1 /**
2 * Copyright (c) 2021-2024 Huawei Device Co., Ltd.
3 * Licensed under the Apache License, Version 2.0 (the "License");
4 * you may not use this file except in compliance with the License.
5 * You may obtain a copy of the License at
6 *
7 * http://www.apache.org/licenses/LICENSE-2.0
8 *
9 * Unless required by applicable law or agreed to in writing, software
10 * distributed under the License is distributed on an "AS IS" BASIS,
11 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 * See the License for the specific language governing permissions and
13 * limitations under the License.
14 */
15
16 #include <algorithm>
17 #include <cstdint>
18 #include <regex>
19 #include "include/mem/panda_string.h"
20 #include "intrinsics.h"
21 #include "libpandabase/utils/logger.h"
22 #include "macros.h"
23 #include "napi/ets_napi.h"
24 #include "runtime/handle_scope-inl.h"
25 #include "runtime/entrypoints/string_index_of.h"
26 #include "runtime/arch/memory_helpers.h"
27 #include "plugins/ets/runtime/types/ets_string.h"
28 #include "plugins/ets/runtime/ets_exceptions.h"
29 #include "plugins/ets/runtime/ets_language_context.h"
30 #include "plugins/ets/runtime/ets_panda_file_items.h"
31
32 #include "unicode/locid.h"
33 #include "unicode/coll.h"
34 #include "unicode/unistr.h"
35 #include "unicode/normalizer2.h"
36 #include "utils/span.h"
37
38 using icu::Normalizer2;
39
40 namespace ark::ets::intrinsics {
41
42 constexpr const uint32_t CHAR0X1FFC00 = 0x1ffc00;
43 constexpr const uint16_t CHAR0XD800 = 0xd800;
44 constexpr const uint16_t CHAR0XDC00 = 0xdc00;
45
StdCoreStringGetDataAsArray(EtsString * s,ets_int begin,ets_int end,bool isUtf16)46 static ObjectHeader *StdCoreStringGetDataAsArray(EtsString *s, ets_int begin, ets_int end, bool isUtf16)
47 {
48 ASSERT(s != nullptr);
49 ets_int length = s->GetLength();
50 if (UNLIKELY(begin > end)) {
51 ark::ThrowStringIndexOutOfBoundsException(begin, length);
52 return nullptr;
53 }
54 if (UNLIKELY(begin > length || begin < 0)) {
55 ark::ThrowStringIndexOutOfBoundsException(begin, length);
56 return nullptr;
57 }
58 if (UNLIKELY(end > length)) {
59 ark::ThrowStringIndexOutOfBoundsException(end, length);
60 return nullptr;
61 }
62
63 auto thread = ManagedThread::GetCurrent();
64 [[maybe_unused]] HandleScope<ObjectHeader *> scope(thread);
65 VMHandle<coretypes::String> sHandle(thread, s->GetCoreType());
66 ets_int n = end - begin;
67 void *array = nullptr;
68 if (isUtf16) {
69 array = EtsCharArray::Create(n);
70 } else {
71 array = EtsByteArray::Create(n);
72 }
73 if (array == nullptr || n == 0) {
74 return reinterpret_cast<ObjectHeader *>(array);
75 }
76 if (isUtf16) {
77 auto charArray = reinterpret_cast<EtsCharArray *>(array);
78 Span<ets_char> out(charArray->GetData<ets_char>(), charArray->GetLength());
79 sHandle.GetPtr()->CopyDataRegionUtf16(&out[0], begin, charArray->GetLength(), sHandle.GetPtr()->GetLength());
80 } else {
81 auto byteArray = reinterpret_cast<EtsByteArray *>(array);
82 Span<uint8_t> out(byteArray->GetData<uint8_t>(), byteArray->GetLength());
83
84 /* as we need only one LSB no sophisticated conversion is needed */
85 if (sHandle.GetPtr()->IsUtf16()) {
86 auto in = sHandle.GetPtr()->GetDataUtf16();
87 for (int i = 0; i < n; ++i) {
88 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
89 out[i] = in[i + begin];
90 }
91 } else {
92 auto in = sHandle.GetPtr()->GetDataMUtf8();
