| 1 |
/*! |
| 2 |
Lower level primitive types that are useful in a variety of circumstances. |
| 3 |
|
| 4 |
# Overview |
| 5 |
|
| 6 |
This list represents the principle types in this module and briefly describes |
| 7 |
when you might want to use them. |
| 8 |
|
| 9 |
* [`PatternID`] - A type that represents the identifier of a regex pattern. |
| 10 |
This is probably the most widely used type in this module (which is why it's |
| 11 |
also re-exported in the crate root). |
| 12 |
* [`StateID`] - A type the represents the identifier of a finite automaton |
| 13 |
state. This is used for both NFAs and DFAs, with the notable exception of |
| 14 |
the hybrid NFA/DFA. (The hybrid NFA/DFA uses a special purpose "lazy" state |
| 15 |
identifier.) |
| 16 |
* [`SmallIndex`] - The internal representation of both a `PatternID` and a |
| 17 |
`StateID`. Its purpose is to serve as a type that can index memory without |
| 18 |
being as big as a `usize` on 64-bit targets. The main idea behind this type |
| 19 |
is that there are many things in regex engines that will, in practice, never |
| 20 |
overflow a 32-bit integer. (For example, like the number of patterns in a regex |
| 21 |
or the number of states in an NFA.) Thus, a `SmallIndex` can be used to index |
| 22 |
memory without peppering `as` casts everywhere. Moreover, it forces callers |
| 23 |
to handle errors in the case where, somehow, the value would otherwise overflow |
| 24 |
either a 32-bit integer or a `usize` (e.g., on 16-bit targets). |
| 25 |
* [`NonMaxUsize`] - Represents a `usize` that cannot be `usize::MAX`. As a |
| 26 |
result, `Option<NonMaxUsize>` has the same size in memory as a `usize`. This |
| 27 |
useful, for example, when representing the offsets of submatches since it |
| 28 |
reduces memory usage by a factor of 2. It is a legal optimization since Rust |
| 29 |
guarantees that slices never have a length that exceeds `isize::MAX`. |
| 30 |
*/ |
| 31 |
|
| 32 |
use core::num::NonZeroUsize; |
| 33 |
|
| 34 |
#[cfg(feature = "alloc")] |
| 35 |
use alloc::vec::Vec; |
| 36 |
|
| 37 |
use crate::util::int::{Usize, U16, U32, U64}; |
| 38 |
|
| 39 |
/// A `usize` that can never be `usize::MAX`. |
| 40 |
/// |
| 41 |
/// This is similar to `core::num::NonZeroUsize`, but instead of not permitting |
| 42 |
/// a zero value, this does not permit a max value. |
| 43 |
/// |
| 44 |
/// This is useful in certain contexts where one wants to optimize the memory |
| 45 |
/// usage of things that contain match offsets. Namely, since Rust slices |
| 46 |
/// are guaranteed to never have a length exceeding `isize::MAX`, we can use |
| 47 |
/// `usize::MAX` as a sentinel to indicate that no match was found. Indeed, |
| 48 |
/// types like `Option<NonMaxUsize>` have exactly the same size in memory as a |
| 49 |
/// `usize`. |
| 50 |
/// |
| 51 |
/// This type is defined to be `repr(transparent)` for |
| 52 |
/// `core::num::NonZeroUsize`, which is in turn defined to be |
| 53 |
/// `repr(transparent)` for `usize`. |
| 54 |
#[derive(Clone, Copy, Eq, Hash, PartialEq, PartialOrd, Ord)] |
| 55 |
#[repr(transparent)] |
| 56 |
pub struct NonMaxUsize(NonZeroUsize); |
| 57 |
|
| 58 |
impl NonMaxUsize { |
| 59 |
/// Create a new `NonMaxUsize` from the given value. |
| 60 |
/// |
| 61 |
/// This returns `None` only when the given value is equal to `usize::MAX`. |
| 62 |
#[inline] |
| 63 |
pub fn new(value: usize) -> Option<NonMaxUsize> { |
| 64 |
NonZeroUsize::new(value.wrapping_add(1)).map(NonMaxUsize) |
| 65 |
} |
| 66 |
|
| 67 |
/// Return the underlying `usize` value. The returned value is guaranteed |
| 68 |
/// to not equal `usize::MAX`. |
| 69 |
#[inline] |
| 70 |
pub fn get(self) -> usize { |
| 71 |
self.0.get().wrapping_sub(1) |
| 72 |
} |
| 73 |
} |
| 74 |
|
| 75 |
// We provide our own Debug impl because seeing the internal repr can be quite |
| 76 |
