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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691
// This file is part of ICU4X. For terms of use, please see the file
// called LICENSE at the top level of the ICU4X source tree
// (online at: https://github.com/unicode-org/icu4x/blob/main/LICENSE ).
use alloc::borrow::Cow;
use alloc::string::String;
use alloc::vec::Vec;
use icu_provider::prelude::*;
use zerotrie::cursor::ZeroAsciiIgnoreCaseTrieCursor;
use crate::{
provider::names::{
Bcp47ToIanaMapV1, Bcp47ToIanaMapV1Marker, IanaToBcp47MapV2, IanaToBcp47MapV2Marker,
},
TimeZoneBcp47Id, TimeZoneError,
};
/// A mapper between IANA time zone identifiers and BCP-47 time zone identifiers.
///
/// This mapper supports two-way mapping, but it is optimized for the case of IANA to BCP-47.
/// It also supports normalizing and canonicalizing the IANA strings.
///
/// There are approximately 600 IANA identifiers and 450 BCP-47 identifiers.
///
/// BCP-47 time zone identifiers are 8 ASCII characters or less and currently
/// average 5.1 characters long. Current IANA time zone identifiers are less than
/// 40 ASCII characters and average 14.2 characters long.
///
/// These lists grow very slowly; in a typical year, 2-3 new identifiers are added.
///
/// # Normalization vs Canonicalization
///
/// Multiple IANA time zone identifiers can refer to the same BCP-47 time zone. For example, the
/// following three IANA identifiers all map to `"usind"`:
///
/// - "America/Fort_Wayne"
/// - "America/Indiana/Indianapolis"
/// - "America/Indianapolis"
/// - "US/East-Indiana"
///
/// There is only one canonical identifier, which is "America/Indiana/Indianapolis". The
/// *canonicalization* operation returns the canonical identifier. You should canonicalize if
/// you need to compare time zones for equality. Note that the canonical identifier can change
/// over time. For example, the identifier "Europe/Kiev" was renamed to the newly-added
/// identifier "Europe/Kyiv" in 2022.
///
/// The *normalization* operation, on the other hand, keeps the input identifier but normalizes
/// the casing. For example, "AMERICA/FORT_WAYNE" normalizes to "America/Fort_Wayne".
/// Normalization is a data-driven operation because there are no algorithmic casing rules that
/// work for all IANA time zone identifiers.
///
/// Normalization is a cheap operation, but canonicalization might be expensive, since it might
/// require searching over all IANA IDs to find the canonicalization. If you need
/// canonicalization that is reliably fast, use [`TimeZoneIdMapperWithFastCanonicalization`].
///
/// # Examples
///
/// ```
/// use icu::timezone::TimeZoneIdMapper;
///
/// let mapper = TimeZoneIdMapper::new();
/// let mapper = mapper.as_borrowed();
///
/// // The IANA zone "Australia/Melbourne" is the BCP-47 zone "aumel":
/// assert_eq!(
/// mapper.iana_to_bcp47("Australia/Melbourne"),
/// Some("aumel".parse().unwrap())
/// );
///
/// // Lookup is ASCII-case-insensitive:
/// assert_eq!(
/// mapper.iana_to_bcp47("australia/melbourne"),
/// Some("aumel".parse().unwrap())
/// );
///
/// // The IANA zone "Australia/Victoria" is an alias:
/// assert_eq!(
/// mapper.iana_to_bcp47("Australia/Victoria"),
/// Some("aumel".parse().unwrap())
/// );
///
/// // We can recover the canonical identifier from the mapper:
/// assert_eq!(
/// mapper.canonicalize_iana("Australia/Victoria").unwrap().0,
/// "Australia/Melbourne"
/// );
/// ```
#[derive(Debug, Clone)]
pub struct TimeZoneIdMapper {
data: DataPayload<IanaToBcp47MapV2Marker>,
}
#[cfg(feature = "compiled_data")]
impl Default for TimeZoneIdMapper {
fn default() -> Self {
Self::new()
}
}
impl TimeZoneIdMapper {
/// Creates a new [`TimeZoneIdMapper`] using compiled data.
///
/// See [`TimeZoneIdMapper`] for an example.