93 for (int i = 0; i < n; ++i) {
94 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
95 out[i] = in[i + begin];
96 }
97 }
98 }
99 return reinterpret_cast<ObjectHeader *>(array);
100 }
101
StdCoreStringGetChars(EtsString * s,ets_int begin,ets_int end)102 ObjectHeader *StdCoreStringGetChars(EtsString *s, ets_int begin, ets_int end)
103 {
104 return StdCoreStringGetDataAsArray(s, begin, end, true);
105 }
106
StdCoreStringGetBytes(EtsString * s,ets_int begin,ets_int end)107 ObjectHeader *StdCoreStringGetBytes(EtsString *s, ets_int begin, ets_int end)
108 {
109 return StdCoreStringGetDataAsArray(s, begin, end, false);
110 }
111
StdCoreStringSubstring(EtsString * str,ets_int begin,ets_int end)112 EtsString *StdCoreStringSubstring(EtsString *str, ets_int begin, ets_int end)
113 {
114 ASSERT(str != nullptr);
115 auto indexes = coretypes::String::NormalizeSubStringIndexes(begin, end, str->GetCoreType());
116 if (UNLIKELY(indexes.first == 0 && indexes.second == str->GetLength())) {
117 return str;
118 }
119 ets_int substrLength = indexes.second - indexes.first;
120 return EtsString::FastSubString(str, static_cast<uint32_t>(indexes.first), static_cast<uint32_t>(substrLength));
121 }
122
StdCoreStringCharAt(EtsString * s,int32_t index)123 uint16_t StdCoreStringCharAt(EtsString *s, int32_t index)
124 {
125 ASSERT(s != nullptr);
126
127 int32_t length = s->GetLength();
128 if (UNLIKELY(index >= length || index < 0)) {
129 ark::ThrowStringIndexOutOfBoundsException(index, length);
130 return 0;
131 }
132
133 if (s->IsUtf16()) {
134 Span<uint16_t> sp(s->GetDataUtf16(), length);
135 return sp[index];
136 }
137
138 Span<uint8_t> sp(s->GetDataMUtf8(), length);
139 return sp[index];
140 }
141
StdCoreStringGetLength(EtsString * s)142 int32_t StdCoreStringGetLength(EtsString *s)
143 {
144 ASSERT(s != nullptr);
145 return s->GetLength();
146 }
147
StdCoreStringLength(EtsString * s)148 double StdCoreStringLength(EtsString *s)
149 {
150 ASSERT(s != nullptr);
151 return static_cast<double>(s->GetLength());
152 }
153
StdCoreStringIsEmpty(EtsString * s)154 EtsBoolean StdCoreStringIsEmpty(EtsString *s)
155 {
156 ASSERT(s != nullptr);
157 return ToEtsBoolean(s->IsEmpty());
158 }
159
StdCoreStringEquals(EtsString * owner,EtsObject * s)160 uint8_t StdCoreStringEquals(EtsString *owner, EtsObject *s)
161 {
162 if ((owner->AsObject()) == s) {
163 return UINT8_C(1);
164 }
165 if (s == nullptr || !(s->GetClass()->IsStringClass())) {
166 return UINT8_C(0);
167 }
168 return static_cast<uint8_t>(owner->StringsAreEqual(s));
169 }
170
StdCoreStringMatch(EtsString * thisStr,EtsString * reg)171 EtsString *StdCoreStringMatch(EtsString *thisStr, EtsString *reg)
172 {
173 PandaVector<uint8_t> buf;
174 auto thisS = std::string(thisStr->ConvertToStringView(&buf));
175 auto regex = std::string(reg->ConvertToStringView(&buf));
176
177 std::regex e(regex);
178 return EtsString::CreateFromMUtf8(std::sregex_iterator(thisS.begin(), thisS.end(), e)->str().c_str());
179 }
180
StringNormalize(EtsString * str,const Normalizer2 * normalizer)181 EtsString *StringNormalize(EtsString *str, const Normalizer2 *normalizer)
182 {
183 auto coroutine = EtsCoroutine::GetCurrent();
184 [[maybe_unused]] HandleScope<ObjectHeader *> scope(coroutine);
185
186 icu::UnicodeString utf16Str;
187 if (str->IsUtf16()) {
188 utf16Str = icu::UnicodeString {str->GetDataUtf16(), static_cast<int32_t>(str->GetUtf16Length())};
