1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486
/* * This Source Code Form is subject to the terms of * the Mozilla Public License, v. 2.0. If a copy of * the MPL was not distributed with this file, You * can obtain one at http://mozilla.org/MPL/2.0/. */ //! WIP see README⚠️ //! hanower is a CLI which calculates exponential backoffs from user input values. //#![deny(missing_docs)] use std::fmt; /// Used to create and work with intervals which are calculated from the user-input CLI values. /// /// - `low` is the starting point of the section from which to find intervals /// - `high` is the inclusive end point of the section from which to find intervals /// - `count` is the total number of desired intervals to be calculated /// - must be a minimum of 2 (`low` and `high`) #[derive(Debug, Clone, Copy)] pub struct Interval { low: f64, high: f64, count: u64, } impl Interval { /// Creates a new Interval, with the range `low..=high`, /// split into `count` number of intervals. pub fn new(low: f64, high: f64, count: u64) -> Result<Self, IntervalError> { if low >= high { Err(IntervalError::InvalidRange) } else if count < 2 { Err(IntervalError::LowCount(count)) } else { Ok(Self { low, high, count }) } } /// Returns the `low` value. pub fn low(&self) -> f64 { self.low } /// Returns the `high` value. pub fn high(&self) -> f64 { self.high } /// Returns the `count` value. pub fn count(&self) -> u64 { self.count } // TODO: fix floating point accuracy bug /// Finds the bucket a given value exists in. /// /// A bucket refers to a range between two values, and including the starting value. /// /// For example, say we have values of `low = 1`, `high = 10`, and `count = 5`, /// and want to know which bucket the number `8` would be in. The output intervals /// would be `2 3 4 6 10`. The first bucket is then `2..<3`, next `3..<4`, etc. /// So, `8` is in the fourth bucket, between `6` and `10`. pub fn bucket(&self, number: f64) -> Option<usize> { if number < self.low() || number >= self.high() { return None; } let bucket = f64::ln(number - self.low() + 1.0) / f64::ln(self.high() - self.low() + 1.0) * self.count() as f64; Some(dbg!(bucket).trunc() as usize) } // /// Iterates through a given list of numbers, and finds the appropriate // /// matching value from the vec of resultant interval values. // /// - you can search for the first, last or average values that fit into a bucket // pub fn in_list(&self, mut list: Vec<i64>) -> Option<Vec<i64>> { // // ensures buckets are in order // // maybe remove this, add error handling for this to the arg itself (when arg is added)? // list.sort(); // // --- calculating resultant interval values --- // let mut interval_values: Vec<i64> = vec![]; // for number in self.intervals().map(|f| f.round() as i64) { // interval_values.push(number); // } // // --- calculating which resultant values fit the specified requirements --- // let in_buckets: Vec<i64> = vec![]; // // for value in list // // if value == list.last() // // return in_buckets // // else // // // if in_buckets == vec![] { // None // } else { // Some(in_buckets) // } // } /// Returns an iterator of lazily evaluated intervals, starting from this /// Interval's `low` value up to and including the `high` value. pub fn iter(&self) -> IntervalIter { self.new_iter() } /// Returns an iterator of lazily evaluated intervals based on the /// `low` and `high points` of this Interval, and skips the floor value. pub fn intervals(&self) -> IntervalIter { let mut iter = self.new_iter(); // Skip the floor value iter.next(); iter } fn new_iter(&self) -> IntervalIter { debug_assert!(self.low < self.high, "Low must be less than high"); debug_assert!