#![allow(dead_code)]
#[cfg(test)]
mod tests;
use core::char::{encode_utf16_raw, encode_utf8_raw};
use core::str::next_code_point;
use crate::borrow::Cow;
use crate::collections::TryReserveError;
use crate::fmt;
use crate::hash::{Hash, Hasher};
use crate::iter::FusedIterator;
use crate::mem;
use crate::ops;
use crate::rc::Rc;
use crate::slice;
use crate::str;
use crate::sync::Arc;
use crate::sys_common::AsInner;
const UTF8_REPLACEMENT_CHARACTER: &str = "\u{FFFD}";
#[derive(Eq, PartialEq, Ord, PartialOrd, Clone, Copy)]
pub struct CodePoint {
value: u32,
}
impl fmt::Debug for CodePoint {
#[inline]
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
write!(formatter, "U+{:04X}", self.value)
}
}
impl CodePoint {
#[inline]
pub unsafe fn from_u32_unchecked(value: u32) -> CodePoint {
CodePoint { value }
}
#[inline]
pub fn from_u32(value: u32) -> Option<CodePoint> {
match value {
0..=0x10FFFF => Some(CodePoint { value }),
_ => None,
}
}
#[inline]
pub fn from_char(value: char) -> CodePoint {
CodePoint { value: value as u32 }
}
#[inline]
pub fn to_u32(&self) -> u32 {
self.value
}
#[inline]
pub fn to_lead_surrogate(&self) -> Option<u16> {
match self.value {
lead @ 0xD800..=0xDBFF => Some(lead as u16),
_ => None,
}
}
#[inline]
pub fn to_trail_surrogate(&self) -> Option<u16> {
match self.value {
trail @ 0xDC00..=0xDFFF => Some(trail as u16),
_ => None,
}
}
#[inline]
pub fn to_char(&self) -> Option<char> {
match self.value {
0xD800..=0xDFFF => None,
_ => Some(unsafe { char::from_u32_unchecked(self.value) }),
}
}
#[inline]
pub fn to_char_lossy(&self) -> char {
self.to_char().unwrap_or('\u{FFFD}')
}
}
#[derive(Eq, PartialEq, Ord, PartialOrd, Clone)]
pub struct Wtf8Buf {
bytes: Vec<u8>,
is_known_utf8: bool,
}
impl ops::Deref for Wtf8Buf {
type Target = Wtf8;
fn deref(&self) -> &Wtf8 {
self.as_slice()
}
}
impl ops::DerefMut for Wtf8Buf {
fn deref_mut(&mut self) -> &mut Wtf8 {
self.as_mut_slice()
}
}
impl fmt::Debug for Wtf8Buf {
#[inline]
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
fmt::Debug::fmt(&**self, formatter)
}
}
impl Wtf8Buf {
#[inline]
pub fn new() -> Wtf8Buf {
Wtf8Buf { bytes: Vec::new(), is_known_utf8: true }
}
#[inline]
pub fn with_capacity(capacity: usize) -> Wtf8Buf {
Wtf8Buf { bytes: Vec::with_capacity(capacity), is_known_utf8: true }
}
#[inline]
pub fn from_string(string: String) -> Wtf8Buf {
Wtf8Buf { bytes: string.into_bytes(), is_known_utf8: true }
}
#[inline]
pub fn from_str(str: &str) -> Wtf8Buf {
Wtf8Buf { bytes: <[_]>::to_vec(str.as_bytes()), is_known_utf8: true }
}
pub fn clear(&mut self) {
self.bytes.clear();
self.is_known_utf8 = true;
}
pub fn from_wide(v: &[u16]) -> Wtf8Buf {
let mut string = Wtf8Buf::with_capacity(v.len());
for item in char::decode_utf16(v.iter().cloned()) {
match item {
Ok(ch) => string.push_char(ch),
Err(surrogate) => {
let surrogate = surrogate.unpaired_surrogate();
let code_point = unsafe { CodePoint::from_u32_unchecked(surrogate as u32) };
string.is_known_utf8 = false;
string.push_code_point_unchecked(code_point);
}
}
}
string
}
fn push_code_point_unchecked(&mut self, code_point: CodePoint) {
let mut bytes = [0; 4];
let bytes = encode_utf8_raw(code_point.value, &mut bytes);
self.bytes.extend_from_slice(bytes)
