acde249bf7
traits: Fix ToPrimitive for f64 -> f32 conversion.
...
It should use the destination type and not the source type to check if
the conversion would be to a value that's in range.
NOTE: A finite f64 value that is larger than the f32 value range now produces
None when converted to f32 with ToPrimitive.
Previously, too large f64 values would produce inf f32 values. This `as`
cast has an undefined result and was not specified to always produce for
example `inf`.
The conversion preserves nan/+-inf specifically.
2016-04-15 13:36:29 +02:00
1e192c31ae
num: add path-dependency versions
2016-04-14 00:45:03 -07:00
0dd410468c
rational: add path-dependency versions
2016-04-14 00:39:07 -07:00
164da50a99
iter: add path-dependency versions
2016-04-14 00:37:04 -07:00
dc733b1402
complex: add path-dependency versions
2016-04-14 00:29:19 -07:00
0b79edc108
bigint: add path-dependency versions
2016-04-14 00:26:54 -07:00
84ffb0ad01
integer: add path-dependency versions
2016-04-14 00:25:51 -07:00
2e8ce33d84
Bump all num crates to 0.1.32
2016-04-14 00:17:42 -07:00
774bf31ec6
Auto merge of #182 - ollie27:ratio_posneg, r=cuviper
...
Correct Ratio::is_negative and Ratio::is_positive
Zero is not positive or negative.
This was broken in 8be7e7bab5
2016-04-14 07:05:42 +09:00
c22e3bf9a2
Correct Ratio::is_negative and Ratio::is_positive
...
Zero is not positive or negative.
This was broken in 8be7e7bab5
2016-04-13 22:47:15 +01:00
095738e7de
Auto merge of #164 - rust-num:split-into-crates, r=cuviper
...
Move segments of library to separate crates
Issue #102
- [x] traits
- [x] bigint
- [x] integer
- [x] complex
- [x] iter
- [x] rational
2016-04-14 05:30:34 +09:00
58b5fe5883
Serializers dependencies
2016-04-11 20:43:07 +02:00
e59ead7b3a
Add Serde crate to `num_bigint`
2016-04-11 12:32:10 +02:00
8450782413
Revert old `num` crate description
2016-04-11 12:30:16 +02:00
3d11940538
Better descriptions for subcrates
2016-04-10 10:31:22 +02:00
a423b39833
.multirust.sh: use the subcrated "make test"
2016-03-25 18:30:27 -07:00
015cd0be43
.travis.yml: add a verbose build
...
I like to have a verbose build log for automation like Travis CI,
because it sometimes helps in diagnosing failures.
2016-03-25 18:30:27 -07:00
c9d82acf00
test_features.sh: re-enable as a simple build
2016-03-25 18:30:27 -07:00
03884fdbcc
bigint: reapply #176
2016-03-25 17:50:51 -07:00
aebbc4fd37
bigint: fix and un-ignore the first doctest
2016-03-25 17:50:51 -07:00
8845ee11ed
integer: reapply the rest of #167
2016-03-25 17:50:51 -07:00
0114559adf
Makefile: add complex and iter
2016-03-25 17:50:51 -07:00
8cb026e273
complex: update testing imports and hash
2016-03-25 17:50:51 -07:00
e672d006a4
iter: update testing imports
2016-03-25 17:50:51 -07:00
dac48aa4da
test_nightly.sh: update the macros path
2016-03-25 17:50:51 -07:00
9d439e7860
num: don't need std::hash even for testing
2016-03-25 17:50:51 -07:00
21a328ad6d
bigint, rational: use std::hash only for testing
2016-03-25 17:50:51 -07:00
529d7634dd
num: use original num_foo crate names for imports
...
Rust 1.0.0 can't seem to find `foo::bar` imports from renamed crates like
`pub use num_foo as foo`. Use `num_foo::bar` instead.
2016-03-25 17:50:38 -07:00
b73cfa57bb
traits: use `cast` items before `int`
...
For some reason, rustc 1.0.0 can't find `PrimInt` if it's before `cast`,
but later versions are fine with this. That may have been a compiler
bug that was fixed. Switching the order seems to work everywhere.
2016-03-25 16:12:56 -07:00
37325eec73
Remove unknown test flag
2016-03-25 12:53:34 +01:00
b962e2b708
Remove leftovers
2016-03-25 12:40:34 +01:00
956bb0f4db
Reapply ebed6756de
2016-03-25 12:34:48 +01:00
72fa7ece48
Reapply #167
2016-03-11 10:24:10 +01:00
7c0ab30bdc
Fix subpackages metadata
2016-03-11 10:22:26 +01:00
96e9166b0a
Extract iter
2016-03-11 01:06:37 +01:00
ed076070e6
Extract complex
2016-03-11 01:06:37 +01:00
2a67a5b86e
Extract num-rational
2016-03-11 01:06:37 +01:00
54685c46a1
Extract rational
2016-03-11 01:06:37 +01:00
2176b7048c
Extract bigint
2016-03-11 01:06:05 +01:00
f1a80857ee
Extract integer module
2016-03-11 01:06:05 +01:00
4361521f5a
Move num-macros to macros to fit new naming system
2016-03-11 01:06:05 +01:00
c124be549f
Move traits to separate crate
2016-03-11 01:06:05 +01:00
7eb666f6b8
Auto merge of #176 - murarth:bigint-bits, r=cuviper
...
Expose `BigUint::bits` through `BigInt`
2016-03-06 06:00:56 +09:00
df34563eda
Expose `BigUint::bits` through `BigInt`
2016-03-05 13:46:22 -07:00
9b0d4c91cb
Auto merge of #171 - erickt:serde, r=hauleth
...
Add support for Serde 0.7.
Serde 0.7 dropped it's dependency on num, so this patch moves the implementations here. For the sake of a better implementation, this just serializes BigUint as a `Vec<u32>`, `BigInt` as a `(u8, Vec<u32>)`, `Complex<T>` as a `(T, T)`, and `Ratio<T>` as a `(T, T)`.
2016-02-27 20:17:18 +09:00
112923eec5
Add support for Serde 0.7.
...
Serde 0.7 dropped it's dependency on num, so this patch moves
the implementations here. For the sake of a better implementation,
this just serializes BigUint as a `Vec<u32>`, `BigInt` as a
`(u8, Vec<u32>)`, `Complex<T>` as a `(T, T)`, and `Ratio<T>`
as a `(T, T)`.
2016-02-27 00:07:32 -08:00
9079b88b1c
Auto merge of #169 - cuviper:rational-overflow, r=hauleth
...
Avoid overflows in Ratio's Ord::cmp
Fixes #7
2016-02-25 20:29:50 +09:00
4e66bbe6a7
Avoid overflows in Ratio's Ord::cmp
...
Fixes #7
2016-02-22 18:10:29 -08:00
8be7e7bab5
Simplify some Ratio methods
2016-02-22 18:09:21 -08:00
5b2cb8df97
Simplify Integer dependant traits
...
- Integer only needs to require Ord explicitly, and then PartialOrd, Eq,
and PartialEq come transitively.
- Generics on Integer can implicitly use all of those comparison traits.
This should not be a breaking change, as it doesn't actually change any
effective trait requirements -- only what's explicit for simplicity.
2016-02-22 15:34:10 -08:00
bluss
Josh Stone
Homu
Oliver Middleton
Łukasz Jan Niemier
Murarth
Erick Tryzelaar