///
/// ✨ *Enabled with the `compiled_data` Cargo feature.*
///
/// [📚 Help choosing a constructor](icu_provider::constructors)
#[cfg(feature = "compiled_data")]
pub fn new() -> Self {
Self {
data: DataPayload::from_static_ref(
crate::provider::Baked::SINGLETON_TIME_ZONE_IANA_TO_BCP47_V2,
),
}
}
icu_provider::gen_any_buffer_data_constructors!(locale: skip, options: skip, error: DataError,
#[cfg(skip)]
functions: [
new,
try_new_with_any_provider,
try_new_with_buffer_provider,
try_new_unstable,
Self,
]
);
#[doc = icu_provider::gen_any_buffer_unstable_docs!(UNSTABLE, Self::new)]
pub fn try_new_unstable<P>(provider: &P) -> Result<Self, DataError>
where
P: DataProvider<IanaToBcp47MapV2Marker> + ?Sized,
{
let data = provider.load(Default::default())?.take_payload()?;
Ok(Self { data })
}
/// Returns a borrowed version of the mapper that can be queried.
///
/// This avoids a small potential indirection cost when querying the mapper.
pub fn as_borrowed(&self) -> TimeZoneIdMapperBorrowed {
TimeZoneIdMapperBorrowed {
data: self.data.get(),
}
}
}
impl AsRef<TimeZoneIdMapper> for TimeZoneIdMapper {
#[inline]
fn as_ref(&self) -> &TimeZoneIdMapper {
self
}
}
/// A borrowed wrapper around the time zone ID mapper, returned by
/// [`TimeZoneIdMapper::as_borrowed()`]. More efficient to query.
#[derive(Debug, Copy, Clone)]
pub struct TimeZoneIdMapperBorrowed<'a> {
data: &'a IanaToBcp47MapV2<'a>,
}
impl<'a> TimeZoneIdMapperBorrowed<'a> {
/// Gets the BCP-47 time zone ID from an IANA time zone ID
/// with a case-insensitive lookup.
///
/// Returns `None` if the IANA ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapper;
///
/// let mapper = TimeZoneIdMapper::new();
/// let mapper = mapper.as_borrowed();
///
/// let result = mapper.iana_to_bcp47("Asia/CALCUTTA").unwrap();
///
/// assert_eq!(*result, "inccu");
///
/// // Unknown IANA time zone ID:
/// assert_eq!(mapper.iana_to_bcp47("America/San_Francisco"), None);
/// ```
pub fn iana_to_bcp47(&self, iana_id: &str) -> Option<TimeZoneBcp47Id> {
self.iana_lookup_quick(iana_id)
.and_then(|trie_value| self.data.bcp47_ids.get(trie_value.index()))
}
/// Same as [`Self::iana_to_bcp47()`] but works with potentially ill-formed UTF-8.
pub fn iana_bytes_to_bcp47(&self, iana_id: &[u8]) -> Option<TimeZoneBcp47Id> {
self.iana_lookup_quick(iana_id)
.and_then(|trie_value| self.data.bcp47_ids.get(trie_value.index()))
}
/// Normalizes the syntax of an IANA time zone ID.
///
/// Also returns the BCP-47 time zone ID.
///
/// Returns `None` if the IANA ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapper;
/// use std::borrow::Cow;
///
/// let mapper = TimeZoneIdMapper::new();
/// let mapper = mapper.as_borrowed();
///
/// let result = mapper.normalize_iana("Asia/CALCUTTA").unwrap();
///
/// assert_eq!(result.0, "Asia/Calcutta");
/// assert!(matches!(result.0, Cow::Owned(_)));
/// assert_eq!(*result.1, "inccu");
///
/// // Borrows when able:
/// let result = mapper.normalize_iana("America/Chicago").unwrap();
/// assert_eq!(result.0, "America/Chicago");
/// assert!(matches!(result.0, Cow::Borrowed(_)));
///
/// // Unknown IANA time zone ID:
/// assert_eq!(mapper.normalize_iana("America/San_Francisco"), None);
/// ```
pub fn normalize_iana<'s>(&self, iana_id: &'s str) -> Option<(Cow<'s, str>, TimeZoneBcp47Id)> {
let (trie_value, string) = self.iana_lookup_with_normalization(iana_id, |_| {})?;
let Some(bcp47_id) = self.data.bcp47_ids.get(trie_value.index()) else {
debug_assert!(false, "index should be in range");
return None;
};
Some((string, bcp47_id))
}
/// Returns the canonical, normalized identifier of the given IANA time zone.
///
/// Also returns the BCP-47 time zone ID.
