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
// Copyright 2023 Colin Finck <colin@reactos.org>
// SPDX-License-Identifier: MIT OR Apache-2.0
use core::cmp::Ordering;
use core::iter::Copied;
use core::marker::PhantomData;
use core::slice::Iter;
use core::{fmt, mem, slice};
use widestring::{U16CStr, U16Str};
use crate::error::{NtStringError, Result};
use crate::helpers::{cmp_iter, RawNtString};
use super::iter::{Chars, CharsLossy};
/// An immutable reference to a `UNICODE_STRING` (equivalent of `&str`).
///
/// See the [module-level documentation](super) for more details.
#[derive(Clone, Copy, Debug)]
#[repr(transparent)]
pub struct NtUnicodeStr<'a> {
raw: RawNtString<*const u16>,
_lifetime: PhantomData<&'a ()>,
}
impl<'a> NtUnicodeStr<'a> {
/// Returns a `*const NtUnicodeStr` pointer
/// (mainly for non-Rust interfaces that expect an immutable `UNICODE_STRING*`).
pub fn as_ptr(&self) -> *const Self {
self as *const Self
}
/// Returns a slice to the raw [`u16`] codepoints of the string.
pub fn as_slice(&self) -> &'a [u16] {
unsafe { slice::from_raw_parts(self.raw.buffer, self.len_in_elements()) }
}
/// Returns a [`U16Str`] reference for this string.
///
/// The [`U16Str`] will only contain the characters and not the NUL terminator.
pub fn as_u16str(&self) -> &'a U16Str {
U16Str::from_slice(self.as_slice())
}
/// Returns the capacity (also known as "maximum length") of this string, in bytes.
pub fn capacity(&self) -> u16 {
self.raw.maximum_length
}
/// Returns the capacity (also known as "maximum length") of this string, in elements.
#[allow(unused)]
pub(crate) fn capacity_in_elements(&self) -> usize {
usize::from(self.raw.maximum_length) / mem::size_of::<u16>()
}
/// Returns an iterator over the [`char`]s of this string.
///
/// As the string may contain invalid UTF-16 characters (unpaired surrogates), the returned iterator is an
/// iterator over `Result<char>`.
/// Unpaired surrogates are returned as an [`NtStringError::UnpairedUtf16Surrogate`] error.
/// If you would like a lossy iterator over [`char`]s directly, use [`chars_lossy`] instead.
///
/// [`chars_lossy`]: Self::chars_lossy
pub fn chars(&self) -> Chars {
Chars::new(self)
}
/// Returns an iterator over the [`char`]s of this string.
///
/// Any invalid UTF-16 characters (unpaired surrogates) are automatically replaced by
/// [`U+FFFD REPLACEMENT CHARACTER`] (�).
/// If you would like to treat them differently, use [`chars`] instead.
///
/// [`chars`]: Self::chars
/// [`U+FFFD REPLACEMENT CHARACTER`]: std::char::REPLACEMENT_CHARACTER
pub fn chars_lossy(&self) -> CharsLossy {
CharsLossy::new(self)
}
/// Creates an [`NtUnicodeStr`] from a [`u16`] string buffer, a byte length of the string,
/// and a buffer capacity in bytes (also known as "maximum length").
///
/// The string is expected to consist of valid UTF-16 characters.
/// The buffer may or may not be NUL-terminated.
/// In any case, `length` does NOT include the terminating NUL character.
///
/// This function is `unsafe` and you are advised to use any of the safe `try_from_*`
/// functions over this one if possible.
///
/// # Safety
///
/// Behavior is undefined if any of the following conditions are violated:
///
/// * `length` must be less than or equal to `maximum_length`.
/// * `buffer` must be valid for at least `maximum_length` bytes.
/// * `buffer` must point to `length` consecutive properly initialized bytes.
/// * `buffer` must be valid for the duration of lifetime `'a`.
///
/// [`try_from_u16`]: Self::try_from_u16
/// [`try_from_u16_until_nul`]: Self::try_from_u16_until_nul
pub const unsafe fn from_raw_parts(
buffer: *const u16,
length: u16,
maximum_length: u16,
) -> Self {
debug_assert!(length <= maximum_length);
Self {
raw: RawNtString {
length,
maximum_length,
buffer,
},
_lifetime: PhantomData,
}
}
/// Returns `true` if this string has a length of zero, and `false` otherwise.
pub fn is_empty(&self) -> bool {
self.raw.length == 0
}
/// Returns the length of this string, in bytes.