189 } else {
190 utf16Str =
191 icu::UnicodeString {utf::Mutf8AsCString(str->GetDataMUtf8()), static_cast<int32_t>(str->GetLength())};
192 }
193
194 UErrorCode errorCode = U_ZERO_ERROR;
195 utf16Str = normalizer->normalize(utf16Str, errorCode);
196
197 if (UNLIKELY(U_FAILURE(errorCode))) {
198 std::string message = "Got error in process of normalization: '" + std::string(u_errorName(errorCode)) + "'";
199 ThrowEtsException(coroutine, panda_file_items::class_descriptors::RANGE_ERROR, message);
200 return nullptr;
201 }
202
203 return EtsString::CreateFromUtf16(reinterpret_cast<const uint16_t *>(utf16Str.getTerminatedBuffer()),
204 utf16Str.length());
205 }
206
StdCoreStringNormalizeNFC(EtsString * thisStr)207 EtsString *StdCoreStringNormalizeNFC(EtsString *thisStr)
208 {
209 UErrorCode errorCode = U_ZERO_ERROR;
210 auto normalizer = Normalizer2::getNFCInstance(errorCode);
211 if (UNLIKELY(U_FAILURE(errorCode))) {
212 std::string message = "Cannot get NFC normalizer: '" + std::string(u_errorName(errorCode)) + "'";
213 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
214 return nullptr;
215 }
216 return StringNormalize(thisStr, normalizer);
217 }
218
StdCoreStringNormalizeNFD(EtsString * thisStr)219 EtsString *StdCoreStringNormalizeNFD(EtsString *thisStr)
220 {
221 UErrorCode errorCode = U_ZERO_ERROR;
222 auto normalizer = Normalizer2::getNFDInstance(errorCode);
223 if (UNLIKELY(U_FAILURE(errorCode))) {
224 std::string message = "Cannot get NFD normalizer: '" + std::string(u_errorName(errorCode)) + "'";
225 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
226 return nullptr;
227 }
228 return StringNormalize(thisStr, normalizer);
229 }
230
StdCoreStringNormalizeNFKC(EtsString * thisStr)231 EtsString *StdCoreStringNormalizeNFKC(EtsString *thisStr)
232 {
233 UErrorCode errorCode = U_ZERO_ERROR;
234 auto normalizer = Normalizer2::getNFKCInstance(errorCode);
235 if (UNLIKELY(U_FAILURE(errorCode))) {
236 std::string message = "Cannot get NFKC normalizer: '" + std::string(u_errorName(errorCode)) + "'";
237 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
238 return nullptr;
239 }
240 return StringNormalize(thisStr, normalizer);
241 }
242
StdCoreStringNormalizeNFKD(EtsString * thisStr)243 EtsString *StdCoreStringNormalizeNFKD(EtsString *thisStr)
244 {
245 UErrorCode errorCode = U_ZERO_ERROR;
246 auto normalizer = Normalizer2::getNFKDInstance(errorCode);
247 if (UNLIKELY(U_FAILURE(errorCode))) {
248 std::string message = "Cannot get NFKD normalizer: '" + std::string(u_errorName(errorCode)) + "'";
249 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
250 return nullptr;
251 }
252 return StringNormalize(thisStr, normalizer);
253 }
254
StdCoreStringIsWellFormed(EtsString * thisStr)255 uint8_t StdCoreStringIsWellFormed(EtsString *thisStr)
256 {
257 if (!thisStr->IsUtf16()) {
258 return UINT8_C(1);
259 }
260 auto length = thisStr->GetUtf16Length();
261 auto codeUnits = Span<uint16_t>(thisStr->GetDataUtf16(), length);
262 for (size_t i = 0; i < length; ++i) {
263 uint16_t codeUnit = codeUnits[i];
264 if ((codeUnit & CHAR0X1FFC00) == CHAR0XD800) {
265 // Code unit is a leading surrogate
266 if (i == length - 1) {
267 return UINT8_C(0);
268 }
269 // Is not trail surrogate
270 if ((codeUnits[i + 1] & CHAR0X1FFC00) != CHAR0XDC00) {
271 return UINT8_C(0);
272 }
273 // Skip the paired trailing surrogate
274 ++i;
275 // Is trail surrogate