///
/// Returns `None` if the IANA ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapper;
/// use std::borrow::Cow;
///
/// let mapper = TimeZoneIdMapper::new();
/// let mapper = mapper.as_borrowed();
///
/// let result = mapper.canonicalize_iana("Asia/CALCUTTA").unwrap();
///
/// assert_eq!(result.0, "Asia/Kolkata");
/// assert!(matches!(result.0, Cow::Owned(_)));
/// assert_eq!(*result.1, "inccu");
///
/// // Borrows when able:
/// let result = mapper.canonicalize_iana("America/Chicago").unwrap();
/// assert_eq!(result.0, "America/Chicago");
/// assert!(matches!(result.0, Cow::Borrowed(_)));
///
/// // Unknown IANA time zone ID:
/// assert_eq!(mapper.canonicalize_iana("America/San_Francisco"), None);
/// ```
pub fn canonicalize_iana<'s>(
&self,
iana_id: &'s str,
) -> Option<(Cow<'s, str>, TimeZoneBcp47Id)> {
// Note: We collect the cursors into a stack so that we start probing
// nearby the input IANA identifier. This should improve lookup time since
// most renames share the same prefix like "Asia" or "Europe".
let mut stack = Vec::with_capacity(iana_id.len());
let (trie_value, string) = self.iana_lookup_with_normalization(iana_id, |cursor| {
stack.push((cursor.clone(), 0, 1));
})?;
let Some(bcp47_id) = self.data.bcp47_ids.get(trie_value.index()) else {
debug_assert!(false, "index should be in range");
return None;
};
if trie_value.is_canonical() {
return Some((string, bcp47_id));
}
// If we get here, we need to walk the trie to find the canonical IANA ID.
let needle = trie_value.to_canonical();
let Some(string) = self.iana_search(needle, string.into_owned(), stack) else {
debug_assert!(false, "every time zone should have a canonical IANA ID");
return None;
};
Some((Cow::Owned(string), bcp47_id))
}
/// Returns the canonical, normalized IANA ID of the given BCP-47 ID.
///
/// This function performs a linear search over all IANA IDs. If this is problematic, consider one of the
/// following functions instead:
///
/// 1. [`TimeZoneIdMapperBorrowed::canonicalize_iana()`]
/// is faster if you have an IANA ID.
/// 2. [`TimeZoneIdMapperWithFastCanonicalizationBorrowed::canonical_iana_from_bcp47()`]
/// is faster, but it requires loading additional data
/// (see [`TimeZoneIdMapperWithFastCanonicalization`]).
///
/// Returns `None` if the BCP-47 ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapper;
/// use std::borrow::Cow;
/// use tinystr::tinystr;
///
/// let mapper = TimeZoneIdMapper::new();
/// let mapper = mapper.as_borrowed();
///
/// let bcp47_id = TimeZoneBcp47Id(tinystr!(8, "inccu"));
/// let result = mapper.find_canonical_iana_from_bcp47(bcp47_id).unwrap();
///
/// assert_eq!(result, "Asia/Kolkata");
///
/// // Unknown BCP-47 time zone ID:
/// let bcp47_id = TimeZoneBcp47Id(tinystr!(8, "ussfo"));
/// assert_eq!(mapper.find_canonical_iana_from_bcp47(bcp47_id), None);
/// ```
pub fn find_canonical_iana_from_bcp47(&self, bcp47_id: TimeZoneBcp47Id) -> Option<String> {
let index = self.data.bcp47_ids.binary_search(&bcp47_id).ok()?;
let stack = alloc::vec![(self.data.map.cursor(), 0, 0)];
let needle = IanaTrieValue::canonical_for_index(index);
let string = self.iana_search(needle, String::new(), stack)?;
Some(string)
}
/// Queries the data for `iana_id` without recording the normalized string.
/// This is a fast, no-alloc lookup.
fn iana_lookup_quick(&self, iana_id: impl AsRef<[u8]>) -> Option<IanaTrieValue> {
self.data.map.get(iana_id).map(IanaTrieValue)
}
/// Queries the data for `iana_id` while keeping track of the normalized string.