// surprising if you aren't expecting it. e.g., 'NonMaxUsize(5)' vs just '5'. |
| 77 |
impl core::fmt::Debug for NonMaxUsize { |
| 78 |
fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result { |
| 79 |
write!(f, "{:?}", self.get()) |
| 80 |
} |
| 81 |
} |
| 82 |
|
| 83 |
/// A type that represents a "small" index. |
| 84 |
/// |
| 85 |
/// The main idea of this type is to provide something that can index memory, |
| 86 |
/// but uses less memory than `usize` on 64-bit systems. Specifically, its |
| 87 |
/// representation is always a `u32` and has `repr(transparent)` enabled. (So |
| 88 |
/// it is safe to transmute between a `u32` and a `SmallIndex`.) |
| 89 |
/// |
| 90 |
/// A small index is typically useful in cases where there is no practical way |
| 91 |
/// that the index will overflow a 32-bit integer. A good example of this is |
| 92 |
/// an NFA state. If you could somehow build an NFA with `2^30` states, its |
| 93 |
/// memory usage would be exorbitant and its runtime execution would be so |
| 94 |
/// slow as to be completely worthless. Therefore, this crate generally deems |
| 95 |
/// it acceptable to return an error if it would otherwise build an NFA that |
| 96 |
/// requires a slice longer than what a 32-bit integer can index. In exchange, |
| 97 |
/// we can use 32-bit indices instead of 64-bit indices in various places. |
| 98 |
/// |
| 99 |
/// This type ensures this by providing a constructor that will return an error |
| 100 |
/// if its argument cannot fit into the type. This makes it much easier to |
| 101 |
/// handle these sorts of boundary cases that are otherwise extremely subtle. |
| 102 |
/// |
| 103 |
/// On all targets, this type guarantees that its value will fit in a `u32`, |
| 104 |
/// `i32`, `usize` and an `isize`. This means that on 16-bit targets, for |
| 105 |
/// example, this type's maximum value will never overflow an `isize`, |
| 106 |
/// which means it will never overflow a `i16` even though its internal |
| 107 |
/// representation is still a `u32`. |
| 108 |
/// |
| 109 |
/// The purpose for making the type fit into even signed integer types like |
| 110 |
/// `isize` is to guarantee that the difference between any two small indices |
| 111 |
/// is itself also a small index. This is useful in certain contexts, e.g., |
| 112 |
/// for delta encoding. |
| 113 |
/// |
| 114 |
/// # Other types |
| 115 |
/// |
| 116 |
/// The following types wrap `SmallIndex` to provide a more focused use case: |
| 117 |
/// |
| 118 |
/// * [`PatternID`] is for representing the identifiers of patterns. |
| 119 |
/// * [`StateID`] is for representing the identifiers of states in finite |
| 120 |
/// automata. It is used for both NFAs and DFAs. |
| 121 |
/// |
| 122 |
/// # Representation |
| 123 |
/// |
| 124 |
/// This type is always represented internally by a `u32` and is marked as |
| 125 |
/// `repr(transparent)`. Thus, this type always has the same representation as |
| 126 |
/// a `u32`. It is thus safe to transmute between a `u32` and a `SmallIndex`. |
| 127 |
/// |
| 128 |
/// # Indexing |
| 129 |
/// |
| 130 |
/// For convenience, callers may use a `SmallIndex` to index slices. |
| 131 |
/// |
| 132 |
/// # Safety |
| 133 |
/// |
| 134 |
/// While a `SmallIndex` is meant to guarantee that its value fits into `usize` |
| 135 |
/// without using as much space as a `usize` on all targets, callers must |
| 136 |
/// not rely on this property for safety. Callers may choose to rely on this |
| 137 |
/// property for correctness however. For example, creating a `SmallIndex` with |
| 138 |
/// an invalid value can be done in entirely safe code. This may in turn result |
| 139 |
/// in panics or silent logical errors. |
| 140 |
#[derive( |
| 141 |
Clone, Copy, Debug, Default, Eq, Hash, PartialEq, PartialOrd, Ord, |
| 142 |
)] |
| 143 |
#[repr(transparent)] |
| 144 |