276 } else if ((codeUnit & CHAR0X1FFC00) == CHAR0XDC00) {
277 return UINT8_C(0);
278 }
279 }
280 return UINT8_C(1);
281 }
282
ToLowerCase(EtsString * thisStr,const icu::Locale & locale)283 EtsString *ToLowerCase(EtsString *thisStr, const icu::Locale &locale)
284 {
285 auto coroutine = EtsCoroutine::GetCurrent();
286 [[maybe_unused]] HandleScope<ObjectHeader *> scope(coroutine);
287
288 icu::UnicodeString utf16Str;
289 if (thisStr->IsUtf16()) {
290 utf16Str = icu::UnicodeString {thisStr->GetDataUtf16(), static_cast<int32_t>(thisStr->GetUtf16Length())};
291 } else {
292 utf16Str = icu::UnicodeString {utf::Mutf8AsCString(thisStr->GetDataMUtf8()),
293 static_cast<int32_t>(thisStr->GetLength())};
294 }
295 auto res = utf16Str.toLower(locale);
296 return EtsString::CreateFromUtf16(reinterpret_cast<const uint16_t *>(res.getTerminatedBuffer()), res.length());
297 }
298
ToUpperCase(EtsString * thisStr,const icu::Locale & locale)299 EtsString *ToUpperCase(EtsString *thisStr, const icu::Locale &locale)
300 {
301 auto coroutine = EtsCoroutine::GetCurrent();
302 [[maybe_unused]] HandleScope<ObjectHeader *> scope(coroutine);
303
304 icu::UnicodeString utf16Str;
305 if (thisStr->IsUtf16()) {
306 utf16Str = icu::UnicodeString {thisStr->GetDataUtf16(), static_cast<int32_t>(thisStr->GetUtf16Length())};
307 } else {
308 utf16Str = icu::UnicodeString {utf::Mutf8AsCString(thisStr->GetDataMUtf8()),
309 static_cast<int32_t>(thisStr->GetLength())};
310 }
311 auto res = utf16Str.toUpper(locale);
312 return EtsString::CreateFromUtf16(reinterpret_cast<const uint16_t *>(res.getTerminatedBuffer()), res.length());
313 }
314
ParseSingleBCP47LanguageTag(EtsString * langTag,icu::Locale & locale)315 UErrorCode ParseSingleBCP47LanguageTag(EtsString *langTag, icu::Locale &locale)
316 {
317 if (langTag == nullptr) {
318 locale = icu::Locale::getDefault();
319 return U_ZERO_ERROR;
320 }
321
322 PandaVector<uint8_t> buf;
323 std::string_view locTag = langTag->ConvertToStringView(&buf);
324 icu::StringPiece sp {locTag.data(), static_cast<int32_t>(locTag.size())};
325 UErrorCode status = U_ZERO_ERROR;
326 locale = icu::Locale::forLanguageTag(sp, status);
327 return status;
328 }
329
StdCoreStringToUpperCase(EtsString * thisStr)330 EtsString *StdCoreStringToUpperCase(EtsString *thisStr)
331 {
332 return ToUpperCase(thisStr, icu::Locale::getDefault());
333 }
334
StdCoreStringToLowerCase(EtsString * thisStr)335 EtsString *StdCoreStringToLowerCase(EtsString *thisStr)
336 {
337 return ToLowerCase(thisStr, icu::Locale::getDefault());
338 }
339
StdCoreStringToLocaleUpperCase(EtsString * thisStr,EtsString * langTag)340 EtsString *StdCoreStringToLocaleUpperCase(EtsString *thisStr, EtsString *langTag)
341 {
342 ASSERT(langTag != nullptr);
343
344 icu::Locale locale;
345 auto localeParseStatus = ParseSingleBCP47LanguageTag(langTag, locale);
346 if (UNLIKELY(U_FAILURE(localeParseStatus))) {
347 auto message = "Language tag '" + ConvertToString(langTag->GetCoreType()) + "' is invalid or not supported";
348 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
349 return nullptr;
350 }
351 return ToUpperCase(thisStr, locale);
352 }
353
StdCoreStringToLocaleLowerCase(EtsString * thisStr,EtsString * langTag)354 EtsString *StdCoreStringToLocaleLowerCase(EtsString *thisStr, EtsString *langTag)
355 {
356 ASSERT(langTag != nullptr);
357
358 icu::Locale locale;
