/rust/registry/src/index.crates.io-1949cf8c6b5b557f/rand-0.8.6/src/distributions/distribution.rs
Line | Count | Source |
1 | | // Copyright 2018 Developers of the Rand project. |
2 | | // Copyright 2013-2017 The Rust Project Developers. |
3 | | // |
4 | | // Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or |
5 | | // https://www.apache.org/licenses/LICENSE-2.0> or the MIT license |
6 | | // <LICENSE-MIT or https://opensource.org/licenses/MIT>, at your |
7 | | // option. This file may not be copied, modified, or distributed |
8 | | // except according to those terms. |
9 | | |
10 | | //! Distribution trait and associates |
11 | | |
12 | | use crate::Rng; |
13 | | use core::iter; |
14 | | #[cfg(feature = "alloc")] |
15 | | use alloc::string::String; |
16 | | |
17 | | /// Types (distributions) that can be used to create a random instance of `T`. |
18 | | /// |
19 | | /// It is possible to sample from a distribution through both the |
20 | | /// `Distribution` and [`Rng`] traits, via `distr.sample(&mut rng)` and |
21 | | /// `rng.sample(distr)`. They also both offer the [`sample_iter`] method, which |
22 | | /// produces an iterator that samples from the distribution. |
23 | | /// |
24 | | /// All implementations are expected to be immutable; this has the significant |
25 | | /// advantage of not needing to consider thread safety, and for most |
26 | | /// distributions efficient state-less sampling algorithms are available. |
27 | | /// |
28 | | /// Implementations are typically expected to be portable with reproducible |
29 | | /// results when used with a PRNG with fixed seed; see the |
30 | | /// [portability chapter](https://rust-random.github.io/book/portability.html) |
31 | | /// of The Rust Rand Book. In some cases this does not apply, e.g. the `usize` |
32 | | /// type requires different sampling on 32-bit and 64-bit machines. |
33 | | /// |
34 | | /// [`sample_iter`]: Distribution::sample_iter |
35 | | pub trait Distribution<T> { |
36 | | /// Generate a random value of `T`, using `rng` as the source of randomness. |
37 | | fn sample<R: Rng + ?Sized>(&self, rng: &mut R) -> T; |
38 | | |
39 | | /// Create an iterator that generates random values of `T`, using `rng` as |
40 | | /// the source of randomness. |
41 | | /// |
42 | | /// Note that this function takes `self` by value. This works since |
43 | | /// `Distribution<T>` is impl'd for `&D` where `D: Distribution<T>`, |
44 | | /// however borrowing is not automatic hence `distr.sample_iter(...)` may |
45 | | /// need to be replaced with `(&distr).sample_iter(...)` to borrow or |
46 | | /// `(&*distr).sample_iter(...)` to reborrow an existing reference. |
47 | | /// |
48 | | /// # Example |
49 | | /// |
50 | | /// ``` |
51 | | /// use rand::thread_rng; |
52 | | /// use rand::distributions::{Distribution, Alphanumeric, Uniform, Standard}; |
53 | | /// |
54 | | /// let mut rng = thread_rng(); |
55 | | /// |
56 | | /// // Vec of 16 x f32: |
57 | | /// let v: Vec<f32> = Standard.sample_iter(&mut rng).take(16).collect(); |
58 | | /// |
59 | | /// // String: |
60 | | /// let s: String = Alphanumeric |
61 | | /// .sample_iter(&mut rng) |
62 | | /// .take(7) |
63 | | /// .map(char::from) |
64 | | /// .collect(); |
65 | | /// |
66 | | /// // Dice-rolling: |
67 | | /// let die_range = Uniform::new_inclusive(1, 6); |
68 | | /// let mut roll_die = die_range.sample_iter(&mut rng); |