pub struct SmallIndex(u32); |
| 145 |
|
| 146 |
impl SmallIndex { |
| 147 |
/// The maximum index value. |
| 148 |
#[cfg(any(target_pointer_width = "32", target_pointer_width = "64"))] |
| 149 |
pub const MAX: SmallIndex = |
| 150 |
// FIXME: Use as_usize() once const functions in traits are stable. |
| 151 |
SmallIndex::new_unchecked(core::i32::MAX as usize - 1); |
| 152 |
|
| 153 |
/// The maximum index value. |
| 154 |
#[cfg(target_pointer_width = "16")] |
| 155 |
pub const MAX: SmallIndex = |
| 156 |
SmallIndex::new_unchecked(core::isize::MAX - 1); |
| 157 |
|
| 158 |
/// The total number of values that can be represented as a small index. |
| 159 |
pub const LIMIT: usize = SmallIndex::MAX.as_usize() + 1; |
| 160 |
|
| 161 |
/// The zero index value. |
| 162 |
pub const ZERO: SmallIndex = SmallIndex::new_unchecked(0); |
| 163 |
|
| 164 |
/// The number of bytes that a single small index uses in memory. |
| 165 |
pub const SIZE: usize = core::mem::size_of::<SmallIndex>(); |
| 166 |
|
| 167 |
/// Create a new small index. |
| 168 |
/// |
| 169 |
/// If the given index exceeds [`SmallIndex::MAX`], then this returns |
| 170 |
/// an error. |
| 171 |
#[inline] |
| 172 |
pub fn new(index: usize) -> Result<SmallIndex, SmallIndexError> { |
| 173 |
SmallIndex::try_from(index) |
| 174 |
} |
| 175 |
|
| 176 |
/// Create a new small index without checking whether the given value |
| 177 |
/// exceeds [`SmallIndex::MAX`]. |
| 178 |
/// |
| 179 |
/// Using this routine with an invalid index value will result in |
| 180 |
/// unspecified behavior, but *not* undefined behavior. In particular, an |
| 181 |
/// invalid index value is likely to cause panics or possibly even silent |
| 182 |
/// logical errors. |
| 183 |
/// |
| 184 |
/// Callers must never rely on a `SmallIndex` to be within a certain range |
| 185 |
/// for memory safety. |
| 186 |
#[inline] |
| 187 |
pub const fn new_unchecked(index: usize) -> SmallIndex { |
| 188 |
// FIXME: Use as_u32() once const functions in traits are stable. |
| 189 |
SmallIndex(index as u32) |
| 190 |
} |
| 191 |
|
| 192 |
/// Like [`SmallIndex::new`], but panics if the given index is not valid. |
| 193 |
#[inline] |
| 194 |
pub fn must(index: usize) -> SmallIndex { |
| 195 |
SmallIndex::new(index).expect("invalid small index") |
| 196 |
} |
| 197 |
|
| 198 |
/// Return this small index as a `usize`. This is guaranteed to never |
| 199 |
/// overflow `usize`. |
| 200 |
#[inline] |
| 201 |
pub const fn as_usize(&self) -> usize { |
| 202 |
// FIXME: Use as_usize() once const functions in traits are stable. |
| 203 |
self.0 as usize |
| 204 |
} |
| 205 |
|
| 206 |
/// Return this small index as a `u64`. This is guaranteed to never |
| 207 |
/// overflow. |
| 208 |
#[inline] |
| 209 |
pub const fn as_u64(&self) -> u64 { |
| 210 |
// FIXME: Use u64::from() once const functions in traits are stable. |
| 211 |
self.0 as u64 |
| 212 |
} |
| 213 |
|
| 214 |
/// Return the internal `u32` of this small index. This is guaranteed to |
| 215 |
/// never overflow `u32`. |
| 216 |
#[inline] |
| 217 |
pub const fn as_u32(&self) -> u32 { |
| 218 |
self.0 |
| 219 |
} |
| 220 |
|
| 221 |
/// Return the internal `u32` of this small index represented as an `i32`. |
| 222 |
/// This is guaranteed to never overflow an `i32`. |
| 223 |
#[inline] |
| 224 |
pub const fn as_i32(&self) -> i32 { |
| 225 |
// This is OK because we guarantee that our max value is <= i32::MAX. |
| 226 |
self.0 as i32 |
| 227 |
} |
| 228 |
|
| 229 |
/// Returns one more than this small index as a usize. |
| 230 |
/// |
| 231 |
/// Since a small index has constraints on its maximum value, adding `1` to |
| 232 |
/// it will always fit in a `usize`, `u32` and a `i32`. |
| 233 |
#[inline] |
| 234 |
pub fn one_more(&self) -> usize { |
| 235 |
self.as_usize() + 1 |
| 236 |
} |
| 237 |
|
| 238 |
/// Decode this small index from the bytes given using the native endian |