359 auto localeParseStatus = ParseSingleBCP47LanguageTag(langTag, locale);
360 if (UNLIKELY(U_FAILURE(localeParseStatus))) {
361 auto message = "Language tag '" + ConvertToString(langTag->GetCoreType()) + "' is invalid or not supported";
362 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
363 return nullptr;
364 }
365 return ToLowerCase(thisStr, locale);
366 }
367
StdCoreStringLocaleCmp(EtsString * thisStr,EtsString * cmpStr,EtsString * langTag)368 ets_double StdCoreStringLocaleCmp(EtsString *thisStr, EtsString *cmpStr, EtsString *langTag)
369 {
370 ASSERT(thisStr != nullptr && cmpStr != nullptr);
371
372 icu::Locale locale;
373 auto status = ParseSingleBCP47LanguageTag(langTag, locale);
374 if (UNLIKELY(U_FAILURE(status))) {
375 auto message = "Language tag '" + ConvertToString(langTag->GetCoreType()) + "' is invalid or not supported";
376 ThrowEtsException(EtsCoroutine::GetCurrent(), panda_file_items::class_descriptors::RANGE_ERROR, message);
377 return 0;
378 }
379
380 icu::UnicodeString source;
381 if (thisStr->IsUtf16()) {
382 source = icu::UnicodeString {thisStr->GetDataUtf16(), static_cast<int32_t>(thisStr->GetUtf16Length())};
383 } else {
384 source = icu::UnicodeString {utf::Mutf8AsCString(thisStr->GetDataMUtf8()),
385 static_cast<int32_t>(thisStr->GetLength())};
386 }
387 icu::UnicodeString target;
388 if (cmpStr->IsUtf16()) {
389 target = icu::UnicodeString {cmpStr->GetDataUtf16(), static_cast<int32_t>(cmpStr->GetUtf16Length())};
390 } else {
391 target =
392 icu::UnicodeString {utf::Mutf8AsCString(cmpStr->GetDataMUtf8()), static_cast<int32_t>(cmpStr->GetLength())};
393 }
394 status = U_ZERO_ERROR;
395 std::unique_ptr<icu::Collator> myCollator(icu::Collator::createInstance(locale, status));
396 if (UNLIKELY(U_FAILURE(status))) {
397 icu::UnicodeString dispName;
398 locale.getDisplayName(dispName);
399 std::string localeName;
400 dispName.toUTF8String(localeName);
401 LOG(FATAL, ETS) << "Failed to create the collator for " << localeName;
402 }
403 return myCollator->compare(source, target);
404 }
405
StdCoreStringIndexOfAfter(EtsString * s,uint16_t ch,ets_int fromIndex)406 ets_int StdCoreStringIndexOfAfter(EtsString *s, uint16_t ch, ets_int fromIndex)
407 {
408 return ark::intrinsics::StringIndexOfU16(s, ch, fromIndex);
409 }
410
StdCoreStringIndexOf(EtsString * s,uint16_t ch)411 ets_int StdCoreStringIndexOf(EtsString *s, uint16_t ch)
412 {
413 return StdCoreStringIndexOfAfter(s, ch, 0);
414 }
415
StdCoreStringIndexOfString(EtsString * thisStr,EtsString * patternStr,ets_int fromIndex)416 ets_int StdCoreStringIndexOfString(EtsString *thisStr, EtsString *patternStr, ets_int fromIndex)
417 {
418 ASSERT(thisStr != nullptr && patternStr != nullptr);
419 return thisStr->GetCoreType()->IndexOf(patternStr->GetCoreType(), fromIndex);
420 }
421
StdCoreStringLastIndexOfString(EtsString * thisStr,EtsString * patternStr,ets_int fromIndex)422 ets_int StdCoreStringLastIndexOfString(EtsString *thisStr, EtsString *patternStr, ets_int fromIndex)
423 {
424 ASSERT(thisStr != nullptr && patternStr != nullptr);
425 // "abc".lastIndexOf("ab", -10) will return 0
426 return thisStr->GetCoreType()->LastIndexOf(patternStr->GetCoreType(), std::max(fromIndex, 0));
427 }
428
StdCoreStringCodePointToChar(ets_int codePoint)429 ets_int StdCoreStringCodePointToChar(ets_int codePoint)
430 {
431 icu::UnicodeString uniStr((UChar32)codePoint);
432 uint32_t ret = bit_cast<uint16_t>(uniStr.charAt(0));
