Merge branch 'master' into work

Author: dhardy

CHANGELOG.md

@@ -8,19 +8,19 @@ A [separate changelog is kept for rand_core](rand_core/CHANGELOG.md).
 
 You may also find the [Upgrade Guide](https://rust-random.github.io/book/update.html) useful.
 
-## [Unreleased]
+## [0.8.4] - 2021-06-15
 ### Additions
 - Use const-generics to support arrays of all sizes (#1104)
 - Implement `Clone` and `Copy` for `Alphanumeric` (#1126)
 - Add `Distribution::map` to derive a distribution using a closure (#1129)
+- Add `Slice` distribution (#1107)
+- Add `DistString` trait with impls for `Standard` and `Alphanumeric` (#1133)
 
 ### Other
 - Reorder asserts in `Uniform` float distributions for easier debugging of non-finite arguments
   (#1094, #1108)
 - Add range overflow check in `Uniform` float distributions (#1108)
-
-### Distributions
-- Add slice distribution (#1107)
+- Deprecate `rngs::adapter::ReadRng` (#1130)
 
 ## [0.8.3] - 2021-01-25
 ### Fixes
@@ -678,4 +678,4 @@ when updating from `rand 0.7.0` without also updating `rand_core`.
 
 ## [0.10-pre] - 2014-03-02
 ### Added
-- Seperate `rand` out of the standard library
+- Separate `rand` out of the standard library

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rand"
-version = "0.8.3"
+version = "0.8.4"
 authors = ["The Rand Project Developers", "The Rust Project Developers"]
 license = "MIT OR Apache-2.0"
 readme = "README.md"

README.md

@@ -143,7 +143,7 @@ unavailable.
 
 The WASM target `wasm32-unknown-unknown` is not *automatically* supported by
 `rand` or `getrandom`. To solve this, either use a different target such as
-`wasm32-wasi` or add a direct dependancy on `getrandom` with the `js` feature
+`wasm32-wasi` or add a direct dependency on `getrandom` with the `js` feature
 (if the target supports JavaScript). See
 [getrandom#WebAssembly support](https://docs.rs/getrandom/latest/getrandom/#webassembly-support).
 

SECURITY.md

@@ -8,7 +8,7 @@ No binding guarantees can be provided.
 ## Security premises
 
 Rand provides the trait `rand_core::CryptoRng` aka `rand::CryptoRng` as a marker
-trait. Generators implementating `RngCore` *and* `CryptoRng`, and given the
+trait. Generators implementing `RngCore` *and* `CryptoRng`, and given the
 additional constraints that:
 
 -   Instances of seedable RNGs (those implementing `SeedableRng`) are

rand_core/CHANGELOG.md

@@ -4,6 +4,11 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.6.3] - 2021-06-15
+### Changed
+- Improved bound for `serde` impls on `BlockRng` (#1130)
+- Minor doc additions (#1118)
+
 ## [0.6.2] - 2021-02-12
 ### Fixed
 - Fixed assertions in `le::read_u32_into` and `le::read_u64_into` which could

rand_core/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rand_core"
-version = "0.6.2"
+version = "0.6.3"
 authors = ["The Rand Project Developers", "The Rust Project Developers"]
 license = "MIT OR Apache-2.0"
 readme = "README.md"

rand_core/src/block.rs

@@ -95,7 +95,7 @@ pub trait BlockRngCore {
 /// [`fill_bytes`] / [`try_fill_bytes`] is called on a large array. These methods
 /// also handle the bookkeeping of when to generate a new batch of values.
 ///
-/// No whole generated `u32` values are thown away and all values are consumed
+/// No whole generated `u32` values are thrown away and all values are consumed
 /// in-order. [`next_u32`] simply takes the next available `u32` value.
 /// [`next_u64`] is implemented by combining two `u32` values, least
 /// significant first. [`fill_bytes`] and [`try_fill_bytes`] consume a whole

rand_distr/CHANGELOG.md

@@ -4,12 +4,14 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## Unreleased
+## [0.4.1] - 2021-06-15
+- Empirically test PDF of normal distribution (#1121)
 - Correctly document `no_std` support (#1100)
 - Add `std_math` feature to prefer `std` over `libm` for floating point math (#1100)
 - Add mean and std_dev accessors to Normal (#1114)
 - Make sure all distributions and their error types implement `Error`, `Display`, `Clone`,
  `Copy`, `PartialEq` and `Eq` as appropriate (#1126)
+- Port benchmarks to use Criterion crate (#1116)
 - Support serde for distributions (#1141)
 
 ## [0.4.0] - 2020-12-18

rand_distr/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rand_distr"
-version = "0.4.0"
+version = "0.4.1"
 authors = ["The Rand Project Developers"]
 license = "MIT OR Apache-2.0"
 readme = "README.md"

rand_distr/README.md

@@ -24,7 +24,7 @@ contrast, this `rand_distr` crate focuses on sampling from distributions.
 