/// This is a fast lookup, but it may require allocating memory.
fn iana_lookup_with_normalization<'l, 's>(
&'l self,
iana_id: &'s str,
mut cursor_fn: impl FnMut(&ZeroAsciiIgnoreCaseTrieCursor<'l>),
) -> Option<(IanaTrieValue, Cow<'s, str>)> {
let mut cursor = self.data.map.cursor();
let mut string = Cow::Borrowed(iana_id);
let mut i = 0;
let trie_value = loop {
cursor_fn(&cursor);
let Some(&input_byte) = string.as_bytes().get(i) else {
break cursor.take_value().map(IanaTrieValue);
};
let Some(matched_byte) = cursor.step(input_byte) else {
break None;
};
if matched_byte != input_byte {
// Safety: we write to input_byte farther down after performing safety checks.
let Some(input_byte) = unsafe { string.to_mut().as_bytes_mut() }.get_mut(i) else {
debug_assert!(false, "the same index was just accessed earlier");
break None;
};
if !input_byte.is_ascii() {
debug_assert!(false, "non-ASCII input byte: {input_byte}");
break None;
}
if !matched_byte.is_ascii() {
debug_assert!(false, "non-ASCII matched byte: {matched_byte}");
break None;
}
// Safety: we just checked that both input_byte and matched_byte are ASCII,
// so the buffer remains UTF-8 when we replace one with the other.
*input_byte = matched_byte;
}
i += 1;
}?;
Some((trie_value, string))
}
/// Performs a reverse lookup by walking the trie with an optional start position.
/// This is not a fast operation since it requires a linear search.
fn iana_search(
&self,
needle: IanaTrieValue,
mut string: String,
mut stack: Vec<(ZeroAsciiIgnoreCaseTrieCursor, usize, usize)>,
) -> Option<String> {
loop {
let Some((mut cursor, index, suffix_len)) = stack.pop() else {
// Nothing left in the trie.
return None;
};
// Check to see if there is a value at the current node.
if let Some(candidate) = cursor.take_value().map(IanaTrieValue) {
if candidate == needle {
// Success! Found what we were looking for.
return Some(string);
}
}
// Now check for children of the current node.
let mut sub_cursor = cursor.clone();
if let Some(probe_result) = sub_cursor.probe(index) {
// Found a child. Add the current byte edge to the string.
if !probe_result.byte.is_ascii() {
debug_assert!(false, "non-ASCII probe byte: {}", probe_result.byte);
return None;
}
// Safety: the byte being added is ASCII as guarded above
unsafe { string.as_mut_vec().push(probe_result.byte) };
// Add the child to the stack, and also add back the current
// node if there are more siblings to visit.
if index + 1 < probe_result.total_siblings as usize {
stack.push((cursor, index + 1, suffix_len));
stack.push((sub_cursor, 0, 1));
} else {
stack.push((sub_cursor, 0, suffix_len + 1));
}
} else {
// No more children. Pop this node's bytes from the string.
for _ in 0..suffix_len {
// Safety: we check that the bytes being removed are ASCII
let removed_byte = unsafe { string.as_mut_vec().pop() };
if let Some(removed_byte) = removed_byte {
if !removed_byte.is_ascii() {
debug_assert!(false, "non-ASCII removed byte: {removed_byte}");
// If we get here for some reason, `string` is not in a valid state,
// so to be extra safe, we can clear it.
string.clear();
return None;
}
} else {
debug_assert!(false, "could not remove another byte");
return None;
}
}
}
}
}
}
/// A mapper that supplements [`TimeZoneIdMapper`] with about 8 KB of additional data to
/// improve the performance of canonical IANA ID lookup.
///
/// The data in [`TimeZoneIdMapper`] is optimized for IANA to BCP-47 lookup; the reverse
/// requires a linear walk over all ~600 IANA identifiers. The data added here allows for
/// constant-time mapping from BCP-47 to IANA.
#[derive(Debug, Clone)]
pub struct TimeZoneIdMapperWithFastCanonicalization<I> {
inner: I,
data: DataPayload<Bcp47ToIanaMapV1Marker>,
}
#[cfg(feature = "compiled_data")]
impl Default for TimeZoneIdMapperWithFastCanonicalization<TimeZoneIdMapper> {
fn default() -> Self {
Self::new()
}
}
impl TimeZoneIdMapperWithFastCanonicalization<TimeZoneIdMapper> {
/// Creates a new [`TimeZoneIdMapperWithFastCanonicalization`] using compiled data.
///
/// See [`TimeZoneIdMapperWithFastCanonicalization`] for an example.