433 // if codepoint contains a surrogate pair
434 // encode it into int with higher bits being second char
435 if (uniStr.length() > 1) {
436 constexpr uint32_t BITS_IN_CHAR = 16;
437 ret |= static_cast<uint32_t>(bit_cast<uint16_t>(uniStr.charAt(1))) << BITS_IN_CHAR;
438 }
439 return bit_cast<ets_int>(ret);
440 }
441
StdCoreStringHashCode(EtsString * thisStr)442 int32_t StdCoreStringHashCode(EtsString *thisStr)
443 {
444 ASSERT(thisStr != nullptr);
445 return thisStr->GetCoreType()->GetHashcode();
446 }
447
StdCoreStringIsCompressed(EtsString * thisStr)448 EtsBoolean StdCoreStringIsCompressed(EtsString *thisStr)
449 {
450 ASSERT(thisStr != nullptr);
451 return ToEtsBoolean(thisStr->GetCoreType()->IsMUtf8());
452 }
453
StdCoreStringConcat2(EtsString * str1,EtsString * str2)454 EtsString *StdCoreStringConcat2(EtsString *str1, EtsString *str2)
455 {
456 auto s1 = reinterpret_cast<coretypes::String *>(str1);
457 auto s2 = reinterpret_cast<coretypes::String *>(str2);
458 return reinterpret_cast<EtsString *>(CoreStringConcat2(s1, s2));
459 }
460
StdCoreStringConcat3(EtsString * str1,EtsString * str2,EtsString * str3)461 EtsString *StdCoreStringConcat3(EtsString *str1, EtsString *str2, EtsString *str3)
462 {
463 auto s1 = reinterpret_cast<coretypes::String *>(str1);
464 auto s2 = reinterpret_cast<coretypes::String *>(str2);
465 auto s3 = reinterpret_cast<coretypes::String *>(str3);
466 return reinterpret_cast<EtsString *>(CoreStringConcat3(s1, s2, s3));
467 }
468
StdCoreStringConcat4(EtsString * str1,EtsString * str2,EtsString * str3,EtsString * str4)469 EtsString *StdCoreStringConcat4(EtsString *str1, EtsString *str2, EtsString *str3, EtsString *str4)
470 {
471 auto s1 = reinterpret_cast<coretypes::String *>(str1);
472 auto s2 = reinterpret_cast<coretypes::String *>(str2);
473 auto s3 = reinterpret_cast<coretypes::String *>(str3);
474 auto s4 = reinterpret_cast<coretypes::String *>(str4);
475 return reinterpret_cast<EtsString *>(CoreStringConcat4(s1, s2, s3, s4));
476 }
477
StdCoreStringCompareTo(EtsString * str1,EtsString * str2)478 ets_int StdCoreStringCompareTo(EtsString *str1, EtsString *str2)
479 {
480 /* corner cases */
481 if (str1->GetLength() == 0) {
482 return -str2->GetLength();
483 }
484 if (str2->GetLength() == 0) {
485 return str1->GetLength();
486 }
487
488 /* use the default implementation otherwise */
489 return str1->GetCoreType()->Compare(str2->GetCoreType());
490 }
491
StdCoreStringTrimLeft(EtsString * thisStr)492 EtsString *StdCoreStringTrimLeft(EtsString *thisStr)
493 {
494 return thisStr->TrimLeft();
495 }
496
StdCoreStringTrimRight(EtsString * thisStr)497 EtsString *StdCoreStringTrimRight(EtsString *thisStr)
498 {
499 return thisStr->TrimRight();
500 }
501
StdCoreStringTrim(EtsString * thisStr)502 EtsString *StdCoreStringTrim(EtsString *thisStr)
503 {
504 return thisStr->Trim();
505 }
506
StdCoreStringStartsWith(EtsString * thisStr,EtsString * prefix,EtsInt fromIndex)507 EtsBoolean StdCoreStringStartsWith(EtsString *thisStr, EtsString *prefix, EtsInt fromIndex)
508 {
509 ASSERT(thisStr != nullptr);
510 return thisStr->StartsWith(prefix, fromIndex);
511 }
512
StdCoreStringEndsWith(EtsString * thisStr,EtsString * suffix,EtsInt endIndex)513 EtsBoolean StdCoreStringEndsWith(EtsString *thisStr, EtsString *suffix, EtsInt endIndex)
514 {
515 ASSERT(thisStr != nullptr);
516 return thisStr->EndsWith(suffix, endIndex);
