forked from denoland/rusty_v8
-
Notifications
You must be signed in to change notification settings - Fork 0
/
string.rs
687 lines (619 loc) · 20.6 KB
/
string.rs
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
use std::borrow::Cow;
use std::convert::TryInto;
use std::default::Default;
use std::mem::MaybeUninit;
use std::slice;
use crate::support::char;
use crate::support::int;
use crate::support::size_t;
use crate::HandleScope;
use crate::Isolate;
use crate::Local;
use crate::String;
extern "C" {
fn v8__String__kMaxLength() -> size_t;
fn v8__String__Empty(isolate: *mut Isolate) -> *const String;
fn v8__String__NewFromUtf8(
isolate: *mut Isolate,
data: *const char,
new_type: NewStringType,
length: int,
) -> *const String;
fn v8__String__NewFromOneByte(
isolate: *mut Isolate,
data: *const u8,
new_type: NewStringType,
length: int,
) -> *const String;
fn v8__String__NewFromTwoByte(
isolate: *mut Isolate,
data: *const u16,
new_type: NewStringType,
length: int,
) -> *const String;
fn v8__String__Length(this: *const String) -> int;
fn v8__String__Utf8Length(this: *const String, isolate: *mut Isolate) -> int;
fn v8__String__Write(
this: *const String,
isolate: *mut Isolate,
buffer: *mut u16,
start: int,
length: int,
options: WriteOptions,
) -> int;
fn v8__String__WriteOneByte(
this: *const String,
isolate: *mut Isolate,
buffer: *mut u8,
start: int,
length: int,
options: WriteOptions,
) -> int;
fn v8__String__WriteUtf8(
this: *const String,
isolate: *mut Isolate,
buffer: *mut char,
length: int,
nchars_ref: *mut int,
options: WriteOptions,
) -> int;
fn v8__String__NewExternalOneByte(
isolate: *mut Isolate,
onebyte_const: *const OneByteConst,
) -> *const String;
fn v8__String__NewExternalOneByteStatic(
isolate: *mut Isolate,
buffer: *const char,
length: int,
) -> *const String;
fn v8__String__NewExternalTwoByteStatic(
isolate: *mut Isolate,
buffer: *const u16,
length: int,
) -> *const String;
#[allow(dead_code)]
fn v8__String__IsExternal(this: *const String) -> bool;
fn v8__String__IsExternalOneByte(this: *const String) -> bool;
fn v8__String__IsExternalTwoByte(this: *const String) -> bool;
#[allow(dead_code)]
fn v8__String__IsOneByte(this: *const String) -> bool;
fn v8__String__ContainsOnlyOneByte(this: *const String) -> bool;
}
#[repr(C)]
pub struct OneByteConst {
vtable: *const OneByteConstNoOp,
cached_data: *const char,
length: int,
}
unsafe impl Sync for OneByteConst {}
extern "C" fn one_byte_const_no_op(_this: &'static OneByteConst) {}
extern "C" fn one_byte_const_is_cacheable(
_this: &'static OneByteConst,
) -> bool {
true
}
extern "C" fn one_byte_const_data(this: &'static OneByteConst) -> *const char {
this.cached_data
}
extern "C" fn one_byte_const_length(this: &'static OneByteConst) -> usize {
this.length as usize
}
type OneByteConstNoOp = extern "C" fn(&'static OneByteConst);
type OneByteConstIsCacheable = extern "C" fn(&'static OneByteConst) -> bool;
type OneByteConstData = extern "C" fn(&'static OneByteConst) -> *const char;
type OneByteConstLength = extern "C" fn(&'static OneByteConst) -> usize;
#[repr(C)]
struct OneByteConstVtable {
_typeinfo: *const (),
_base_offset: usize,
delete1: OneByteConstNoOp,
#[cfg(not(target_family = "windows"))]
delete2: OneByteConstNoOp,
is_cacheable: OneByteConstIsCacheable,
dispose: OneByteConstNoOp,
lock: OneByteConstNoOp,
unlock: OneByteConstNoOp,
data: OneByteConstData,
length: OneByteConstLength,
}
const ONE_BYTE_CONST_VTABLE: OneByteConstVtable = OneByteConstVtable {
_typeinfo: std::ptr::null(),
_base_offset: 0,
delete1: one_byte_const_no_op,
#[cfg(not(target_family = "windows"))]
delete2: one_byte_const_no_op,
is_cacheable: one_byte_const_is_cacheable,
dispose: one_byte_const_no_op,
lock: one_byte_const_no_op,
unlock: one_byte_const_no_op,
data: one_byte_const_data,
length: one_byte_const_length,
};
/// Compile-time function to determine if a string is ASCII. Note that UTF-8 chars
/// longer than one byte have the high-bit set and thus, are not ASCII.