///
/// Note that a single character may occupy more than one byte.
/// In other words, the returned value might not be what a human considers the length of the string.
pub fn len(&self) -> u16 {
self.raw.length
}
/// Returns the length of this string, in elements.
///
/// Note that a single character may occupy more than one element.
/// In other words, the returned value might not be what a human considers the length of the string.
pub(crate) fn len_in_elements(&self) -> usize {
usize::from(self.raw.length) / mem::size_of::<u16>()
}
/// Returns the remaining capacity of this string, in bytes.
#[allow(unused)]
pub(crate) fn remaining_capacity(&self) -> u16 {
debug_assert!(self.raw.maximum_length >= self.raw.length);
self.raw.maximum_length - self.raw.length
}
/// Creates an [`NtUnicodeStr`] from an existing [`u16`] string buffer without a terminating NUL character.
///
/// The string is expected to consist of valid UTF-16 characters.
///
/// The given buffer becomes the internal buffer of the [`NtUnicodeStr`] and therefore won't be NUL-terminated.
/// See the [module-level documentation](super) for the implications of that.
///
/// This function has *O*(1) complexity.
///
/// If you have a NUL-terminated buffer, either use [`try_from_u16_until_nul`] or convert from a [`U16CStr`]
/// using the corresponding [`TryFrom`] implementation.
///
/// [`try_from_u16_until_nul`]: Self::try_from_u16_until_nul
pub fn try_from_u16(buffer: &'a [u16]) -> Result<Self> {
let elements = buffer.len();
let length_usize = elements
.checked_mul(mem::size_of::<u16>())
.ok_or(NtStringError::BufferSizeExceedsU16)?;
let length =
u16::try_from(length_usize).map_err(|_| NtStringError::BufferSizeExceedsU16)?;
Ok(Self {
raw: RawNtString {
length,
maximum_length: length,
buffer: buffer.as_ptr(),
},
_lifetime: PhantomData,
})
}
/// Creates an [`NtUnicodeStr`] from an existing [`u16`] string buffer that contains at least one NUL character.
///
/// The string is expected to consist of valid UTF-16 characters.
///
/// The string will be terminated at the NUL character.
/// An [`NtStringError::NulNotFound`] error is returned if no NUL character could be found.
/// As a consequence, this function has *O*(*n*) complexity.
///
/// The resulting internal `buffer` of [`NtUnicodeStr`] will be NUL-terminated.
/// See the [module-level documentation](super) for the implications of that.
///
/// Use [`try_from_u16`] if you have a buffer that is not NUL-terminated.
/// You can also convert from a NUL-terminated [`U16CStr`] in *O*(1) via the corresponding [`TryFrom`] implementation.
///
/// [`try_from_u16`]: Self::try_from_u16
pub fn try_from_u16_until_nul(buffer: &'a [u16]) -> Result<Self> {
let length;
let maximum_length;
match buffer.iter().position(|x| *x == 0) {
Some(nul_pos) => {
// Include the terminating NUL character in `maximum_length` ...
let maximum_elements = nul_pos
.checked_add(1)
.ok_or(NtStringError::BufferSizeExceedsU16)?;
let maximum_length_usize = maximum_elements
.checked_mul(mem::size_of::<u16>())
.ok_or(NtStringError::BufferSizeExceedsU16)?;
maximum_length = u16::try_from(maximum_length_usize)
.map_err(|_| NtStringError::BufferSizeExceedsU16)?;
// ... but not in `length`
length = maximum_length - mem::size_of::<u16>() as u16;
}
None => return Err(NtStringError::NulNotFound),
};
Ok(Self {
raw: RawNtString {
length,
maximum_length,
buffer: buffer.as_ptr(),
},
_lifetime: PhantomData,
})
}
pub(crate) fn u16_iter(&'a self) -> Copied<Iter<'a, u16>> {
self.as_slice().iter().copied()
}
}
impl<'a> fmt::Display for NtUnicodeStr<'a> {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
for single_char in self.chars_lossy() {
single_char.fmt(f)?;
}
Ok(())
}
}
impl<'a> Eq for NtUnicodeStr<'a> {}
impl<'a> Ord for NtUnicodeStr<'a> {
fn cmp(&self, other: &Self) -> Ordering {
cmp_iter(self.u16_iter(), other.u16_iter())
}
}
impl<'a, 'b> PartialEq<NtUnicodeStr<'a>> for NtUnicodeStr<'b> {
/// Checks that two strings are a (case-sensitive!) match.