517 }
518
519 /* the allocation routine to create an unitialized string of the given size */
AllocateStringObject(size_t length,bool compressed)520 extern "C" EtsString *AllocateStringObject(size_t length, bool compressed)
521 {
522 auto ctx = Runtime::GetCurrent()->GetLanguageContext(panda_file::SourceLang::ETS);
523 auto vm = Runtime::GetCurrent()->GetPandaVM();
524 ASSERT(vm != nullptr);
525 auto *stringClass = Runtime::GetCurrent()->GetClassLinker()->GetExtension(ctx)->GetClassRoot(ClassRoot::STRING);
526 size_t size =
527 compressed ? coretypes::String::ComputeSizeMUtf8(length) : coretypes::String::ComputeSizeUtf16(length);
528 auto string = reinterpret_cast<EtsString *>(vm->GetHeapManager()->AllocateObject(
529 stringClass, size, DEFAULT_ALIGNMENT, nullptr, mem::ObjectAllocatorBase::ObjMemInitPolicy::NO_INIT));
530 if (string != nullptr) {
531 // After setting length we should have a full barrier, so this write should happens-before barrier
532 TSAN_ANNOTATE_IGNORE_WRITES_BEGIN();
533 auto len = ToNativePtr<uint32_t>(ToUintPtr(string) + coretypes::String::GetLengthOffset());
534 auto hashcode = ToNativePtr<uint32_t>(ToUintPtr(string) + coretypes::String::GetHashcodeOffset());
535 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
536 len[0] = compressed ? (length << 1U) : (length << 1U) | 1U;
537 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
538 hashcode[0] = 0U;
539 TSAN_ANNOTATE_IGNORE_WRITES_END();
540 // Witout full memory barrier it is possible that architectures with
541 // weak memory order can try fetching string legth before it's set
542 arch::FullMemoryBarrier();
543 }
544 return string;
545 }
546
StdCoreStringRepeat(EtsString * str,EtsInt count)547 EtsString *StdCoreStringRepeat(EtsString *str, EtsInt count)
548 {
549 auto length = str->GetLength();
550
551 if (UNLIKELY(count < 0)) {
552 PandaString message = "repeat: count is negative";
553 auto coroutine = EtsCoroutine::GetCurrent();
554 ThrowEtsException(coroutine, panda_file_items::class_descriptors::RANGE_ERROR, message);
555 return nullptr;
556 }
557
558 if (length == 0 || count == 0) {
559 return EtsString::CreateFromUtf8(nullptr, 0);
560 }
561
562 auto thread = ManagedThread::GetCurrent();
563 [[maybe_unused]] HandleScope<ObjectHeader *> scope(thread);
564 VMHandle<coretypes::String> sHandle(thread, str->GetCoreType());
565
566 int size = length * count;
567 auto compressed = str->GetCoreType()->IsMUtf8();
568 auto rep = AllocateStringObject(size, compressed);
569 if (UNLIKELY(rep == nullptr)) {
570 PandaString message = "repeat: memory allocation failed";
571 auto coroutine = EtsCoroutine::GetCurrent();
572 ThrowEtsException(coroutine, panda_file_items::class_descriptors::OUT_OF_MEMORY_ERROR, message);
573 return nullptr;
574 }
575
576 if (compressed) {
577 auto strData = sHandle.GetPtr()->GetDataMUtf8();
578 auto repData = rep->GetDataMUtf8();
579 for (int i = 0; i < count; ++i) {
580 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
581 std::copy_n(strData, length, repData + i * length);
582 }
583 } else {
584 auto strData = sHandle.GetPtr()->GetDataUtf16();
585 auto repData = rep->GetDataUtf16();
586 for (int i = 0; i < count; ++i) {
587 // NOLINTNEXTLINE(cppcoreguidelines-pro-bounds-pointer-arithmetic)
588 std::copy_n(strData, length, repData + i * length);
589 }
590 }
591 return rep;
592 }
593
594 } // namespace ark::ets::intrinsics
595