| 239 |
/// byte order for the current target. |
| 240 |
/// |
| 241 |
/// If the decoded integer is not representable as a small index for the |
| 242 |
/// current target, then this returns an error. |
| 243 |
#[inline] |
| 244 |
pub fn from_ne_bytes( |
| 245 |
bytes: [u8; 4], |
| 246 |
) -> Result<SmallIndex, SmallIndexError> { |
| 247 |
let id = u32::from_ne_bytes(bytes); |
| 248 |
if id > SmallIndex::MAX.as_u32() { |
| 249 |
return Err(SmallIndexError { attempted: u64::from(id) }); |
| 250 |
} |
| 251 |
Ok(SmallIndex::new_unchecked(id.as_usize())) |
| 252 |
} |
| 253 |
|
| 254 |
/// Decode this small index from the bytes given using the native endian |
| 255 |
/// byte order for the current target. |
| 256 |
/// |
| 257 |
/// This is analogous to [`SmallIndex::new_unchecked`] in that is does not |
| 258 |
/// check whether the decoded integer is representable as a small index. |
| 259 |
#[inline] |
| 260 |
pub fn from_ne_bytes_unchecked(bytes: [u8; 4]) -> SmallIndex { |
| 261 |
SmallIndex::new_unchecked(u32::from_ne_bytes(bytes).as_usize()) |
| 262 |
} |
| 263 |
|
| 264 |
/// Return the underlying small index integer as raw bytes in native endian |
| 265 |
/// format. |
| 266 |
#[inline] |
| 267 |
pub fn to_ne_bytes(&self) -> [u8; 4] { |
| 268 |
self.0.to_ne_bytes() |
| 269 |
} |
| 270 |
} |
| 271 |
|
| 272 |
impl<T> core::ops::Index<SmallIndex> for [T] { |
| 273 |
type Output = T; |
| 274 |
|
| 275 |
#[inline] |
| 276 |
fn index(&self, index: SmallIndex) -> &T { |
| 277 |
&self[index.as_usize()] |
| 278 |
} |
| 279 |
} |
| 280 |
|
| 281 |
impl<T> core::ops::IndexMut<SmallIndex> for [T] { |
| 282 |
#[inline] |
| 283 |
fn index_mut(&mut self, index: SmallIndex) -> &mut T { |
| 284 |
&mut self[index.as_usize()] |
| 285 |
} |
| 286 |
} |
| 287 |
|
| 288 |
#[cfg(feature = "alloc")] |
| 289 |
impl<T> core::ops::Index<SmallIndex> for Vec<T> { |
| 290 |
type Output = T; |
| 291 |
|
| 292 |
#[inline] |
| 293 |
fn index(&self, index: SmallIndex) -> &T { |
| 294 |
&self[index.as_usize()] |
| 295 |
} |
| 296 |
} |
| 297 |
|
| 298 |
#[cfg(feature = "alloc")] |
| 299 |
impl<T> core::ops::IndexMut<SmallIndex> for Vec<T> { |
| 300 |
#[inline] |
| 301 |
fn index_mut(&mut self, index: SmallIndex) -> &mut T { |
| 302 |
&mut self[index.as_usize()] |
| 303 |
} |
| 304 |
} |
| 305 |
|
| 306 |
impl From<u8> for SmallIndex { |
| 307 |
fn from(index: u8) -> SmallIndex { |
| 308 |
SmallIndex::new_unchecked(usize::from(index)) |
| 309 |
} |
| 310 |
} |
| 311 |
|
| 312 |
impl TryFrom<u16> for SmallIndex { |
| 313 |
type Error = SmallIndexError; |
| 314 |
|
| 315 |
fn try_from(index: u16) -> Result<SmallIndex, SmallIndexError> { |
| 316 |
if u32::from(index) > SmallIndex::MAX.as_u32() { |
| 317 |
return Err(SmallIndexError { attempted: u64::from(index) }); |
| 318 |
} |
| 319 |
Ok(SmallIndex::new_unchecked(index.as_usize())) |
| 320 |
} |
| 321 |
} |
| 322 |
|
| 323 |
impl TryFrom<u32> for SmallIndex { |
| 324 |
type Error = SmallIndexError; |
| 325 |
|
| 326 |
fn try_from(index: u32) -> Result<SmallIndex, SmallIndexError> { |
| 327 |
if index > SmallIndex::MAX.as_u32() { |
| 328 |
return Err(SmallIndexError { attempted: u64::from(index) }); |
| 329 |
} |
| 330 |
Ok(SmallIndex::new_unchecked(index.as_usize())) |
| 331 |
} |
| 332 |
} |
| 333 |
|
| 334 |
impl TryFrom<u64> for SmallIndex { |
| 335 |
type Error = SmallIndexError; |
| 336 |
|
| 337 |
fn try_from(index: u64) -> Result<SmallIndex, SmallIndexError> { |
| 338 |
if index > SmallIndex::MAX.as_u64() { |
| 339 |
return Err(SmallIndexError { attempted: index }); |
| 340 |
} |
| 341 |
Ok(SmallIndex::new_unchecked(index.as_usize())) |
| 342 |
} |
| 343 |
} |
| 344 |
|
| 345 |
impl TryFrom<usize> for SmallIndex { |
| 346 |
type Error = SmallIndexError; |
| 347 |
|
| 348 |
fn try_from(index: usize) -> Result<SmallIndex, SmallIndexError> { |
| 349 |