const fn is_ascii(s: &'static [u8]) -> bool {
let mut i = 0;
while i < s.len() {
if !s[i].is_ascii() {
return false;
}
i += 1;
}
true
}
#[repr(C)]
#[derive(Debug)]
pub enum NewStringType {
Normal,
Internalized,
}
impl Default for NewStringType {
fn default() -> Self {
NewStringType::Normal
}
}
bitflags! {
#[derive(Default)]
#[repr(transparent)]
pub struct WriteOptions: int {
const NO_OPTIONS = 0;
const HINT_MANY_WRITES_EXPECTED = 1;
const NO_NULL_TERMINATION = 2;
const PRESERVE_ONE_BYTE_NULL = 4;
// Used by WriteUtf8 to replace orphan surrogate code units with the
// unicode replacement character. Needs to be set to guarantee valid UTF-8
// output.
const REPLACE_INVALID_UTF8 = 8;
}
}
impl String {
/// The maximum length (in bytes) of a buffer that a v8::String can be built
/// from. Attempting to create a v8::String from a larger buffer will result
/// in None being returned.
#[inline(always)]
pub fn max_length() -> usize {
unsafe { v8__String__kMaxLength() }
}
#[inline(always)]
pub fn empty<'s>(scope: &mut HandleScope<'s, ()>) -> Local<'s, String> {
// FIXME(bnoordhuis) v8__String__Empty() is infallible so there
// is no need to box up the result, only to unwrap it again.
unsafe { scope.cast_local(|sd| v8__String__Empty(sd.get_isolate_ptr())) }
.unwrap()
}
/// Allocates a new string from UTF-8 data. Only returns an empty value when
/// length > kMaxLength
#[inline(always)]
pub fn new_from_utf8<'s>(
scope: &mut HandleScope<'s, ()>,
buffer: &[u8],
new_type: NewStringType,
) -> Option<Local<'s, String>> {
if buffer.is_empty() {
return Some(Self::empty(scope));
}
let buffer_len = buffer.len().try_into().ok()?;
unsafe {
scope.cast_local(|sd| {
v8__String__NewFromUtf8(
sd.get_isolate_ptr(),
buffer.as_ptr() as *const char,
new_type,
buffer_len,
)
})
}
}
/// Allocates a new string from Latin-1 data. Only returns an empty value when
/// length > kMaxLength.
#[inline(always)]
pub fn new_from_one_byte<'s>(
scope: &mut HandleScope<'s, ()>,
buffer: &[u8],
new_type: NewStringType,
) -> Option<Local<'s, String>> {
let buffer_len = buffer.len().try_into().ok()?;
unsafe {
scope.cast_local(|sd| {
v8__String__NewFromOneByte(
sd.get_isolate_ptr(),
buffer.as_ptr(),
new_type,
buffer_len,
)
})
}
}
/// Allocates a new string from UTF-16 data. Only returns an empty value when
/// length > kMaxLength.
#[inline(always)]
pub fn new_from_two_byte<'s>(
scope: &mut HandleScope<'s, ()>,
buffer: &[u16],
new_type: NewStringType,
) -> Option<Local<'s, String>> {
let buffer_len = buffer.len().try_into().ok()?;
unsafe {
scope.cast_local(|sd| {
v8__String__NewFromTwoByte(
sd.get_isolate_ptr(),
buffer.as_ptr(),
new_type,
buffer_len,
)
})
}
}
/// Returns the number of characters (UTF-16 code units) in this string.
#[inline(always)]
pub fn length(&self) -> usize {
unsafe { v8__String__Length(self) as usize }
}
/// Returns the number of bytes in the UTF-8 encoded representation of this
/// string.
#[inline(always)]
pub fn utf8_length(&self, scope: &mut Isolate) -> usize {
unsafe { v8__String__Utf8Length(self, scope) as usize }
}
/// Writes the contents of the string to an external buffer, as 16-bit
/// (UTF-16) character codes.