69 | | /// while roll_die.next().unwrap() != 6 { |
70 | | /// println!("Not a 6; rolling again!"); |
71 | | /// } |
72 | | /// ``` |
73 | 0 | fn sample_iter<R>(self, rng: R) -> DistIter<Self, R, T> |
74 | 0 | where |
75 | 0 | R: Rng, |
76 | 0 | Self: Sized, |
77 | | { |
78 | 0 | DistIter { |
79 | 0 | distr: self, |
80 | 0 | rng, |
81 | 0 | phantom: ::core::marker::PhantomData, |
82 | 0 | } |
83 | 0 | } |
84 | | |
85 | | /// Create a distribution of values of 'S' by mapping the output of `Self` |
86 | | /// through the closure `F` |
87 | | /// |
88 | | /// # Example |
89 | | /// |
90 | | /// ``` |
91 | | /// use rand::thread_rng; |
92 | | /// use rand::distributions::{Distribution, Uniform}; |
93 | | /// |
94 | | /// let mut rng = thread_rng(); |
95 | | /// |
96 | | /// let die = Uniform::new_inclusive(1, 6); |
97 | | /// let even_number = die.map(|num| num % 2 == 0); |
98 | | /// while !even_number.sample(&mut rng) { |
99 | | /// println!("Still odd; rolling again!"); |
100 | | /// } |
101 | | /// ``` |
102 | 0 | fn map<F, S>(self, func: F) -> DistMap<Self, F, T, S> |
103 | 0 | where |
104 | 0 | F: Fn(T) -> S, |
105 | 0 | Self: Sized, |
106 | | { |
107 | 0 | DistMap { |
108 | 0 | distr: self, |
109 | 0 | func, |
110 | 0 | phantom: ::core::marker::PhantomData, |
111 | 0 | } |
112 | 0 | } |
113 | | } |
114 | | |
115 | | impl<'a, T, D: Distribution<T>> Distribution<T> for &'a D { |
116 | 0 | fn sample<R: Rng + ?Sized>(&self, rng: &mut R) -> T { |
117 | 0 | (*self).sample(rng) |
118 | 0 | } |
119 | | } |
120 | | |
121 | | /// An iterator that generates random values of `T` with distribution `D`, |
122 | | /// using `R` as the source of randomness. |
123 | | /// |
124 | | /// This `struct` is created by the [`sample_iter`] method on [`Distribution`]. |
125 | | /// See its documentation for more. |
126 | | /// |
127 | | /// [`sample_iter`]: Distribution::sample_iter |
128 | | #[derive(Debug)] |
129 | | pub struct DistIter<D, R, T> { |
130 | | distr: D, |
131 | | rng: R, |
132 | | phantom: ::core::marker::PhantomData<T>, |
133 | | } |
134 | | |
135 | | impl<D, R, T> Iterator for DistIter<D, R, T> |
136 | | where |
137 | | D: Distribution<T>, |
138 | | R: Rng, |
139 | | { |
140 | | type Item = T; |
141 | | |
142 | | #[inline(always)] |
143 | 0 | fn next(&mut self) -> Option<T> { |
144 | | // Here, self.rng may be a reference, but we must take &mut anyway. |
145 | | // Even if sample could take an R: Rng by value, we would need to do this |
146 | | // since Rng is not copyable and we cannot enforce that this is "reborrowable". |
147 | 0 | Some(self.distr.sample(&mut self.rng)) |
148 | 0 | } |
149 | | |
150 | 0 | fn size_hint(&self) -> (usize, Option<usize>) { |
151 | 0 | (usize::max_value(), None) |
152 | 0 | } |
153 | | } |
154 | | |
155 | | impl<D, R, T> iter::FusedIterator for DistIter<D, R, T> |
156 | | where |
157 | | D: Distribution<T>, |
158 | | R: Rng, |
159 | | { |
160 | | } |
161 | | |
162 | | #[cfg(feature = "nightly")] |
163 | | unsafe impl<D, R, T> iter::TrustedLen for DistIter<D, R, T> |
164 | | where |
165 | | D: Distribution<T>, |
166 | | R: Rng, |
167 | | { |
168 | | } |
169 | | |
170 | | /// A distribution of values of type `S` derived from the distribution `D` |
171 | | /// by mapping its output of type `T` through the closure `F`. |
172 | | /// |
173 | | /// This `struct` is created by the [`Distribution::map`] method. |