///
/// ✨ *Enabled with the `compiled_data` Cargo feature.*
///
/// [📚 Help choosing a constructor](icu_provider::constructors)
#[cfg(feature = "compiled_data")]
pub fn new() -> Self {
const _: () = assert!(
crate::provider::Baked::SINGLETON_TIME_ZONE_IANA_TO_BCP47_V2.bcp47_ids_checksum
== crate::provider::Baked::SINGLETON_TIME_ZONE_BCP47_TO_IANA_V1.bcp47_ids_checksum,
);
Self {
inner: TimeZoneIdMapper {
data: DataPayload::from_static_ref(
crate::provider::Baked::SINGLETON_TIME_ZONE_IANA_TO_BCP47_V2,
),
},
data: DataPayload::from_static_ref(
crate::provider::Baked::SINGLETON_TIME_ZONE_BCP47_TO_IANA_V1,
),
}
}
icu_provider::gen_any_buffer_data_constructors!(locale: skip, options: skip, error: TimeZoneError,
#[cfg(skip)]
functions: [
new,
try_new_with_any_provider,
try_new_with_buffer_provider,
try_new_unstable,
Self,
]
);
#[doc = icu_provider::gen_any_buffer_unstable_docs!(UNSTABLE, Self::new)]
pub fn try_new_unstable<P>(provider: &P) -> Result<Self, TimeZoneError>
where
P: DataProvider<IanaToBcp47MapV2Marker> + DataProvider<Bcp47ToIanaMapV1Marker> + ?Sized,
{
let mapper = TimeZoneIdMapper::try_new_unstable(provider)?;
Self::try_new_with_mapper_unstable(provider, mapper)
}
}
impl<I> TimeZoneIdMapperWithFastCanonicalization<I>
where
I: AsRef<TimeZoneIdMapper>,
{
/// Creates a new [`TimeZoneIdMapperWithFastCanonicalization`] using compiled data
/// and a pre-existing [`TimeZoneIdMapper`], which can be borrowed.
///
/// See [`TimeZoneIdMapperWithFastCanonicalization`] for an example.
///
/// ✨ *Enabled with the `compiled_data` Cargo feature.*
///
/// [📚 Help choosing a constructor](icu_provider::constructors)
#[cfg(feature = "compiled_data")]
pub fn try_new_with_mapper(mapper: I) -> Result<Self, TimeZoneError> {
Self {
inner: mapper,
data: DataPayload::from_static_ref(
crate::provider::Baked::SINGLETON_TIME_ZONE_BCP47_TO_IANA_V1,
),
}
.validated()
}
icu_provider::gen_any_buffer_data_constructors!(locale: skip, mapper: I, error: TimeZoneError,
#[cfg(skip)]
functions: [
try_new_with_mapper,
try_new_with_mapper_with_any_provider,
try_new_with_mapper_with_buffer_provider,
try_new_with_mapper_unstable,
Self,
]
);
#[doc = icu_provider::gen_any_buffer_unstable_docs!(UNSTABLE, Self::new)]
pub fn try_new_with_mapper_unstable<P>(provider: &P, mapper: I) -> Result<Self, TimeZoneError>
where
P: DataProvider<IanaToBcp47MapV2Marker> + DataProvider<Bcp47ToIanaMapV1Marker> + ?Sized,
{
let data = provider.load(Default::default())?.take_payload()?;
Self {
inner: mapper,
data,
}
.validated()
}
fn validated(self) -> Result<Self, TimeZoneError> {
if self.inner.as_ref().data.get().bcp47_ids_checksum != self.data.get().bcp47_ids_checksum {
return Err(TimeZoneError::MismatchedChecksums);
}
Ok(self)
}
/// Gets the inner [`TimeZoneIdMapper`] for performing queries.
pub fn inner(&self) -> &TimeZoneIdMapper {
self.inner.as_ref()
}
/// Returns a borrowed version of the mapper that can be queried.
///
/// This avoids a small potential indirection cost when querying the mapper.
pub fn as_borrowed(&self) -> TimeZoneIdMapperWithFastCanonicalizationBorrowed {
TimeZoneIdMapperWithFastCanonicalizationBorrowed {
inner: self.inner.as_ref().as_borrowed(),
data: self.data.get(),
}
}
}
/// A borrowed wrapper around the time zone ID mapper, returned by
/// [`TimeZoneIdMapperWithFastCanonicalization::as_borrowed()`]. More efficient to query.