if index > SmallIndex::MAX.as_usize() { |
| 350 |
return Err(SmallIndexError { attempted: index.as_u64() }); |
| 351 |
} |
| 352 |
Ok(SmallIndex::new_unchecked(index)) |
| 353 |
} |
| 354 |
} |
| 355 |
|
| 356 |
#[cfg(test)] |
| 357 |
impl quickcheck::Arbitrary for SmallIndex { |
| 358 |
fn arbitrary(gen: &mut quickcheck::Gen) -> SmallIndex { |
| 359 |
use core::cmp::max; |
| 360 |
|
| 361 |
let id = max(i32::MIN + 1, i32::arbitrary(gen)).abs(); |
| 362 |
if id > SmallIndex::MAX.as_i32() { |
| 363 |
SmallIndex::MAX |
| 364 |
} else { |
| 365 |
SmallIndex::new(usize::try_from(id).unwrap()).unwrap() |
| 366 |
} |
| 367 |
} |
| 368 |
} |
| 369 |
|
| 370 |
/// This error occurs when a small index could not be constructed. |
| 371 |
/// |
| 372 |
/// This occurs when given an integer exceeding the maximum small index value. |
| 373 |
/// |
| 374 |
/// When the `std` feature is enabled, this implements the `Error` trait. |
| 375 |
#[derive(Clone, Debug, Eq, PartialEq)] |
| 376 |
pub struct SmallIndexError { |
| 377 |
attempted: u64, |
| 378 |
} |
| 379 |
|
| 380 |
impl SmallIndexError { |
| 381 |
/// Returns the value that could not be converted to a small index. |
| 382 |
pub fn attempted(&self) -> u64 { |
| 383 |
self.attempted |
| 384 |
} |
| 385 |
} |
| 386 |
|
| 387 |
#[cfg(feature = "std")] |
| 388 |
impl std::error::Error for SmallIndexError {} |
| 389 |
|
| 390 |
impl core::fmt::Display for SmallIndexError { |
| 391 |
fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result { |
| 392 |
write!( |
| 393 |
f, |
| 394 |
"failed to create small index from {:?}, which exceeds {:?}", |
| 395 |
self.attempted(), |
| 396 |
SmallIndex::MAX, |
| 397 |
) |
| 398 |
} |
| 399 |
} |
| 400 |
|
| 401 |
#[derive(Clone, Debug)] |
| 402 |
pub(crate) struct SmallIndexIter { |
| 403 |
rng: core::ops::Range<usize>, |
| 404 |
} |
| 405 |
|
| 406 |
impl Iterator for SmallIndexIter { |
| 407 |
type Item = SmallIndex; |
| 408 |
|
| 409 |
fn next(&mut self) -> Option<SmallIndex> { |
| 410 |
if self.rng.start >= self.rng.end { |
| 411 |
return None; |
| 412 |
} |
| 413 |
let next_id = self.rng.start + 1; |
| 414 |
let id = core::mem::replace(&mut self.rng.start, next_id); |
| 415 |
// new_unchecked is OK since we asserted that the number of |
| 416 |
// elements in this iterator will fit in an ID at construction. |
| 417 |
Some(SmallIndex::new_unchecked(id)) |
| 418 |
} |
| 419 |
} |
| 420 |
|
| 421 |
macro_rules! index_type_impls { |
| 422 |
($name:ident, $err:ident, $iter:ident, $withiter:ident) => { |
| 423 |
impl $name { |
| 424 |
/// The maximum value. |
| 425 |
pub const MAX: $name = $name(SmallIndex::MAX); |
| 426 |
|
| 427 |
/// The total number of values that can be represented. |
| 428 |
pub const LIMIT: usize = SmallIndex::LIMIT; |
| 429 |
|
| 430 |
/// The zero value. |
| 431 |
pub const ZERO: $name = $name(SmallIndex::ZERO); |
| 432 |
|
| 433 |
/// The number of bytes that a single value uses in memory. |
| 434 |
pub const SIZE: usize = SmallIndex::SIZE; |
| 435 |
|
| 436 |
/// Create a new value that is represented by a "small index." |
| 437 |
/// |
| 438 |
/// If the given index exceeds the maximum allowed value, then this |
| 439 |
/// returns an error. |
| 440 |
#[inline] |
| 441 |
pub fn new(value: usize) -> Result<$name, $err> { |
| 442 |
SmallIndex::new(value).map($name).map_err($err) |
| 443 |
} |
| 444 |
|
| 445 |
/// Create a new value without checking whether the given argument |
| 446 |
/// exceeds the maximum. |
| 447 |
/// |
| 448 |
/// Using this routine with an invalid value will result in |
| 449 |
/// unspecified behavior, but *not* undefined behavior. In |
| 450 |
/// particular, an invalid ID value is likely to cause panics or |
| 451 |
/// possibly even silent logical errors. |
| 452 |
/// |
| 453 |
/// Callers must never rely on this type to be within a certain |
| 454 |
/// range for memory safety. |