 The floating point functions from `num_traits` and `libm` are used to support
 `no_std` environments and ensure reproducibility. If the floating point
-functions from `std` are prefered, which may provide better accuracy and
+functions from `std` are preferred, which may provide better accuracy and
 performance but may produce different random values, the `std_math` feature
 can be enabled.
 

rand_distr/src/binomial.rs

@@ -106,7 +106,7 @@ impl Distribution<u64> for Binomial {
         // random variate generation. Commun. ACM 31, 2 (February 1988),
         // 216-222. http://dx.doi.org/10.1145/42372.42381
 
-        // Threshold for prefering the BINV algorithm. The paper suggests 10,
+        // Threshold for preferring the BINV algorithm. The paper suggests 10,
         // Ranlib uses 30, and GSL uses 14.
         const BINV_THRESHOLD: f64 = 10.;
 
@@ -243,7 +243,7 @@ impl Distribution<u64> for Binomial {
                     }
                 }
 
-                // Step 5.2: Squeezing. Check the value of ln(v) againts upper and
+                // Step 5.2: Squeezing. Check the value of ln(v) against upper and
                 // lower bound of ln(f(y)).
                 let k = k as f64;
                 let rho = (k / npq) * ((k * (k / 3. + 0.625) + 1. / 6.) / npq + 0.5);

rand_distr/src/hypergeometric.rs

@@ -30,7 +30,7 @@ enum SamplingMethod {
 /// `f(k) = binomial(K, k) * binomial(N-K, n-k) / binomial(N, n)`,
 /// where `binomial(a, b) = a! / (b! * (a - b)!)`.
 /// 
-/// The [binomial distribution](crate::Binomial) is the analagous distribution
+/// The [binomial distribution](crate::Binomial) is the analogous distribution
 /// for sampling with replacement. It is a good approximation when the population
 /// size is much larger than the sample size.
 /// 

rand_distr/src/poisson.rs

@@ -110,7 +110,7 @@ where F: Float + FloatConst, Standard: Distribution<F>
                 loop {
                     // draw from the Cauchy distribution
                     comp_dev = rng.sample(cauchy);
-                    // shift the peak of the comparison ditribution
+                    // shift the peak of the comparison distribution
                     result = self.sqrt_2lambda * comp_dev + self.lambda;
                     // repeat the drawing until we are in the range of possible values
                     if result >= F::zero() {

rand_distr/src/utils.rs

@@ -91,14 +91,14 @@ where
         let i = bits as usize & 0xff;
 
         let u = if symmetric {
-            // Convert to a value in the range [2,4) and substract to get [-1,1)
+            // Convert to a value in the range [2,4) and subtract to get [-1,1)
             // We can't convert to an open range directly, that would require
-            // substracting `3.0 - EPSILON`, which is not representable.
+            // subtracting `3.0 - EPSILON`, which is not representable.
             // It is possible with an extra step, but an open range does not
-            // seem neccesary for the ziggurat algorithm anyway.
+            // seem necessary for the ziggurat algorithm anyway.
             (bits >> 12).into_float_with_exponent(1) - 3.0
         } else {
-            // Convert to a value in the range [1,2) and substract to get (0,1)
+            // Convert to a value in the range [1,2) and subtract to get (0,1)
             (bits >> 12).into_float_with_exponent(0) - (1.0 - core::f64::EPSILON / 2.0)
         };
         let x = u * x_tab[i];

rand_hc/CHANGELOG.md

@@ -4,6 +4,9 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.3.1] - 2021-06-15
+- Adjust crate links
+
 ## [0.3.0] - 2020-12-08
 - Bump `rand_core` version to 0.6.0
 - Bump MSRV to 1.36 (#1011)

rand_hc/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rand_hc"
-version = "0.3.0"
+version = "0.3.1"
 authors = ["The Rand Project Developers"]
 license = "MIT OR Apache-2.0"
 readme = "README.md"