#[inline(always)]
pub fn write(
&self,
scope: &mut Isolate,
buffer: &mut [u16],
start: usize,
options: WriteOptions,
) -> usize {
unsafe {
v8__String__Write(
self,
scope,
buffer.as_mut_ptr(),
start.try_into().unwrap_or(int::max_value()),
buffer.len().try_into().unwrap_or(int::max_value()),
options,
) as usize
}
}
/// Writes the contents of the string to an external buffer, as one-byte
/// (Latin-1) characters.
#[inline(always)]
pub fn write_one_byte(
&self,
scope: &mut Isolate,
buffer: &mut [u8],
start: usize,
options: WriteOptions,
) -> usize {
unsafe {
v8__String__WriteOneByte(
self,
scope,
buffer.as_mut_ptr(),
start.try_into().unwrap_or(int::max_value()),
buffer.len().try_into().unwrap_or(int::max_value()),
options,
) as usize
}
}
/// Writes the contents of the string to an external [`MaybeUninit`] buffer, as one-byte
/// (Latin-1) characters.
#[inline(always)]
pub fn write_one_byte_uninit(
&self,
scope: &mut Isolate,
buffer: &mut [MaybeUninit<u8>],
start: usize,
options: WriteOptions,
) -> usize {
unsafe {
v8__String__WriteOneByte(
self,
scope,
buffer.as_mut_ptr() as *mut u8,
start.try_into().unwrap_or(int::max_value()),
buffer.len().try_into().unwrap_or(int::max_value()),
options,
) as usize
}
}
/// Writes the contents of the string to an external buffer, as UTF-8.
#[inline(always)]
pub fn write_utf8(
&self,
scope: &mut Isolate,
buffer: &mut [u8],
nchars_ref: Option<&mut usize>,
options: WriteOptions,
) -> usize {
unsafe {
// SAFETY:
// We assume that v8 will overwrite the buffer without de-initializing any byte in it.
// So the type casting of the buffer is safe.
let buffer = {
let len = buffer.len();
let data = buffer.as_mut_ptr().cast();
slice::from_raw_parts_mut(data, len)
};
self.write_utf8_uninit(scope, buffer, nchars_ref, options)
}
}
/// Writes the contents of the string to an external [`MaybeUninit`] buffer, as UTF-8.
pub fn write_utf8_uninit(
&self,
scope: &mut Isolate,
buffer: &mut [MaybeUninit<u8>],
nchars_ref: Option<&mut usize>,
options: WriteOptions,
) -> usize {
let mut nchars_ref_int: int = 0;
let bytes = unsafe {
v8__String__WriteUtf8(
self,
scope,
buffer.as_mut_ptr() as *mut char,
buffer.len().try_into().unwrap_or(int::max_value()),
&mut nchars_ref_int,
options,
)
};
if let Some(r) = nchars_ref {
*r = nchars_ref_int as usize;
}
bytes as usize
}
// Convenience function not present in the original V8 API.
#[inline(always)]
pub fn new<'s>(
scope: &mut HandleScope<'s, ()>,
value: &str,
) -> Option<Local<'s, String>> {
Self::new_from_utf8(scope, value.as_ref(), NewStringType::Normal)
}
// Compile-time function to create an external string resource.
// This resource must be assigned into a 'static mut'.
#[inline(always)]
pub const fn create_external_onebyte_const(
buffer: &'static [u8],
) -> OneByteConst {
is_ascii(buffer);
OneByteConst {
vtable: &ONE_BYTE_CONST_VTABLE.delete1,
cached_data: buffer.as_ptr() as *const char,
length: buffer.len() as i32,
}
}
// Creates a v8::String from a `&'static [u8]`,
// must be Latin-1 or ASCII, not UTF-8 !
#[inline(always)]
pub fn new_from_onebyte_const<'s>(
scope: &mut HandleScope<'s, ()>,
onebyte_const: &'static OneByteConst,
) -> Option<Local<'s, String>> {
unsafe {
scope.cast_local(|sd| {
v8__String__NewExternalOneByte(sd.get_isolate_ptr(), onebyte_const)
})
}
}
// Creates a v8::String from a `&'static [u8]`,
// must be Latin-1 or ASCII, not UTF-8 !
#[inline(always)]
pub fn new_external_onebyte_static<'s>(
scope: &mut HandleScope<'s, ()>,
buffer: &'static [u8],
) -> Option<Local<'s, String>> {
let buffer_len = buffer.len().try_into().ok()?;
unsafe {
scope.cast_local(|sd| {
v8__String__NewExternalOneByteStatic(
sd.get_isolate_ptr(),
buffer.as_ptr() as *const char,
buffer_len,
)
})
}
}
// Creates a v8::String from a `&'static [u16]`.