| 455 |
#[inline] |
| 456 |
pub const fn new_unchecked(value: usize) -> $name { |
| 457 |
$name(SmallIndex::new_unchecked(value)) |
| 458 |
} |
| 459 |
|
| 460 |
/// Like `new`, but panics if the given value is not valid. |
| 461 |
#[inline] |
| 462 |
pub fn must(value: usize) -> $name { |
| 463 |
$name::new(value).expect(concat!( |
| 464 |
"invalid ", |
| 465 |
stringify!($name), |
| 466 |
" value" |
| 467 |
)) |
| 468 |
} |
| 469 |
|
| 470 |
/// Return the internal value as a `usize`. This is guaranteed to |
| 471 |
/// never overflow `usize`. |
| 472 |
#[inline] |
| 473 |
pub const fn as_usize(&self) -> usize { |
| 474 |
self.0.as_usize() |
| 475 |
} |
| 476 |
|
| 477 |
/// Return the internal value as a `u64`. This is guaranteed to |
| 478 |
/// never overflow. |
| 479 |
#[inline] |
| 480 |
pub const fn as_u64(&self) -> u64 { |
| 481 |
self.0.as_u64() |
| 482 |
} |
| 483 |
|
| 484 |
/// Return the internal value as a `u32`. This is guaranteed to |
| 485 |
/// never overflow `u32`. |
| 486 |
#[inline] |
| 487 |
pub const fn as_u32(&self) -> u32 { |
| 488 |
self.0.as_u32() |
| 489 |
} |
| 490 |
|
| 491 |
/// Return the internal value as a i32`. This is guaranteed to |
| 492 |
/// never overflow an `i32`. |
| 493 |
#[inline] |
| 494 |
pub const fn as_i32(&self) -> i32 { |
| 495 |
self.0.as_i32() |
| 496 |
} |
| 497 |
|
| 498 |
/// Returns one more than this value as a usize. |
| 499 |
/// |
| 500 |
/// Since values represented by a "small index" have constraints |
| 501 |
/// on their maximum value, adding `1` to it will always fit in a |
| 502 |
/// `usize`, `u32` and a `i32`. |
| 503 |
#[inline] |
| 504 |
pub fn one_more(&self) -> usize { |
| 505 |
self.0.one_more() |
| 506 |
} |
| 507 |
|
| 508 |
/// Decode this value from the bytes given using the native endian |
| 509 |
/// byte order for the current target. |
| 510 |
/// |
| 511 |
/// If the decoded integer is not representable as a small index |
| 512 |
/// for the current target, then this returns an error. |
| 513 |
#[inline] |
| 514 |
pub fn from_ne_bytes(bytes: [u8; 4]) -> Result<$name, $err> { |
| 515 |
SmallIndex::from_ne_bytes(bytes).map($name).map_err($err) |
| 516 |
} |
| 517 |
|
| 518 |
/// Decode this value from the bytes given using the native endian |
| 519 |
/// byte order for the current target. |
| 520 |
/// |
| 521 |
/// This is analogous to `new_unchecked` in that is does not check |
| 522 |
/// whether the decoded integer is representable as a small index. |
| 523 |
#[inline] |
| 524 |
pub fn from_ne_bytes_unchecked(bytes: [u8; 4]) -> $name { |
| 525 |
$name(SmallIndex::from_ne_bytes_unchecked(bytes)) |
| 526 |
} |
| 527 |
|
| 528 |
/// Return the underlying integer as raw bytes in native endian |
| 529 |
/// format. |
| 530 |
#[inline] |
| 531 |
pub fn to_ne_bytes(&self) -> [u8; 4] { |
| 532 |
self.0.to_ne_bytes() |
| 533 |
} |
| 534 |
|
| 535 |
/// Returns an iterator over all values from 0 up to and not |
| 536 |
/// including the given length. |
| 537 |
/// |
| 538 |
/// If the given length exceeds this type's limit, then this |
| 539 |
/// panics. |
| 540 |
pub(crate) fn iter(len: usize) -> $iter { |
| 541 |
$iter::new(len) |
| 542 |
} |
| 543 |
} |
| 544 |
|
| 545 |
// We write our own Debug impl so that we get things like PatternID(5) |
| 546 |
// instead of PatternID(SmallIndex(5)). |
| 547 |
impl core::fmt::Debug for $name { |
| 548 |
fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result { |
| 549 |
f.debug_tuple(stringify!($name)).field(&self.as_u32()).finish() |
| 550 |
} |
| 551 |
} |
| 552 |
|
| 553 |
impl<T> core::ops::Index<$name> for [T] { |
| 554 |
type Output = T; |
| 555 |
|
| 556 |
#[inline] |
| 557 |
fn index(&self, index: $name) -> &T { |
| 558 |
&self[index.as_usize()] |
| 559 |
} |
| 560 |
} |
| 561 |
|
| 562 |
impl<T> core::ops::IndexMut<$name> for [T] { |
| 563 |
#[inline] |
| 564 |
