e1361fe3e0
Fix copysign tests for 1.8.0
2022-04-29 18:37:16 -07:00
2fe191843c
Make sure test_copysignf uses Float
2022-04-29 18:37:16 -07:00
1b04b78740
Update float.rs
2022-04-29 18:10:11 -07:00
fdd43fad3d
Update float.rs
2022-04-29 18:10:11 -07:00
fac306842a
Update float.rs
2022-04-29 18:10:11 -07:00
37575361a6
Add copysign
2022-04-29 18:10:11 -07:00
05fd16d2a0
Forward euclid methods when possible
2022-04-29 17:53:50 -07:00
039fb91251
tweak euclid tests
2022-04-29 17:22:07 -07:00
d68870321f
take references as parameters
2022-04-29 17:22:07 -07:00
7ece7edabf
fixes codes structure etc.
2022-04-29 17:22:07 -07:00
a0a6154a75
Fixes epsilon
2022-04-29 17:22:07 -07:00
e4b037c2c0
fixes MIN
2022-04-29 17:22:07 -07:00
59a72a8fb9
Fixes methods
2022-04-29 17:22:07 -07:00
f09717c82b
fixes method impl
2022-04-29 17:22:07 -07:00
d4191d236c
Fixes use core::float
2022-04-29 17:22:07 -07:00
dde1bdda60
fixes example
2022-04-29 17:22:07 -07:00
4533ad7bb0
fixes ::Float
2022-04-29 17:22:07 -07:00
6f778fafdf
fixes float::Float and fmt
2022-04-29 17:22:07 -07:00
b3ba355daa
Add support for euclidean division and modulation
2022-04-29 17:22:07 -07:00
da62ee51e6
Skip check_numassignref_ops test for old rustc
2022-04-29 16:58:43 -07:00
527da0fa21
Highlight difference in traits' abstraction level
...
Short descriptions for NumOps sort of blended in with NumRef and RefNum.
This suggests to clarify the difference a little bit more.
2022-04-29 16:57:11 -07:00
ad307574de
add test: check_numassignref_ops
2022-04-29 16:57:11 -07:00
7220b0bcd3
impl Num Wrapping<T> can rely on NumOps instead
2022-04-29 16:57:11 -07:00
78aae4c700
Fix cfg typo
2022-01-21 02:23:32 +01:00
c24991f3f3
Reimplement is_sign_negative as a bitcast, libm not necessary.
2021-07-01 00:39:26 +03:00
622263ab9f
Implement is_sign_* and signum methods in terms of libm.
2021-06-24 19:40:54 +03:00
5bf2599fef
Merge branch 'master' into reverse-bits
2021-06-15 17:30:17 -07:00
89cf792ed1
Merge #214
...
214: Ignore case for float parsing of text special values r=cuviper a=tspiteri
This copies some of the standard library fixes of <https://github.com/rust-lang/rust/pull/78618 >.
Co-authored-by: Trevor Spiteri <tspiteri@ieee.org >
2021-06-16 00:18:05 +00:00
4154fb5af9
Ignore case for float parsing of text special values
...
This copies some of the standard library fixes of
<https://github.com/rust-lang/rust/pull/78618 >.
2021-04-15 20:15:46 +02:00
9b9d0ef5d7
PrimInt: add unit tests for reverse_bits
2021-04-13 22:30:09 +02:00
d989634fda
Add leading_ones and trailing_ones to PrimInt
2021-04-09 21:43:48 -04:00
c235ad63f0
Add LowerBounded/UpperBounded traits
2021-03-22 15:23:51 -04:00
54a371e00c
PrimInt: add fallback for reverse_bits on rustc<1.37
2021-03-02 14:41:28 +01:00
534a8170b6
Merge #196
...
196: Use libm in Float::{min,max} and in FloatCore r=cuviper a=cuviper
Co-authored-by: Josh Stone <cuviper@gmail.com >
2021-03-02 00:01:34 +00:00
235f67cd6f
PrimInt: add reverse_bits() method
2021-02-12 14:27:18 +01:00
3b7fd3af79
match standard library float parsing of "-NaN"
2021-02-05 19:18:57 +01:00
00b5a1d41e
use standard library for float from_str_radix with radix 10
2021-02-05 19:03:58 +01:00
c54841995a
Use libm in Float::{min,max} and in FloatCore
2020-11-13 18:26:06 -08:00
ba57faeacd
Update the names for overflowing results
2020-10-29 15:19:35 -07:00
e4723c3e39
Document relaxed requirements for Num::from_str_radix
2020-10-29 15:02:46 -07:00
9f672817d3
Merge #185 #186 #190
...
185: Trust the "i128" feature r=cuviper a=cuviper
If the "i128" feature is explicity requested, don't bother probing for
it. It will still cause a build error if that was set improperly.
186: Allow large f64-to-f32 to saturate to infinity r=cuviper a=cuviper
The implementation of `<f64 as ToPrimitive>::to_f32` was written at a
time when float-to-float overflow was though to be undefined behavior,
per rust-lang/rust#15536 , but this was later determined to be fine.
Casting a large `f64` to `f32` just results in an infinity with the
matching sign. The sign gives more information than if `to_f32` just
returns `None`, so now we let these infinities through as a result.
See also rust-num/num-bigint#163 and rust-num/num-rational#83 .
190: Normalize the comment style r=cuviper a=cuviper
Co-authored-by: Josh Stone <cuviper@gmail.com >
2020-10-29 20:52:48 +00:00
e9683c775d
Merge #180
...
180: Add Overflowing trait r=cuviper a=KaczuH
Closes issue #168
I've been inspired by `saturated.rs`
Co-authored-by: Kamil Niski <kamil.niski@gmail.com >
2020-10-29 20:46:52 +00:00
b2a6df9869
Correct docs for overflowing docs
2020-10-24 14:36:38 +02:00
3aeede4e29
Normalize the comment style
2020-10-06 14:46:22 -07:00
c2997d6c3d
Allow large f64-to-f32 to saturate to infinity
...
The implementation of `<f64 as ToPrimitive>::to_f32` was written at a
time when float-to-float overflow was though to be undefined behavior,
per rust-lang/rust#15536 , but this was later determined to be fine.
Casting a large `f64` to `f32` just results in an infinity with the
matching sign. The sign gives more information than if `to_f32` just
returns `None`, so now we let these infinities through as a result.
See also rust-num/num-bigint#163 and rust-num/num-rational#83 .
2020-08-27 17:17:32 -07:00
f3b54e0c95
remove example which was never UB from AsPrimitive docs
2020-08-21 04:59:29 +02:00
90a8de74ee
update AsPrimitive Safety docs for rust 1.45
2020-07-30 23:42:53 +02:00
7d66a73e7f
Remove std dep from integer MAX, MIN
2020-07-17 08:33:54 +02:00
98a050fcc0
Add Overflowing trait
...
Closes issue #168
2020-07-17 08:18:13 +02:00
2997cd8cda
#175 document cast traits about precision loss
...
- Document on `ToPrimitive`, `FromPrimitive` and `NumCast` that
scalars can be represented by the target type
when they are within the range admitted by the target type,
and so truncation/precision loss may happen
2020-07-05 23:07:07 +01:00
Josh Stone