}
#[inline]
pub fn as_slice(&self) -> &Wtf8 {
unsafe { Wtf8::from_bytes_unchecked(&self.bytes) }
}
#[inline]
pub fn as_mut_slice(&mut self) -> &mut Wtf8 {
unsafe { Wtf8::from_mut_bytes_unchecked(&mut self.bytes) }
}
#[inline]
pub fn reserve(&mut self, additional: usize) {
self.bytes.reserve(additional)
}
#[inline]
pub fn try_reserve(&mut self, additional: usize) -> Result<(), TryReserveError> {
self.bytes.try_reserve(additional)
}
#[inline]
pub fn reserve_exact(&mut self, additional: usize) {
self.bytes.reserve_exact(additional)
}
#[inline]
pub fn try_reserve_exact(&mut self, additional: usize) -> Result<(), TryReserveError> {
self.bytes.try_reserve_exact(additional)
}
#[inline]
pub fn shrink_to_fit(&mut self) {
self.bytes.shrink_to_fit()
}
#[inline]
pub fn shrink_to(&mut self, min_capacity: usize) {
self.bytes.shrink_to(min_capacity)
}
#[inline]
pub fn capacity(&self) -> usize {
self.bytes.capacity()
}
#[inline]
pub fn push_str(&mut self, other: &str) {
self.bytes.extend_from_slice(other.as_bytes())
}
#[inline]
pub fn push_wtf8(&mut self, other: &Wtf8) {
match ((&*self).final_lead_surrogate(), other.initial_trail_surrogate()) {
(Some(lead), Some(trail)) => {
let len_without_lead_surrogate = self.len() - 3;
self.bytes.truncate(len_without_lead_surrogate);
let other_without_trail_surrogate = &other.bytes[3..];
self.bytes.reserve(4 + other_without_trail_surrogate.len());
self.push_char(decode_surrogate_pair(lead, trail));
self.bytes.extend_from_slice(other_without_trail_surrogate);
}
_ => {
if other.next_surrogate(0).is_some() {
self.is_known_utf8 = false;
}
self.bytes.extend_from_slice(&other.bytes);
}
}
}
#[inline]
pub fn push_char(&mut self, c: char) {
self.push_code_point_unchecked(CodePoint::from_char(c))
}
#[inline]
pub fn push(&mut self, code_point: CodePoint) {
if let Some(trail) = code_point.to_trail_surrogate() {
if let Some(lead) = (&*self).final_lead_surrogate() {
let len_without_lead_surrogate = self.len() - 3;
self.bytes.truncate(len_without_lead_surrogate);
self.push_char(decode_surrogate_pair(lead, trail));
return;
}
self.is_known_utf8 = false;
} else if code_point.to_lead_surrogate().is_some() {
self.is_known_utf8 = false;
}
self.push_code_point_unchecked(code_point)
}
#[inline]
pub fn truncate(&mut self, new_len: usize) {
assert!(is_code_point_boundary(self, new_len));
self.bytes.truncate(new_len)
}
pub fn into_string(self) -> Result<String, Wtf8Buf> {
if self.is_known_utf8 || self.next_surrogate(0).is_none() {
Ok(unsafe { String::from_utf8_unchecked(self.bytes) })
} else {
Err(self)
}
}
pub fn into_string_lossy(mut self) -> String {
if self.is_known_utf8 {
return unsafe { String::from_utf8_unchecked(self.bytes) };
}
let mut pos = 0;
loop {
match self.next_surrogate(pos) {
Some((surrogate_pos, _)) => {
pos = surrogate_pos + 3;
self.bytes[surrogate_pos..pos]
.copy_from_slice(UTF8_REPLACEMENT_CHARACTER.as_bytes());
}
None => return unsafe { String::from_utf8_unchecked(self.bytes) },
}
}
}
#[inline]
pub fn into_box(self) -> Box<Wtf8> {
unsafe { mem::transmute(self.bytes.into_boxed_slice()) }
}
pub fn from_box(boxed: Box<Wtf8>) -> Wtf8Buf {
let bytes: Box<[u8]> = unsafe { mem::transmute(boxed) };