fn index_mut(&mut self, index: $name) -> &mut T { |
| 565 |
&mut self[index.as_usize()] |
| 566 |
} |
| 567 |
} |
| 568 |
|
| 569 |
#[cfg(feature = "alloc")] |
| 570 |
impl<T> core::ops::Index<$name> for Vec<T> { |
| 571 |
type Output = T; |
| 572 |
|
| 573 |
#[inline] |
| 574 |
fn index(&self, index: $name) -> &T { |
| 575 |
&self[index.as_usize()] |
| 576 |
} |
| 577 |
} |
| 578 |
|
| 579 |
#[cfg(feature = "alloc")] |
| 580 |
impl<T> core::ops::IndexMut<$name> for Vec<T> { |
| 581 |
#[inline] |
| 582 |
fn index_mut(&mut self, index: $name) -> &mut T { |
| 583 |
&mut self[index.as_usize()] |
| 584 |
} |
| 585 |
} |
| 586 |
|
| 587 |
impl From<u8> for $name { |
| 588 |
fn from(value: u8) -> $name { |
| 589 |
$name(SmallIndex::from(value)) |
| 590 |
} |
| 591 |
} |
| 592 |
|
| 593 |
impl TryFrom<u16> for $name { |
| 594 |
type Error = $err; |
| 595 |
|
| 596 |
fn try_from(value: u16) -> Result<$name, $err> { |
| 597 |
SmallIndex::try_from(value).map($name).map_err($err) |
| 598 |
} |
| 599 |
} |
| 600 |
|
| 601 |
impl TryFrom<u32> for $name { |
| 602 |
type Error = $err; |
| 603 |
|
| 604 |
fn try_from(value: u32) -> Result<$name, $err> { |
| 605 |
SmallIndex::try_from(value).map($name).map_err($err) |
| 606 |
} |
| 607 |
} |
| 608 |
|
| 609 |
impl TryFrom<u64> for $name { |
| 610 |
type Error = $err; |
| 611 |
|
| 612 |
fn try_from(value: u64) -> Result<$name, $err> { |
| 613 |
SmallIndex::try_from(value).map($name).map_err($err) |
| 614 |
} |
| 615 |
} |
| 616 |
|
| 617 |
impl TryFrom<usize> for $name { |
| 618 |
type Error = $err; |
| 619 |
|
| 620 |
fn try_from(value: usize) -> Result<$name, $err> { |
| 621 |
SmallIndex::try_from(value).map($name).map_err($err) |
| 622 |
} |
| 623 |
} |
| 624 |
|
| 625 |
#[cfg(test)] |
| 626 |
impl quickcheck::Arbitrary for $name { |
| 627 |
fn arbitrary(gen: &mut quickcheck::Gen) -> $name { |
| 628 |
$name(SmallIndex::arbitrary(gen)) |
| 629 |
} |
| 630 |
} |
| 631 |
|
| 632 |
/// This error occurs when a value could not be constructed. |
| 633 |
/// |
| 634 |
/// This occurs when given an integer exceeding the maximum allowed |
| 635 |
/// value. |
| 636 |
/// |
| 637 |
/// When the `std` feature is enabled, this implements the `Error` |
| 638 |
/// trait. |
| 639 |
#[derive(Clone, Debug, Eq, PartialEq)] |
| 640 |
pub struct $err(SmallIndexError); |
| 641 |
|
| 642 |
impl $err { |
| 643 |
/// Returns the value that could not be converted to an ID. |
| 644 |
pub fn attempted(&self) -> u64 { |
| 645 |
self.0.attempted() |
| 646 |
} |
| 647 |
} |
| 648 |
|
| 649 |
#[cfg(feature = "std")] |
| 650 |
impl std::error::Error for $err {} |
| 651 |
|
| 652 |
impl core::fmt::Display for $err { |
| 653 |
fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result { |
| 654 |
write!( |
| 655 |
f, |
| 656 |
"failed to create {} from {:?}, which exceeds {:?}", |
| 657 |
stringify!($name), |
| 658 |
self.attempted(), |
| 659 |
$name::MAX, |
| 660 |
) |
| 661 |
} |
| 662 |
} |
| 663 |
|
| 664 |
#[derive(Clone, Debug)] |
| 665 |
pub(crate) struct $iter(SmallIndexIter); |
| 666 |
|
| 667 |
impl $iter { |
| 668 |
fn new(len: usize) -> $iter { |
| 669 |
assert!( |
| 670 |
len <= $name::LIMIT, |
| 671 |
"cannot create iterator for {} when number of \ |
| 672 |
elements exceed {:?}", |
| 673 |
stringify!($name), |
| 674 |
$name::LIMIT, |
| 675 |
); |
| 676 |
$iter(SmallIndexIter { rng: 0..len }) |
| 677 |
} |
| 678 |
} |
| 679 |
|
| 680 |
impl Iterator for $iter { |
| 681 |
type Item = $name; |
| 682 |
|
| 683 |
fn next(&mut self) -> Option<$name> { |
| 684 |
self.0.next().map($name) |
| 685 |
} |
| 686 |
} |
| 687 |
|
| 688 |
/// An iterator adapter that is like std::iter::Enumerate, but attaches |
| 689 |
/// small index values instead. It requires `ExactSizeIterator`. At |
| 690 |
/// construction, it ensures that the index of each element in the |
| 691 |