#[inline(always)]
pub fn new_external_twobyte_static<'s>(
scope: &mut HandleScope<'s, ()>,
buffer: &'static [u16],
) -> Option<Local<'s, String>> {
let buffer_len = buffer.len().try_into().ok()?;
unsafe {
scope.cast_local(|sd| {
v8__String__NewExternalTwoByteStatic(
sd.get_isolate_ptr(),
buffer.as_ptr(),
buffer_len,
)
})
}
}
/// True if string is external
#[inline(always)]
pub fn is_external(&self) -> bool {
// TODO: re-enable on next v8-release
// Right now it fallbacks to Value::IsExternal, which is incorrect
// See: https://source.chromium.org/chromium/_/chromium/v8/v8.git/+/1dd8624b524d14076160c1743f7da0b20fbe68e0
// unsafe { v8__String__IsExternal(self) }
// Fallback for now (though functionally identical)
self.is_external_onebyte() || self.is_external_twobyte()
}
/// True if string is external & one-byte
/// (e.g: created with new_external_onebyte_static)
#[inline(always)]
pub fn is_external_onebyte(&self) -> bool {
unsafe { v8__String__IsExternalOneByte(self) }
}
/// True if string is external & two-byte
/// (e.g: created with new_external_twobyte_static)
#[inline(always)]
pub fn is_external_twobyte(&self) -> bool {
unsafe { v8__String__IsExternalTwoByte(self) }
}
/// Will return true if and only if string is known for certain to contain only one-byte data,
/// ie: Latin-1, a.k.a. ISO-8859-1 code points. Doesn't read the string so can return false
/// negatives, and a return value of false does not mean this string is not one-byte data.
///
/// For a method that will not return false negatives at the cost of
/// potentially reading the entire string, use [`contains_only_onebyte()`].
///
/// [`contains_only_onebyte()`]: String::contains_only_onebyte
#[inline(always)]
pub fn is_onebyte(&self) -> bool {
unsafe { v8__String__IsOneByte(self) }
}
/// True if the string contains only one-byte data.
/// Will read the entire string in some cases.
#[inline(always)]
pub fn contains_only_onebyte(&self) -> bool {
unsafe { v8__String__ContainsOnlyOneByte(self) }
}
/// Creates a copy of a [`crate::String`] in a [`std::string::String`].
/// Convenience function not present in the original V8 API.
pub fn to_rust_string_lossy(
&self,
scope: &mut Isolate,
) -> std::string::String {
let len_utf8 = self.utf8_length(scope);
let len_utf16 = self.length();
// If len_utf8 == len_utf16 and the string is one-byte, we can take the fast memcpy path. This is true iff the
// string is 100% 7-bit ASCII.
if self.is_onebyte() && len_utf8 == len_utf16 {
unsafe {
// Create an uninitialized buffer of `capacity` bytes. We need to be careful here to avoid
// accidentally creating a slice of u8 which would be invalid.
let layout = std::alloc::Layout::from_size_align(len_utf16, 1).unwrap();
let data = std::alloc::alloc(layout) as *mut MaybeUninit<u8>;
let buffer = std::ptr::slice_from_raw_parts_mut(data, len_utf16);
// Write to this MaybeUninit buffer, assuming we're going to fill this entire buffer
let length = self.write_one_byte_uninit(
scope,
&mut *buffer,
0,
WriteOptions::NO_NULL_TERMINATION
| WriteOptions::REPLACE_INVALID_UTF8,
);
debug_assert!(length == len_utf16);
// Return an owned string from this guaranteed now-initialized data
let buffer = data as *mut u8;
return std::string::String::from_raw_parts(buffer, length, len_utf16);
}
}
// SAFETY: This allocates a buffer manually using the default allocator using the string's capacity.
