122] limit minimum and maximum values
while preserving input `NAN`s.
- [Fixed a panic in floating point `from_str_radix` on invalid signs][126].
- Miscellaneous documentation updates.
**Contributors**: @cuviper, @dingelish, @HeroicKatora, @jturner314, @ocstl,
@Shnatsel, @termoshtt, @waywardmonkeys, @yoanlcq
# Release 0.2.8 (2019-05-21)
- [Fixed feature detection on `no_std` targets][116].
**Contributors**: @cuviper
# Release 0.2.7 (2019-05-20)
- [Documented when `CheckedShl` and `CheckedShr` return `None`][90].
- [The new `Zero::set_zero` and `One::set_one`][104] will set values to their
identities in place, possibly optimized better than direct assignment.
- [Documented general features and intentions of `PrimInt`][108].
**Contributors**: @cuviper, @dvdhrm, @ignatenkobrain, @lcnr, @samueltardieu
# Release 0.2.6 (2018-09-13)
- [Documented that `pow(0, 0)` returns `1`][79]. Mathematically, this is not
strictly defined, but the current behavior is a pragmatic choice that has
precedent in Rust `core` for the primitives and in many other languages.
- [The new `WrappingShl` and `WrappingShr` traits][81] will wrap the shift count
if it exceeds the bit size of the type.
**Contributors**: @cuviper, @edmccard, @meltinglava
# Release 0.2.5 (2018-06-20)
- [Documentation for `mul_add` now clarifies that it's not always faster.][70]
- [The default methods in `FromPrimitive` and `ToPrimitive` are more robust.][73]
**Contributors**: @cuviper, @frewsxcv
# Release 0.2.4 (2018-05-11)
- [Support for 128-bit integers is now automatically detected and enabled.][69]
Setting the `i128` crate feature now causes the build script to panic if such
support is not detected.
**Contributors**: @cuviper
# Release 0.2.3 (2018-05-10)
- [The new `CheckedNeg` and `CheckedRem` traits][63] perform checked `Neg` and
`Rem`, returning `Some(output)` or `None` on overflow.
- [The `no_std` implementation of `FloatCore::to_degrees` for `f32`][61] now
uses a constant for greater accuracy, mirroring [rust#47919]. (With `std` it
just calls the inherent `f32::to_degrees` in the standard library.)
- [The new `MulAdd` and `MulAddAssign` traits][59] perform a fused multiply-
add. For integer types this is just a convenience, but for floating point
types this produces a more accurate result than the separate operations.
- [All applicable traits are now implemented for 128-bit integers][60] starting
with Rust 1.26, enabled by the new `i128` crate feature. The `FromPrimitive`
and `ToPrimitive` traits now also have corresponding 128-bit methods, which
default to converting via 64-bit integers for compatibility.
**Contributors**: @cuviper, @LEXUGE, @regexident, @vks
# Release 0.2.2 (2018-03-18)
- [Casting from floating point to integers now returns `None` on overflow][52],
avoiding [rustc's undefined behavior][rust-10184]. This applies to the `cast`
function and the traits `NumCast`, `FromPrimitive`, and `ToPrimitive`.
**Contributors**: @apopiak, @cuviper, @dbarella
# Release 0.2.1 (2018-03-01)
- [The new `FloatCore` trait][32] offers a subset of `Float` for `#![no_std]` use.
[This includes everything][41] except the transcendental functions and FMA.
- [The new `Inv` trait][37] returns the multiplicative inverse, or reciprocal.
- [The new `Pow` trait][37] performs exponentiation, much like the existing `pow`
function, but with generic exponent types.
- [The new `One::is_one` method][39] tests if a value equals 1. Implementers
should override this method if there's a more efficient way to check for 1,
rather than comparing with a temporary `one()`.
**Contributors**: @clarcharr, @cuviper, @vks
# Release 0.2.0 (2018-02-06)
- **breaking change**: [There is now a `std` feature][30], enabled by default, along
with the implication that building *without* this feature makes this a
`#![no_std]` crate.
- The `Float` and `Real` traits are only available when `std` is enabled.
- Otherwise, the API is unchanged, and num-traits 0.1.43 now re-exports its
items from num-traits 0.2 for compatibility (the [semver-trick]).
**Contributors**: @cuviper, @termoshtt, @vks
# Release 0.1.43 (2018-02-06)
- All items are now [re-exported from num-traits 0.2][31] for compatibility.
# Release 0.1.42 (2018-01-22)
- [num-traits now has its own source repository][num-356] at [rust-num/num-traits][home].
- [`ParseFloatError` now implements `Display`][22].
- [The new `AsPrimitive` trait][17] implements generic casting with the `as` operator.
- [The new `CheckedShl` and `CheckedShr` traits][21] implement generic
support for the `checked_shl` and `checked_shr` methods on primitive integers.
- [The new `Real` trait][23] offers a subset of `Float` functionality that may be applicable to more
types, with a blanket implementation for all existing `T: Float` types.
Thanks to @cuviper, @Enet4, @fabianschuiki, @svartalf, and @yoanlcq for their contributions!
# Prior releases
No prior release notes were kept. Thanks all the same to the many
contributors that have made this crate what it is!