(self.count >= 2, "Interval count must be >= 2."); IntervalIter::new(self.low, self.high, self.count) } } /// An iterator version of the [`Interval`] struct, which calculates intervals from the user-input CLI values. /// /// - `low` is the starting point of the section from which to find intervals /// - `high` is the inclusive end point of the section from which to find intervals /// - `count` is the total number of desired intervals to be calculated /// - must be a minimum of 2 (`low` and `high`) /// - `idx_front` and `idx_back` are used to keep track of where the iterator is #[derive(Debug, Clone)] pub struct IntervalIter { low: f64, high: f64, count: u64, // Used by next() idx_front: u64, // Used by next_back() idx_back: u64, } impl IntervalIter { fn new(low: f64, high: f64, count: u64) -> Self { Self { low, high, count, idx_front: 0, idx_back: 0, } } fn idx(&self) -> u64 { self.idx_front + self.idx_back } fn calculate_interval(&self, index: u64) -> f64 { // scales `high` value down according to `low` value // `low` must always move down to 1.0 let nlog = (self.high - self.low + 1.0).ln() / self.count as f64; let expo = (nlog * index as f64).exp(); expo + self.low - 1.0 } } impl Iterator for IntervalIter { type Item = f64; fn next(&mut self) -> Option<Self::Item> { if self.idx() > self.count { return None; } match self.idx_front { 0 => { self.idx_front += 1; Some(self.low) } index => { let interval = self.calculate_interval(index); self.idx_front += 1; Some(interval) } } } fn size_hint(&self) -> (usize, Option<usize>) { // Because we iterate over `low` *and* `count` number // of intervals we need to add one let len = (self.count + 1) - self.idx(); let len = len as usize; (len, Some(len)) } } impl DoubleEndedIterator for IntervalIter { fn next_back(&mut self) -> Option<Self::Item> { if self.idx() > self.count { return None; } match self.count.checked_sub(self.idx_back) { Some(0) => { self.idx_back += 1; Some(self.low) } Some(index) => { let interval = self.calculate_interval(index); self.idx_back += 1; Some(interval) } None => { self.idx_back += 1; None } } } } impl ExactSizeIterator for IntervalIter {} impl std::iter::FusedIterator for IntervalIter {} /// Error kinds for command line arguments. #[derive(Debug)] pub enum IntervalError { /// Occurs when the user provides a `count` value below 2. LowCount(u64), /// Occurs when the user gives a `low` value >= `high`. InvalidRange, } impl fmt::Display for IntervalError { fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { match self { Self::LowCount(bad) => write!( f, "Invalid count. Ensure `number` value is >= 2 (was: {})", bad ), Self::InvalidRange => { write!(f, "Invalid range. Ensure `start` value is less than `end`") } } } } impl std::error::Error for IntervalError {} // fn read_csv(path: String) -> Result<Vec<f64>, Box<dyn Error>> { // let mut reader = csv::Reader::from_path(path)?; // let mut index_vec = vec![0.0]; // for result in reader.records() { // let record = result?; // let value: f64 = record[0].parse().unwrap(); // index_vec.push(value); // } // Ok(index_vec) // } #[cfg(test)] mod tests { #![allow(unused_imports)] // All functions (excluding main), structs and traits that are defined above // can be used in this module use super::*; use anyhow::{bail as error, Error as AnyError}; /// Typedef of the Results our #[test] functions return type TestResult = std::result::Result<(), AnyError>; /* * For reference this is how I calculated the expected values in `test_data`: * * -- INPUTS * * $START : The low value * * $END : The high value * * $COUNT : The number of fences to divide $START..=$END into * * -- OUTPUT * * $OUTPUT : An array containing $COUNT fence points * * -- FUNCTIONS * * ln : See [f64::ln](https://doc.rust-lang.org/std/primitive.f64.html#method.ln) * * exp : See [f64::exp](https://doc.rust-lang.org/std/primitive.f64.html#method.ln) * * -- ALGORITHM * $CEILING := ln($END - $START + 1) * $NLOG := $CEILING / $COUNT * * for $IDX in 1..=$COUNT: * $EXP := exp($NLOG * $IDX) * $POST := $EXP + $START - 1 * $OUTPUT += $POST * * return $OUTPUT */ /* --- TESTS --- */ #[test] /// Checks that the Interval struct correctly detects and refuses invalid input values. fn start_after_end_err() -> TestResult { let args = Interval::new(10.0, 1.0, 5); // Assert that bad inputs lead to an error assert!