rand_hc/src/hc128.rs

@@ -142,7 +142,7 @@ impl BlockRngCore for Hc128Core {
         let dd = (cc + 16) % 512;
         let ee = cc.wrapping_sub(16) % 512;
         // These asserts let the compiler optimize out the bounds checks.
-        // Some of them may be superflous, and that's fine:
+        // Some of them may be superfluous, and that's fine:
         // they'll be optimized out if that's the case.
         assert!(ee + 15 < 512);
         assert!(cc + 15 < 512);
@@ -238,7 +238,7 @@ impl Hc128Core {
         let dd = (cc + 16) % 512;
         let ee = cc.wrapping_sub(16) % 512;
         // These asserts let the compiler optimize out the bounds checks.
-        // Some of them may be superflous, and that's fine:
+        // Some of them may be superfluous, and that's fine:
         // they'll be optimized out if that's the case.
         assert!(ee + 15 < 512);
         assert!(cc + 15 < 512);

rand_pcg/CHANGELOG.md

@@ -4,6 +4,10 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/)
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.3.1] - 2021-06-15
+- Add `advance` methods to RNGs (#1111)
+- Document dependencies between streams (#1122)
+
 ## [0.3.0] - 2020-12-08
 - Bump `rand_core` version to 0.6.0
 - Bump MSRV to 1.36 (#1011)

rand_pcg/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "rand_pcg"
-version = "0.3.0"
+version = "0.3.1"
 authors = ["The Rand Project Developers"]
 license = "MIT OR Apache-2.0"
 readme = "README.md"

rand_pcg/src/pcg128.rs

@@ -93,7 +93,7 @@ impl Lcg128Xsl64 {
     #[inline]
     fn from_state_incr(state: u128, increment: u128) -> Self {
         let mut pcg = Lcg128Xsl64 { state, increment };
-        // Move away from inital value:
+        // Move away from initial value:
         pcg.state = pcg.state.wrapping_add(pcg.increment);
         pcg.step();
         pcg

rand_pcg/src/pcg64.rs

@@ -94,7 +94,7 @@ impl Lcg64Xsh32 {
     #[inline]
     fn from_state_incr(state: u64, increment: u64) -> Self {
         let mut pcg = Lcg64Xsh32 { state, increment };
-        // Move away from inital value:
+        // Move away from initial value:
         pcg.state = pcg.state.wrapping_add(pcg.increment);
         pcg.step();
         pcg

src/distributions/bernoulli.rs

@@ -49,7 +49,7 @@ pub struct Bernoulli {
 // `f64` only has 53 bits of precision, and the next largest value of `p` will
 // result in `2^64 - 2048`.
 //
-// Also there is a 100% theoretical concern: if someone consistenly wants to
+// Also there is a 100% theoretical concern: if someone consistently wants to
 // generate `true` using the Bernoulli distribution (i.e. by using a probability
 // of `1.0`), just using `u64::MAX` is not enough. On average it would return
 // false once every 2^64 iterations. Some people apparently care about this

src/distributions/float.rs

@@ -78,7 +78,7 @@ pub struct Open01;
 pub trait IntoFloat {
     type F;
 
-    /// Helper method to combine the fraction and a contant exponent into a
+    /// Helper method to combine the fraction and a constant exponent into a
     /// float.
     ///
     /// Only the least significant bits of `self` may be set, 23 for `f32` and

src/distributions/uniform.rs

@@ -158,7 +158,7 @@ use serde::{Serialize, Deserialize};
 /// println!("{}", sum);
 /// ```
 ///
-/// For a single sample, [`Rng::gen_range`] may be prefered:
+/// For a single sample, [`Rng::gen_range`] may be preferred:
 ///
 /// ```
 /// use rand::Rng;

src/rngs/thread.rs

@@ -33,7 +33,7 @@ use crate::{CryptoRng, Error, RngCore, SeedableRng};
 
 
 // Number of generated bytes after which to reseed `ThreadRng`.
-// According to benchmarks, reseeding has a noticable impact with thresholds
+// According to benchmarks, reseeding has a noticeable impact with thresholds
 // of 32 kB and less. We choose 64 kB to avoid significant overhead.
 const THREAD_RNG_RESEED_THRESHOLD: u64 = 1024 * 64;
 
@@ -59,7 +59,7 @@ const THREAD_RNG_RESEED_THRESHOLD: u64 = 1024 * 64;
 #[cfg_attr(doc_cfg, doc(cfg(all(feature = "std", feature = "std_rng"))))]
 #[derive(Clone, Debug)]
 pub struct ThreadRng {
-    // Rc is explictly !Send and !Sync
+    // Rc is explicitly !Send and !Sync
     rng: Rc<UnsafeCell<ReseedingRng<Core, OsRng>>>,
 }