#[derive(Debug, Copy, Clone)]
pub struct TimeZoneIdMapperWithFastCanonicalizationBorrowed<'a> {
inner: TimeZoneIdMapperBorrowed<'a>,
data: &'a Bcp47ToIanaMapV1<'a>,
}
impl<'a> TimeZoneIdMapperWithFastCanonicalizationBorrowed<'a> {
/// Gets the inner [`TimeZoneIdMapperBorrowed`] for performing queries.
pub fn inner(&self) -> TimeZoneIdMapperBorrowed<'a> {
self.inner
}
/// Returns the canonical, normalized identifier of the given IANA time zone.
///
/// Also returns the BCP-47 time zone ID.
///
/// This is a faster version of [`TimeZoneIdMapperBorrowed::canonicalize_iana()`]
/// and it always returns borrowed IANA strings, but it requires loading additional data
/// (see [`TimeZoneIdMapperWithFastCanonicalization`]).
///
/// Returns `None` if the IANA ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapperWithFastCanonicalization;
/// use std::borrow::Cow;
///
/// let mapper = TimeZoneIdMapperWithFastCanonicalization::new();
/// let mapper = mapper.as_borrowed();
///
/// let result = mapper.canonicalize_iana("Asia/CALCUTTA").unwrap();
///
/// // The Cow is always returned borrowed:
/// assert_eq!(result.0, "Asia/Kolkata");
/// assert_eq!(*result.1, "inccu");
///
/// // Unknown IANA time zone ID:
/// assert_eq!(mapper.canonicalize_iana("America/San_Francisco"), None);
/// ```
pub fn canonicalize_iana(&self, iana_id: &str) -> Option<(&str, TimeZoneBcp47Id)> {
let trie_value = self.inner.iana_lookup_quick(iana_id)?;
let Some(bcp47_id) = self.inner.data.bcp47_ids.get(trie_value.index()) else {
debug_assert!(false, "index should be in range");
return None;
};
let Some(string) = self.data.canonical_iana_ids.get(trie_value.index()) else {
debug_assert!(false, "index should be in range");
return None;
};
Some((string, bcp47_id))
}
/// Returns the canonical, normalized IANA ID of the given BCP-47 ID.
///
/// This is a faster version of [`TimeZoneIdMapperBorrowed::find_canonical_iana_from_bcp47()`]
/// and it always returns borrowed IANA strings, but it requires loading additional data
/// (see [`TimeZoneIdMapperWithFastCanonicalization`]).
///
/// Returns `None` if the BCP-47 ID is not found.
///
/// # Examples
///
/// ```
/// use icu_timezone::TimeZoneBcp47Id;
/// use icu_timezone::TimeZoneIdMapperWithFastCanonicalization;
/// use std::borrow::Cow;
/// use tinystr::tinystr;
///
/// let mapper = TimeZoneIdMapperWithFastCanonicalization::new();
/// let mapper = mapper.as_borrowed();
///
/// let bcp47_id = TimeZoneBcp47Id(tinystr!(8, "inccu"));
/// let result = mapper.canonical_iana_from_bcp47(bcp47_id).unwrap();
///
/// // The Cow is always returned borrowed:
/// assert_eq!(result, "Asia/Kolkata");
///
/// // Unknown BCP-47 time zone ID:
/// let bcp47_id = TimeZoneBcp47Id(tinystr!(8, "ussfo"));
/// assert_eq!(mapper.canonical_iana_from_bcp47(bcp47_id), None);
/// ```
pub fn canonical_iana_from_bcp47(&self, bcp47_id: TimeZoneBcp47Id) -> Option<&str> {
let index = self.inner.data.bcp47_ids.binary_search(&bcp47_id).ok()?;
let Some(string) = self.data.canonical_iana_ids.get(index) else {
debug_assert!(false, "index should be in range");
return None;
};
Some(string)
}
}
#[derive(Copy, Clone, PartialEq, Eq)]
#[repr(transparent)]
struct IanaTrieValue(usize);
impl IanaTrieValue {
#[inline]
pub(crate) fn to_canonical(self) -> Self {
Self(self.0 | 1)
}
#[inline]
pub(crate) fn canonical_for_index(index: usize) -> Self {
Self(index << 1).to_canonical()
}
#[inline]
pub(crate) fn index(self) -> usize {
self.0 >> 1
}
#[inline]
pub(crate) fn is_canonical(self) -> bool {
(self.0 & 0x1) != 0
}
}