Crates.io | js_int |
lib.rs | js_int |
version | 0.2.2 |
source | src |
created_at | 2019-06-16 15:45:25.310386 |
updated_at | 2022-03-23 11:28:04.02811 |
description | JavaScript-interoperable integer types |
homepage | |
repository | https://github.com/ruma/js_int |
max_upload_size | |
id | 141566 |
size | 63,391 |
Crate js_int
provides JavaScript-interoperable integer types.
JavaScript does not have native integers. Instead it represents all numeric
values with a single Number
type which is represented as an
IEEE 754 floating-point value.*
Rust's i64
and u64
types can contain values outside the range of what can be
represented in a JavaScript Number
.
This crate provides the types Int
and UInt
which wrap i64
and u64
,
respectively. These types add bounds checking to ensure the contained value is
within the range representable by a JavaScript Number
. They provide useful
trait implementations to easily convert from Rust's primitive integer types.
* In the upcoming ECMAScript 2020, JavaScript will probably gain support
for integers. There is a proposal for a BigInt
type type that is not
far from becoming part of the JavaScript specification. It won't make this crate
obsolete in any way though, since there will still be lots of JS code using
Number
, and code in other languages that assumes its use.
This crate requires rustc >= 1.35.
This crate is no_std
-compatible with default-features = false
. This will
disable the std
feature, which at the time of writing will only omit the
implementations of std::error::Error
for ParseIntError
and
TryFromIntError
.
(De-)Serialization via serde
is supported via the serde
feature, even
without the std
feature.
Deserialization can be routed through f64
instead of u64
with the
float_deserialize
feature. This will still not deserialize numbers with a
non-zero fractional component. Enabling the lax_deserialize
feature will discard the
fractional part instead of declining to deserialize. Please be aware that
serde_json
doesn't losslessly parse large floats with a fractional part by
default (even if the fractional part is .0
). To fix that, enable its
float_roundtrip
feature.