fn eq(&self, other: &NtUnicodeStr<'a>) -> bool {
self.as_slice() == other.as_slice()
}
}
impl<'a> PartialEq<str> for NtUnicodeStr<'a> {
fn eq(&self, other: &str) -> bool {
cmp_iter(self.u16_iter(), other.encode_utf16()) == Ordering::Equal
}
}
impl<'a> PartialEq<NtUnicodeStr<'a>> for str {
fn eq(&self, other: &NtUnicodeStr<'a>) -> bool {
cmp_iter(self.encode_utf16(), other.u16_iter()) == Ordering::Equal
}
}
impl<'a> PartialEq<&str> for NtUnicodeStr<'a> {
fn eq(&self, other: &&str) -> bool {
cmp_iter(self.u16_iter(), other.encode_utf16()) == Ordering::Equal
}
}
impl<'a> PartialEq<NtUnicodeStr<'a>> for &str {
fn eq(&self, other: &NtUnicodeStr<'a>) -> bool {
cmp_iter(self.encode_utf16(), other.u16_iter()) == Ordering::Equal
}
}
impl<'a> PartialOrd for NtUnicodeStr<'a> {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
impl<'a> PartialOrd<str> for NtUnicodeStr<'a> {
fn partial_cmp(&self, other: &str) -> Option<Ordering> {
Some(cmp_iter(self.u16_iter(), other.encode_utf16()))
}
}
impl<'a> PartialOrd<NtUnicodeStr<'a>> for str {
fn partial_cmp(&self, other: &NtUnicodeStr<'a>) -> Option<Ordering> {
Some(cmp_iter(self.encode_utf16(), other.u16_iter()))
}
}
impl<'a> PartialOrd<&str> for NtUnicodeStr<'a> {
fn partial_cmp(&self, other: &&str) -> Option<Ordering> {
Some(cmp_iter(self.u16_iter(), other.encode_utf16()))
}
}
impl<'a> PartialOrd<NtUnicodeStr<'a>> for &str {
fn partial_cmp(&self, other: &NtUnicodeStr<'a>) -> Option<Ordering> {
Some(cmp_iter(self.encode_utf16(), other.u16_iter()))
}
}
impl<'a> TryFrom<&'a U16CStr> for NtUnicodeStr<'a> {
type Error = NtStringError;
/// Converts a [`U16CStr`] reference into an [`NtUnicodeStr`].
///
/// The internal buffer will be NUL-terminated.
/// See the [module-level documentation](super) for the implications of that.
fn try_from(value: &'a U16CStr) -> Result<Self> {
let buffer = value.as_slice_with_nul();
// Include the terminating NUL character in `maximum_length` ...
let maximum_length_in_elements = buffer.len();
let maximum_length_in_bytes = maximum_length_in_elements
.checked_mul(mem::size_of::<u16>())
.ok_or(NtStringError::BufferSizeExceedsU16)?;
let maximum_length = u16::try_from(maximum_length_in_bytes)
.map_err(|_| NtStringError::BufferSizeExceedsU16)?;
// ... but not in `length`
debug_assert!(maximum_length >= mem::size_of::<u16>() as u16);
let length = maximum_length - mem::size_of::<u16>() as u16;
Ok(Self {
raw: RawNtString {
length,
maximum_length,
buffer: buffer.as_ptr(),
},
_lifetime: PhantomData,
})
}
}
impl<'a> TryFrom<&'a U16Str> for NtUnicodeStr<'a> {
type Error = NtStringError;
/// Converts a [`U16Str`] reference into an [`NtUnicodeStr`].
///
/// The internal buffer will NOT be NUL-terminated.
/// See the [module-level documentation](super) for the implications of that.
fn try_from(value: &'a U16Str) -> Result<Self> {
Self::try_from_u16(value.as_slice())
}
}