/// iterator is representable in the corresponding small index type. |
| 692 |
#[derive(Clone, Debug)] |
| 693 |
pub(crate) struct $withiter<I> { |
| 694 |
it: I, |
| 695 |
ids: $iter, |
| 696 |
} |
| 697 |
|
| 698 |
impl<I: Iterator + ExactSizeIterator> $withiter<I> { |
| 699 |
fn new(it: I) -> $withiter<I> { |
| 700 |
let ids = $name::iter(it.len()); |
| 701 |
$withiter { it, ids } |
| 702 |
} |
| 703 |
} |
| 704 |
|
| 705 |
impl<I: Iterator + ExactSizeIterator> Iterator for $withiter<I> { |
| 706 |
type Item = ($name, I::Item); |
| 707 |
|
| 708 |
fn next(&mut self) -> Option<($name, I::Item)> { |
| 709 |
let item = self.it.next()?; |
| 710 |
// Number of elements in this iterator must match, according |
| 711 |
// to contract of ExactSizeIterator. |
| 712 |
let id = self.ids.next().unwrap(); |
| 713 |
Some((id, item)) |
| 714 |
} |
| 715 |
} |
| 716 |
}; |
| 717 |
} |
| 718 |
|
| 719 |
/// The identifier of a regex pattern, represented by a [`SmallIndex`]. |
| 720 |
/// |
| 721 |
/// The identifier for a pattern corresponds to its relative position among |
| 722 |
/// other patterns in a single finite state machine. Namely, when building |
| 723 |
/// a multi-pattern regex engine, one must supply a sequence of patterns to |
| 724 |
/// match. The position (starting at 0) of each pattern in that sequence |
| 725 |
/// represents its identifier. This identifier is in turn used to identify and |
| 726 |
/// report matches of that pattern in various APIs. |
| 727 |
/// |
| 728 |
/// See the [`SmallIndex`] type for more information about what it means for |
| 729 |
/// a pattern ID to be a "small index." |
| 730 |
/// |
| 731 |
/// Note that this type is defined in the |
| 732 |
/// [`util::primitives`](crate::util::primitives) module, but it is also |
| 733 |
/// re-exported at the crate root due to how common it is. |
| 734 |
#[derive(Clone, Copy, Default, Eq, Hash, PartialEq, PartialOrd, Ord)] |
| 735 |
#[repr(transparent)] |
| 736 |
pub struct PatternID(SmallIndex); |
| 737 |
|
| 738 |
/// The identifier of a finite automaton state, represented by a |
| 739 |
/// [`SmallIndex`]. |
| 740 |
/// |
| 741 |
/// Most regex engines in this crate are built on top of finite automata. Each |
| 742 |
/// state in a finite automaton defines transitions from its state to another. |
| 743 |
/// Those transitions point to other states via their identifiers, i.e., a |
| 744 |
/// `StateID`. Since finite automata tend to contain many transitions, it is |
| 745 |
/// much more memory efficient to define state IDs as small indices. |
| 746 |
/// |
| 747 |
/// See the [`SmallIndex`] type for more information about what it means for |
| 748 |
/// a state ID to be a "small index." |
| 749 |
#[derive(Clone, Copy, Default, Eq, Hash, PartialEq, PartialOrd, Ord)] |
| 750 |
#[repr(transparent)] |
| 751 |
pub struct StateID(SmallIndex); |
| 752 |
|
| 753 |
index_type_impls!(PatternID, PatternIDError, PatternIDIter, WithPatternIDIter); |
| 754 |
index_type_impls!(StateID, StateIDError, StateIDIter, WithStateIDIter); |
| 755 |
|
| 756 |
/// A utility trait that defines a couple of adapters for making it convenient |
| 757 |
/// to access indices as "small index" types. We require ExactSizeIterator so |
| 758 |
/// that iterator construction can do a single check to make sure the index of |
| 759 |
/// each element is representable by its small index type. |
| 760 |
pub(crate) trait IteratorIndexExt: Iterator { |
| 761 |
fn with_pattern_ids(self) -> WithPatternIDIter<Self> |
| 762 |
where |
| 763 |
Self: Sized + ExactSizeIterator, |
| 764 |
{ |
| 765 |
WithPatternIDIter::new(self) |
| 766 |
} |
| 767 |
|
| 768 |
fn with_state_ids(self) -> WithStateIDIter<Self> |
| 769 |
where |
| 770 |
Self: Sized + ExactSizeIterator, |
| 771 |
{ |
| 772 |
WithStateIDIter::new(self) |
| 773 |
} |
| 774 |
} |
| 775 |
|
| 776 |
impl<I: Iterator> IteratorIndexExt for I {} |