Wtf8Buf { bytes: bytes.into_vec(), is_known_utf8: false }
}
}
impl FromIterator<CodePoint> for Wtf8Buf {
fn from_iter<T: IntoIterator<Item = CodePoint>>(iter: T) -> Wtf8Buf {
let mut string = Wtf8Buf::new();
string.extend(iter);
string
}
}
impl Extend<CodePoint> for Wtf8Buf {
fn extend<T: IntoIterator<Item = CodePoint>>(&mut self, iter: T) {
let iterator = iter.into_iter();
let (low, _high) = iterator.size_hint();
self.bytes.reserve(low);
iterator.for_each(move |code_point| self.push(code_point));
}
#[inline]
fn extend_one(&mut self, code_point: CodePoint) {
self.push(code_point);
}
#[inline]
fn extend_reserve(&mut self, additional: usize) {
self.bytes.reserve(additional);
}
}
#[derive(Eq, Ord, PartialEq, PartialOrd)]
pub struct Wtf8 {
bytes: [u8],
}
impl AsInner<[u8]> for Wtf8 {
#[inline]
fn as_inner(&self) -> &[u8] {
&self.bytes
}
}
impl fmt::Debug for Wtf8 {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
fn write_str_escaped(f: &mut fmt::Formatter<'_>, s: &str) -> fmt::Result {
use crate::fmt::Write;
for c in s.chars().flat_map(|c| c.escape_debug()) {
f.write_char(c)?
}
Ok(())
}
formatter.write_str("\"")?;
let mut pos = 0;
while let Some((surrogate_pos, surrogate)) = self.next_surrogate(pos) {
write_str_escaped(formatter, unsafe {
str::from_utf8_unchecked(&self.bytes[pos..surrogate_pos])
})?;
write!(formatter, "\\u{{{:x}}}", surrogate)?;
pos = surrogate_pos + 3;
}
write_str_escaped(formatter, unsafe { str::from_utf8_unchecked(&self.bytes[pos..]) })?;
formatter.write_str("\"")
}
}
impl fmt::Display for Wtf8 {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
let wtf8_bytes = &self.bytes;
let mut pos = 0;
loop {
match self.next_surrogate(pos) {
Some((surrogate_pos, _)) => {
formatter.write_str(unsafe {
str::from_utf8_unchecked(&wtf8_bytes[pos..surrogate_pos])
})?;
formatter.write_str(UTF8_REPLACEMENT_CHARACTER)?;
pos = surrogate_pos + 3;
}
None => {
let s = unsafe { str::from_utf8_unchecked(&wtf8_bytes[pos..]) };
if pos == 0 { return s.fmt(formatter) } else { return formatter.write_str(s) }
}
}
}
}
}
impl Wtf8 {
#[inline]
pub fn from_str(value: &str) -> &Wtf8 {
unsafe { Wtf8::from_bytes_unchecked(value.as_bytes()) }
}
#[inline]
unsafe fn from_bytes_unchecked(value: &[u8]) -> &Wtf8 {
mem::transmute(value)
}
#[inline]
unsafe fn from_mut_bytes_unchecked(value: &mut [u8]) -> &mut Wtf8 {
mem::transmute(value)
}
#[inline]
pub fn len(&self) -> usize {
self.bytes.len()
}
#[inline]
pub fn is_empty(&self) -> bool {
self.bytes.is_empty()
}
#[inline]
pub fn ascii_byte_at(&self, position: usize) -> u8 {
match self.bytes[position] {
ascii_byte @ 0x00..=0x7F => ascii_byte,
_ => 0xFF,
}
}
#[inline]
pub fn code_points(&self) -> Wtf8CodePoints<'_> {
Wtf8CodePoints { bytes: self.bytes.iter() }
}
#[inline]
pub fn as_str(&self) -> Option<&str> {
match self.next_surrogate(0) {
None => Some(unsafe { str::from_utf8_unchecked(&self.bytes) }),
Some(_) => None,
}
}
pub fn to_owned(&self) -> Wtf8Buf {
Wtf8Buf { bytes: self.bytes.to_vec(), is_known_utf8: false }
}
pub fn to_string_lossy(&self) -> Cow<'_, str> {
let surrogate_pos = match self.next_surrogate(0) {
None => return Cow::Borrowed(unsafe { str::from_utf8_unchecked(&self.bytes) }),
Some((pos, _)) => pos,
};