(args.is_err()); Ok(()) } #[test] /// Checks that the Interval struct correctly detects and refuses invalid count values fn count_less_than_two_err() -> TestResult { let args = Interval::new(1.0, 10.0, 1); // Assert that a bad count leads to an error assert!(args.is_err()); Ok(()) } #[test] /// Runs the program's computed intervals against a series of precomputed data sets, /// checking that all of the actual outputs match the expected values fn hanower_algorithm_iter() -> TestResult { // For each set of args and precomputed outputs for (args, expected_list) in test_data().into_iter() { // Generate the actual outputs let actual_list: Vec<f64> = args.intervals().collect(); // For each expected and actual data sets actual_list.iter().zip(expected_list.iter()).enumerate().try_for_each(|(idx, (&actual, &expected))| { // Round the actual item let rounded = actual.round() as i64; // Check that the rounded item matches the precomputed item if rounded != expected { let msg = format!( "@{} => Expected {}, received {} ({})\nExpected Values | {:?}\nActual Values | {:?}", idx, expected, rounded, actual, expected_list, actual_list ); error!(msg); } Ok(()) })?; // Assert the lists are the same length, if they aren't the actual data is wrong assert_output_length(actual_list.len(), expected_list.len())? } Ok(()) } #[test] fn hanower_algorithm_iter_back() -> TestResult { for (interval, expected_list) in test_data().into_iter() { let actual_list: IntervalIter = interval.intervals(); interval.intervals().rev().zip(expected_list.iter().rev()).enumerate().try_for_each(|(idx, (actual, &expected))| { // Round the actual item let rounded = actual.round() as i64; // Check that the rounded item matches the precomputed item if rounded != expected { let msg = format!( "@{} => Expected {}, received {} ({})\nExpected Values | {:?}\nActual Values | {:?}", idx, expected, rounded, actual, expected_list, actual_list ); error!(msg); } Ok(()) })?; assert_output_length(actual_list.len(), expected_list.len())? } Ok(()) } // tests the bucket method of Interval #[test] fn interval_bucket_method() { let data = vec![ BucketTestData::new(Interval::new(1.0, 10.0, 5).unwrap(), 7.0, Some(4)), BucketTestData::new(Interval::new(30.0, 100.0, 10).unwrap(), 1000.0, None), BucketTestData::new(Interval::new(30.0, 100.0, 10).unwrap(), 0.0, None), BucketTestData::new(Interval::new(30.0, 100.0, 10).unwrap(), 10.0 * 10.0, None), BucketTestData::new(Interval::new(-100.0, 100.0, 10).unwrap(), 0.0, Some(8)), BucketTestData::new(Interval::new(-100.0, 100.0, 10).unwrap(), -100.0, Some(0)), ]; for test in data { let actual = Interval::bucket(&test.interval, test.number); assert_eq!(test.expected, actual) } } // #[test] // fn first_in_buckets() { // let expected: Option<Vec<i64>> = Some(vec![24, 46, 67]); // let interval = Interval::new(10.0, 100.0, 10).unwrap(); // let actual = interval.in_list(vec![20, 40, 100, 60]); // assert_eq!(expected, actual) // } /* --- HELPER STRUCTS & IMPLEMENTATIONS --- */ struct BucketTestData { interval: Interval, number: f64, expected: Option<usize>, } impl BucketTestData { fn new(interval: Interval, number: f64, expected: Option<usize>) -> Self { Self { interval, number, expected, } } } /* --- HELPER FUNCTIONS --- */ fn assert_output_length(actual: usize, expected: usize) -> TestResult { if let false = actual == expected { let msg = format!("Expected {} fences, but received: {}", expected, actual); error!(msg); } Ok(()) } // helper function for `hanower_algorithm_iter` test // unwrap helps validate that the input data (Interval::new) is correct for test fn test_data() -> Vec<(Interval, Vec<i64>)> { vec![ (Interval::new(1.0, 16.0, 4).unwrap(), vec![2, 4, 8, 16]), ( Interval::new(100.0, 1000.0, 15).unwrap(), vec![ 101, 101, 103, 105, 109, 114, 123, 137, 158, 192, 246, 330, 463, 671, 1000, ], ), ( Interval::new(3.0, 72.0, 9).unwrap(), vec![4, 5, 6, 9, 13, 19, 29, 46, 72], ), (Interval::new(-19.0, 12.0, 3).unwrap(), vec![-17, -10, 12]), ( Interval::new(-11000.0, -1200.0, 16).unwrap(), vec![ -10999, -10998, -10995, -10991, -10983, -10970, -10945, -10902, -10825, -10689, -10446, -10016, -9252, -7894, -5483, -1200, ], ), ] } }