174 | | /// See its documentation for more. |
175 | | #[derive(Debug)] |
176 | | pub struct DistMap<D, F, T, S> { |
177 | | distr: D, |
178 | | func: F, |
179 | | phantom: ::core::marker::PhantomData<fn(T) -> S>, |
180 | | } |
181 | | |
182 | | impl<D, F, T, S> Distribution<S> for DistMap<D, F, T, S> |
183 | | where |
184 | | D: Distribution<T>, |
185 | | F: Fn(T) -> S, |
186 | | { |
187 | 0 | fn sample<R: Rng + ?Sized>(&self, rng: &mut R) -> S { |
188 | 0 | (self.func)(self.distr.sample(rng)) |
189 | 0 | } |
190 | | } |
191 | | |
192 | | /// `String` sampler |
193 | | /// |
194 | | /// Sampling a `String` of random characters is not quite the same as collecting |
195 | | /// a sequence of chars. This trait contains some helpers. |
196 | | #[cfg(feature = "alloc")] |
197 | | pub trait DistString { |
198 | | /// Append `len` random chars to `string` |
199 | | fn append_string<R: Rng + ?Sized>(&self, rng: &mut R, string: &mut String, len: usize); |
200 | | |
201 | | /// Generate a `String` of `len` random chars |
202 | | #[inline] |
203 | 0 | fn sample_string<R: Rng + ?Sized>(&self, rng: &mut R, len: usize) -> String { |
204 | 0 | let mut s = String::new(); |
205 | 0 | self.append_string(rng, &mut s, len); |
206 | 0 | s |
207 | 0 | } |
208 | | } |
209 | | |
210 | | #[cfg(test)] |
211 | | mod tests { |
212 | | use crate::distributions::{Distribution, Uniform}; |
213 | | use crate::Rng; |
214 | | |
215 | | #[test] |
216 | | fn test_distributions_iter() { |
217 | | use crate::distributions::Open01; |
218 | | let mut rng = crate::test::rng(210); |
219 | | let distr = Open01; |
220 | | let mut iter = Distribution::<f32>::sample_iter(distr, &mut rng); |
221 | | let mut sum: f32 = 0.; |
222 | | for _ in 0..100 { |
223 | | sum += iter.next().unwrap(); |
224 | | } |
225 | | assert!(0. < sum && sum < 100.); |
226 | | } |
227 | | |
228 | | #[test] |
229 | | fn test_distributions_map() { |
230 | | let dist = Uniform::new_inclusive(0, 5).map(|val| val + 15); |
231 | | |
232 | | let mut rng = crate::test::rng(212); |
233 | | let val = dist.sample(&mut rng); |
234 | | assert!((15..=20).contains(&val)); |
235 | | } |
236 | | |
237 | | #[test] |
238 | | fn test_make_an_iter() { |
239 | | fn ten_dice_rolls_other_than_five<R: Rng>( |
240 | | rng: &mut R, |
241 | | ) -> impl Iterator<Item = i32> + '_ { |
242 | | Uniform::new_inclusive(1, 6) |
243 | | .sample_iter(rng) |
244 | | .filter(|x| *x != 5) |
245 | | .take(10) |
246 | | } |
247 | | |
248 | | let mut rng = crate::test::rng(211); |
249 | | let mut count = 0; |
250 | | for val in ten_dice_rolls_other_than_five(&mut rng) { |
251 | | assert!((1..=6).contains(&val) && val != 5); |
252 | | count += 1; |
253 | | } |
254 | | assert_eq!(count, 10); |
255 | | } |
256 | | |
257 | | #[test] |
258 | | #[cfg(feature = "alloc")] |
259 | | fn test_dist_string() { |
260 | | use core::str; |
261 | | use crate::distributions::{Alphanumeric, DistString, Standard}; |
262 | | let mut rng = crate::test::rng(213); |
263 | | |
264 | | let s1 = Alphanumeric.sample_string(&mut rng, 20); |
265 | | assert_eq!(s1.len(), 20); |
266 | | assert_eq!(str::from_utf8(s1.as_bytes()), Ok(s1.as_str())); |
267 | | |
268 | | let s2 = Standard.sample_string(&mut rng, 20); |
269 | | assert_eq!(s2.chars().count(), 20); |
270 | | assert_eq!(str::from_utf8(s2.as_bytes()), Ok(s2.as_str())); |
271 | | } |
272 | | } |