// We have a large number of invariants to uphold, so please check changes to this code carefully
unsafe {
// Create an uninitialized buffer of `capacity` bytes. We need to be careful here to avoid
// accidentally creating a slice of u8 which would be invalid.
let layout = std::alloc::Layout::from_size_align(len_utf8, 1).unwrap();
let data = std::alloc::alloc(layout) as *mut MaybeUninit<u8>;
let buffer = std::ptr::slice_from_raw_parts_mut(data, len_utf8);
// Write to this MaybeUninit buffer, assuming we're going to fill this entire buffer
let length = self.write_utf8_uninit(
scope,
&mut *buffer,
None,
WriteOptions::NO_NULL_TERMINATION | WriteOptions::REPLACE_INVALID_UTF8,
);
debug_assert!(length == len_utf8);
// Return an owned string from this guaranteed now-initialized data
let buffer = data as *mut u8;
std::string::String::from_raw_parts(buffer, length, len_utf8)
}
}
/// Converts a [`crate::String`] to either an owned [`std::string::String`], or a borrowed [`str`], depending on whether it fits into the
/// provided buffer.
pub fn to_rust_cow_lossy<'a, const N: usize>(
&self,
scope: &mut Isolate,
buffer: &'a mut [MaybeUninit<u8>; N],
) -> Cow<'a, str> {
// TODO(mmastrac): Ideally we should be able to access the string's internal representation
let len_utf8 = self.utf8_length(scope);
let len_utf16 = self.length();
// If len_utf8 == len_utf16 and the string is one-byte, we can take the fast memcpy path. This is true iff the
// string is 100% 7-bit ASCII.
if self.is_onebyte() && len_utf8 == len_utf16 {
if len_utf16 <= N {
let length = self.write_one_byte_uninit(
scope,
buffer,
0,
WriteOptions::NO_NULL_TERMINATION,
);
debug_assert!(length == len_utf16);
unsafe {
// Get a slice of &[u8] of what we know is initialized now
let buffer = &mut buffer[..length];
let buffer = &mut *(buffer as *mut [_] as *mut [u8]);
// We know it's valid UTF-8, so make a string
return Cow::Borrowed(std::str::from_utf8_unchecked(buffer));
}
}
unsafe {
// Create an uninitialized buffer of `capacity` bytes. We need to be careful here to avoid
// accidentally creating a slice of u8 which would be invalid.
let layout = std::alloc::Layout::from_size_align(len_utf16, 1).unwrap();
let data = std::alloc::alloc(layout) as *mut MaybeUninit<u8>;
let buffer = std::ptr::slice_from_raw_parts_mut(data, len_utf16);
// Write to this MaybeUninit buffer, assuming we're going to fill this entire buffer
let length = self.write_one_byte_uninit(
scope,
&mut *buffer,
0,
WriteOptions::NO_NULL_TERMINATION
| WriteOptions::REPLACE_INVALID_UTF8,
);
debug_assert!(length == len_utf16);
// Return an owned string from this guaranteed now-initialized data
let buffer = data as *mut u8;
return Cow::Owned(std::string::String::from_raw_parts(
buffer, length, len_utf16,
));
}
}
if len_utf8 <= N {
// No malloc path
let length = self.write_utf8_uninit(
scope,
buffer,
None,
WriteOptions::NO_NULL_TERMINATION | WriteOptions::REPLACE_INVALID_UTF8,
);
debug_assert!(length == len_utf8);
// SAFETY: We know that we wrote `length` UTF-8 bytes. See `slice_assume_init_mut` for additional guarantee information.
unsafe {
// Get a slice of &[u8] of what we know is initialized now
let buffer = &mut buffer[..length];
let buffer = &mut *(buffer as *mut [_] as *mut [u8]);
// We know it's valid UTF-8, so make a string
return Cow::Borrowed(std::str::from_utf8_unchecked(buffer));
}
}
// SAFETY: This allocates a buffer manually using the default allocator using the string's capacity.
// We have a large number of invariants to uphold, so please check changes to this code carefully
unsafe {
// Create an uninitialized buffer of `capacity` bytes. We need to be careful here to avoid
// accidentally creating a slice of u8 which would be invalid.
let layout = std::alloc::Layout::from_size_align(len_utf8, 1).unwrap();
let data = std::alloc::alloc(layout) as *mut MaybeUninit<u8>;
let buffer = std::ptr::slice_from_raw_parts_mut(data, len_utf8);
// Write to this MaybeUninit buffer, assuming we're going to fill this entire buffer
let length = self.write_utf8_uninit(
scope,
&mut *buffer,
None,
WriteOptions::NO_NULL_TERMINATION | WriteOptions::REPLACE_INVALID_UTF8,
);
debug_assert!(length == len_utf8);
// Return an owned string from this guaranteed now-initialized data
let buffer = data as *mut u8;
Cow::Owned(std::string::String::from_raw_parts(
buffer, length, len_utf8,
))
}
}
}