let wtf8_bytes = &self.bytes;
let mut utf8_bytes = Vec::with_capacity(self.len());
utf8_bytes.extend_from_slice(&wtf8_bytes[..surrogate_pos]);
utf8_bytes.extend_from_slice(UTF8_REPLACEMENT_CHARACTER.as_bytes());
let mut pos = surrogate_pos + 3;
loop {
match self.next_surrogate(pos) {
Some((surrogate_pos, _)) => {
utf8_bytes.extend_from_slice(&wtf8_bytes[pos..surrogate_pos]);
utf8_bytes.extend_from_slice(UTF8_REPLACEMENT_CHARACTER.as_bytes());
pos = surrogate_pos + 3;
}
None => {
utf8_bytes.extend_from_slice(&wtf8_bytes[pos..]);
return Cow::Owned(unsafe { String::from_utf8_unchecked(utf8_bytes) });
}
}
}
}
#[inline]
pub fn encode_wide(&self) -> EncodeWide<'_> {
EncodeWide { code_points: self.code_points(), extra: 0 }
}
#[inline]
fn next_surrogate(&self, mut pos: usize) -> Option<(usize, u16)> {
let mut iter = self.bytes[pos..].iter();
loop {
let b = *iter.next()?;
if b < 0x80 {
pos += 1;
} else if b < 0xE0 {
iter.next();
pos += 2;
} else if b == 0xED {
match (iter.next(), iter.next()) {
(Some(&b2), Some(&b3)) if b2 >= 0xA0 => {
return Some((pos, decode_surrogate(b2, b3)));
}
_ => pos += 3,
}
} else if b < 0xF0 {
iter.next();
iter.next();
pos += 3;
} else {
iter.next();
iter.next();
iter.next();
pos += 4;
}
}
}
#[inline]
fn final_lead_surrogate(&self) -> Option<u16> {
match self.bytes {
[.., 0xED, b2 @ 0xA0..=0xAF, b3] => Some(decode_surrogate(b2, b3)),
_ => None,
}
}
#[inline]
fn initial_trail_surrogate(&self) -> Option<u16> {
match self.bytes {
[0xED, b2 @ 0xB0..=0xBF, b3, ..] => Some(decode_surrogate(b2, b3)),
_ => None,
}
}
pub fn clone_into(&self, buf: &mut Wtf8Buf) {
buf.is_known_utf8 = false;
self.bytes.clone_into(&mut buf.bytes);
}
#[inline]
pub fn into_box(&self) -> Box<Wtf8> {
let boxed: Box<[u8]> = self.bytes.into();
unsafe { mem::transmute(boxed) }
}
pub fn empty_box() -> Box<Wtf8> {
let boxed: Box<[u8]> = Default::default();
unsafe { mem::transmute(boxed) }
}
#[inline]
pub fn into_arc(&self) -> Arc<Wtf8> {
let arc: Arc<[u8]> = Arc::from(&self.bytes);
unsafe { Arc::from_raw(Arc::into_raw(arc) as *const Wtf8) }
}
#[inline]
pub fn into_rc(&self) -> Rc<Wtf8> {
let rc: Rc<[u8]> = Rc::from(&self.bytes);
unsafe { Rc::from_raw(Rc::into_raw(rc) as *const Wtf8) }
}
#[inline]
pub fn make_ascii_lowercase(&mut self) {
self.bytes.make_ascii_lowercase()
}
#[inline]
pub fn make_ascii_uppercase(&mut self) {
self.bytes.make_ascii_uppercase()
}
#[inline]
pub fn to_ascii_lowercase(&self) -> Wtf8Buf {
Wtf8Buf { bytes: self.bytes.to_ascii_lowercase(), is_known_utf8: false }
}
#[inline]
pub fn to_ascii_uppercase(&self) -> Wtf8Buf {
Wtf8Buf { bytes: self.bytes.to_ascii_uppercase(), is_known_utf8: false }
}
#[inline]
pub fn is_ascii(&self) -> bool {
self.bytes.is_ascii()
}
#[inline]
pub fn eq_ignore_ascii_case(&self, other: &Self) -> bool {
self.bytes.eq_ignore_ascii_case(&other.bytes)
}
}
impl ops::Index<ops::Range<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::Range<usize>) -> &Wtf8 {
if range.start <= range.end
&& is_code_point_boundary(self, range.start)
&& is_code_point_boundary(self, range.end)
{
unsafe { slice_unchecked(self, range.start, range.end) }
} else {
slice_error_fail(self, range.start, range.end)
}
}
}
