Source code
Revision control
Copy as Markdown
Other Tools
// 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
use crate::complex::ComplexPayloadsBorrowed;
use crate::indices::{Latin1Indices, Utf16Indices};
use crate::options::WordType;
use crate::provider::*;
use core::str::CharIndices;
use utf8_iter::Utf8CharIndices;
/// A trait allowing for RuleBreakIterator to be generalized to multiple string
/// encoding methods and granularity such as grapheme cluster, word, etc.
///
/// <div class="stab unstable">
/// 🚫 This trait is sealed; it cannot be implemented by user code. If an API requests an item that implements this
/// trait, please consider using a type from the implementors listed below.
/// </div>
pub trait RuleBreakType: crate::private::Sealed + Sized {
/// The iterator over characters.
type IterAttr<'s>: Iterator<Item = (usize, Self::CharType)> + Clone + core::fmt::Debug;
/// The character type.
type CharType: Copy + Into<u32> + core::fmt::Debug;
#[doc(hidden)]
fn char_len(ch: Self::CharType) -> usize;
}
/// Implements the [`Iterator`] trait over the segmenter boundaries of the given string.
///
/// Lifetimes:
///
/// - `'l` = lifetime of the segmenter object from which this iterator was created
/// - `'data` = lifetime of data borrowed by segmenter object
/// (this largely exists because segmenter data is invariant due to ZeroMap constraints,
/// think of it as a second 'l)
/// - `'s` = lifetime of the string being segmented
///
/// The [`Iterator::Item`] is an [`usize`] representing index of a code unit
/// _after_ the boundary (for a boundary at the end of text, this index is the length
/// of the [`str`] or array of code units).
#[derive(Debug)]
pub struct RuleBreakIterator<'data, 's, Y: RuleBreakType> {
pub(crate) iter: Y::IterAttr<'s>,
pub(crate) len: usize,
pub(crate) current_pos_data: Option<(usize, Y::CharType)>,
pub(crate) result_cache: alloc::vec::Vec<usize>,
pub(crate) data: &'data RuleBreakData<'data>,
pub(crate) complex: Option<ComplexPayloadsBorrowed<'data>>,
pub(crate) boundary_property: u8,
pub(crate) locale_override: Option<&'data RuleBreakDataOverride<'data>>,
// Should return None if there is no complex language handling
pub(crate) handle_complex_language:
fn(&mut RuleBreakIterator<'data, 's, Y>, Y::CharType) -> Option<usize>,
}
pub(crate) fn empty_handle_complex_language<Y: RuleBreakType>(
_i: &mut RuleBreakIterator<'_, '_, Y>,
_c: Y::CharType,
) -> Option<usize> {
debug_assert!(
false,
"grapheme/sentence segmenters should never need complex language handling"
);
None
}
impl<Y: RuleBreakType> Iterator for RuleBreakIterator<'_, '_, Y> {
type Item = usize;
fn next(&mut self) -> Option<Self::Item> {
// If we have break point cache by previous run, return this result
if let Some(&first_result) = self.result_cache.first() {
let mut i = 0;
loop {
if i == first_result {
self.result_cache = self.result_cache.iter().skip(1).map(|r| r - i).collect();
return self.get_current_position();
}
i += self.get_current_codepoint().map_or(0, Y::char_len);
self.advance_iter();
if self.is_eof() {
self.result_cache.clear();
self.boundary_property = self.data.complex_property;
return Some(self.len);
}
}
}
if self.is_eof() {
self.advance_iter();
if self.is_eof() && self.len == 0 {
// Empty string. Since `self.current_pos_data` is always going to be empty,
// we never read `self.len` except for here, so we can use it to mark that
// we have already returned the single empty-string breakpoint.
self.len = 1;
return Some(0);
}
let Some(right_prop) = self.get_current_break_property() else {
// iterator already reaches to EOT. Reset boundary property for word-like.
self.boundary_property = 0;
return None;
};
// SOT x anything
if matches!(
self.get_break_state_from_table(self.data.sot_property, right_prop),
BreakState::Break | BreakState::NoMatch
) {
self.boundary_property = 0; // SOT is special type
return self.get_current_position();
}
}
'a: loop {
debug_assert!(!self.is_eof());
let left_codepoint = self.get_current_codepoint()?;
let left_prop = self.get_break_property(left_codepoint);
self.advance_iter();
let Some(right_prop) = self.get_current_break_property() else {
self.boundary_property = left_prop;
return Some(self.len);
};
// Some segmenter rules doesn't have language-specific rules, we have to use LSTM (or dictionary) segmenter.
// If property is marked as SA, use it
if right_prop == self.data.complex_property {
if left_prop != self.data.complex_property {
// break before SA
self.boundary_property = left_prop;
return self.get_current_position();
}
let break_offset = (self.handle_complex_language)(self, left_codepoint);
if break_offset.is_some() {
return break_offset;
}
}
match self.get_break_state_from_table(left_prop, right_prop) {
BreakState::Keep => continue,
BreakState::Break | BreakState::NoMatch => {
self.boundary_property = left_prop;
return self.get_current_position();
}
BreakState::Index(mut index) | BreakState::Intermediate(mut index) => {
// This isn't simple rule set. We need marker to restore iterator to previous position.
let mut previous_iter = self.iter.clone();
let mut previous_pos_data = self.current_pos_data;
let mut previous_left_prop = left_prop;
loop {
self.advance_iter();
let Some(prop) = self.get_current_break_property() else {
// Reached EOF. But we are analyzing multiple characters now, so next break may be previous point.
self.boundary_property = index;
if self.get_break_state_from_table(index, self.data.eot_property)
== BreakState::NoMatch
{
self.boundary_property = previous_left_prop;
self.iter = previous_iter;
self.current_pos_data = previous_pos_data;
return self.get_current_position();
}
// EOF
return Some(self.len);
};
let previous_break_state_is_cp_prop =
index <= self.data.last_codepoint_property;
match self.get_break_state_from_table(index, prop) {
BreakState::Keep => continue 'a,
BreakState::NoMatch => {
self.boundary_property = previous_left_prop;
self.iter = previous_iter;
self.current_pos_data = previous_pos_data;
return self.get_current_position();
}
BreakState::Break => return self.get_current_position(),
BreakState::Intermediate(i) => {
index = i;
if previous_break_state_is_cp_prop {
// Move marker
previous_left_prop = index;
}
previous_iter = self.iter.clone();
previous_pos_data = self.current_pos_data;
}
BreakState::Index(i) => {
index = i;
if previous_break_state_is_cp_prop {
// Move marker
previous_iter = self.iter.clone();
previous_pos_data = self.current_pos_data;
previous_left_prop = index;
}
}
}
}
}
}
}
}
}
impl<Y: RuleBreakType> RuleBreakIterator<'_, '_, Y> {
pub(crate) fn advance_iter(&mut self) {
self.current_pos_data = self.iter.next();
}
pub(crate) fn is_eof(&self) -> bool {
self.current_pos_data.is_none()
}
pub(crate) fn get_current_break_property(&self) -> Option<u8> {
self.get_current_codepoint()
.map(|c| self.get_break_property(c))
}
pub(crate) fn get_current_position(&self) -> Option<usize> {
self.current_pos_data.map(|(pos, _)| pos)
}
pub(crate) fn get_current_codepoint(&self) -> Option<Y::CharType> {
self.current_pos_data.map(|(_, codepoint)| codepoint)
}
fn get_break_property(&self, codepoint: Y::CharType) -> u8 {
// Note: Default value is 0 == UNKNOWN
if let Some(locale_override) = &self.locale_override {
let property = locale_override
.property_table_override
.get32(codepoint.into());
if property != 0 {
return property;
}
}
self.data.property_table.get32(codepoint.into())
}
fn get_break_state_from_table(&self, left: u8, right: u8) -> BreakState {
let idx = left as usize * self.data.property_count as usize + right as usize;
// We use unwrap_or to fall back to the base case and prevent panics on bad data.
self.data
.break_state_table
.get(idx)
.unwrap_or(BreakState::Keep)
}
/// Return the status value of break boundary.
/// If segmenter isn't word, always return WordType::None
pub fn word_type(&self) -> WordType {
if !self.result_cache.is_empty() {
// Dictionary type (CJ and East Asian) is letter.
return WordType::Letter;
}
if self.boundary_property == 0 {
// break position is SOT / Any
return WordType::None;
}
self.data
.word_type_table
.get((self.boundary_property - 1) as usize)
.unwrap_or(WordType::None)
}
/// Return true when break boundary is word-like such as letter/number/CJK
/// If segmenter isn't word, return false
pub fn is_word_like(&self) -> bool {
self.word_type().is_word_like()
}
}
#[derive(Debug)]
#[non_exhaustive]
/// [`RuleBreakType`] for UTF-8 strings
pub struct Utf8;
impl crate::private::Sealed for Utf8 {}
impl RuleBreakType for Utf8 {
type IterAttr<'s> = CharIndices<'s>;
type CharType = char;
fn char_len(ch: Self::CharType) -> usize {
ch.len_utf8()
}
}
#[derive(Debug)]
#[non_exhaustive]
/// [`RuleBreakType`] for potentially ill-formed UTF-8 strings
pub struct PotentiallyIllFormedUtf8;
impl crate::private::Sealed for PotentiallyIllFormedUtf8 {}
impl RuleBreakType for PotentiallyIllFormedUtf8 {
type IterAttr<'s> = Utf8CharIndices<'s>;
type CharType = char;
fn char_len(ch: Self::CharType) -> usize {
ch.len_utf8()
}
}
#[derive(Debug)]
#[non_exhaustive]
/// [`RuleBreakType`] for Latin-1 strings
pub struct Latin1;
impl crate::private::Sealed for Latin1 {}
impl RuleBreakType for Latin1 {
type IterAttr<'s> = Latin1Indices<'s>;
type CharType = u8;
fn char_len(_ch: Self::CharType) -> usize {
unreachable!()
}
}
#[derive(Debug)]
#[non_exhaustive]
/// [`RuleBreakType`] for UTF-16 strings
pub struct Utf16;
impl crate::private::Sealed for Utf16 {}
impl RuleBreakType for Utf16 {
type IterAttr<'s> = Utf16Indices<'s>;
type CharType = u32;
fn char_len(ch: Self::CharType) -> usize {
if ch >= 0x10000 {
2
} else {
1
}
}
}