impl ops::Index<ops::RangeFrom<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::RangeFrom<usize>) -> &Wtf8 {
if is_code_point_boundary(self, range.start) {
unsafe { slice_unchecked(self, range.start, self.len()) }
} else {
slice_error_fail(self, range.start, self.len())
}
}
}
impl ops::Index<ops::RangeTo<usize>> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, range: ops::RangeTo<usize>) -> &Wtf8 {
if is_code_point_boundary(self, range.end) {
unsafe { slice_unchecked(self, 0, range.end) }
} else {
slice_error_fail(self, 0, range.end)
}
}
}
impl ops::Index<ops::RangeFull> for Wtf8 {
type Output = Wtf8;
#[inline]
fn index(&self, _range: ops::RangeFull) -> &Wtf8 {
self
}
}
#[inline]
fn decode_surrogate(second_byte: u8, third_byte: u8) -> u16 {
0xD800 | (second_byte as u16 & 0x3F) << 6 | third_byte as u16 & 0x3F
}
#[inline]
fn decode_surrogate_pair(lead: u16, trail: u16) -> char {
let code_point = 0x10000 + ((((lead - 0xD800) as u32) << 10) | (trail - 0xDC00) as u32);
unsafe { char::from_u32_unchecked(code_point) }
}
#[inline]
pub fn is_code_point_boundary(slice: &Wtf8, index: usize) -> bool {
if index == slice.len() {
return true;
}
match slice.bytes.get(index) {
None => false,
Some(&b) => b < 128 || b >= 192,
}
}
#[inline]
pub unsafe fn slice_unchecked(s: &Wtf8, begin: usize, end: usize) -> &Wtf8 {
Wtf8::from_bytes_unchecked(slice::from_raw_parts(s.bytes.as_ptr().add(begin), end - begin))
}
#[inline(never)]
pub fn slice_error_fail(s: &Wtf8, begin: usize, end: usize) -> ! {
assert!(begin <= end);
panic!("index {begin} and/or {end} in `{s:?}` do not lie on character boundary");
}
#[derive(Clone)]
pub struct Wtf8CodePoints<'a> {
bytes: slice::Iter<'a, u8>,
}
impl<'a> Iterator for Wtf8CodePoints<'a> {
type Item = CodePoint;
#[inline]
fn next(&mut self) -> Option<CodePoint> {
unsafe { next_code_point(&mut self.bytes).map(|c| CodePoint { value: c }) }
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let len = self.bytes.len();
(len.saturating_add(3) / 4, Some(len))
}
}
#[stable(feature = "rust1", since = "1.0.0")]
#[derive(Clone)]
pub struct EncodeWide<'a> {
code_points: Wtf8CodePoints<'a>,
extra: u16,
}
#[stable(feature = "rust1", since = "1.0.0")]
impl<'a> Iterator for EncodeWide<'a> {
type Item = u16;
#[inline]
fn next(&mut self) -> Option<u16> {
if self.extra != 0 {
let tmp = self.extra;
self.extra = 0;
return Some(tmp);
}
let mut buf = [0; 2];
self.code_points.next().map(|code_point| {
let n = encode_utf16_raw(code_point.value, &mut buf).len();
if n == 2 {
self.extra = buf[1];
}
buf[0]
})
}
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let (low, high) = self.code_points.size_hint();
let ext = (self.extra != 0) as usize;
(low + ext, high.and_then(|n| n.checked_mul(2)).and_then(|n| n.checked_add(ext)))
}
}
#[stable(feature = "encode_wide_fused_iterator", since = "1.62.0")]
impl FusedIterator for EncodeWide<'_> {}
impl Hash for CodePoint {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
self.value.hash(state)
}
}
impl Hash for Wtf8Buf {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
state.write(&self.bytes);
0xfeu8.hash(state)
}
}
impl Hash for Wtf8 {
#[inline]
fn hash<H: Hasher>(&self, state: &mut H) {
state.write(&self.bytes);
0xfeu8.hash(state)
}
}