1 /*
2 * Copyright (c) 2023 Shenzhen Kaihong Digital Industry Development 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 "ecmascript/js_segmenter.h"
17
18 #include <cstring>
19
20 #include "ecmascript/intl/locale_helper.h"
21 #include "ecmascript/object_factory-inl.h"
22
23 namespace panda::ecmascript {
24
SetIcuBreakIterator(JSThread * thread,const JSHandle<JSSegmenter> & segmenter,icu::BreakIterator * icuBreakIterator,const NativePointerCallback & callback)25 void JSSegmenter::SetIcuBreakIterator(JSThread *thread, const JSHandle<JSSegmenter> &segmenter,
26 icu::BreakIterator* icuBreakIterator, const NativePointerCallback &callback)
27 {
28 EcmaVM *ecmaVm = thread->GetEcmaVM();
29 ObjectFactory *factory = ecmaVm->GetFactory();
30
31 ASSERT(icuBreakIterator != nullptr);
32 JSTaggedValue data = segmenter->GetIcuField();
33 if (data.IsJSNativePointer()) {
34 JSNativePointer *native = JSNativePointer::Cast(data.GetTaggedObject());
35 native->ResetExternalPointer(thread, icuBreakIterator);
36 return;
37 }
38 JSHandle<JSNativePointer> pointer = factory->NewJSNativePointer(icuBreakIterator, callback);
39 segmenter->SetIcuField(thread, pointer.GetTaggedValue());
40 }
41
GetAvailableLocales(JSThread * thread)42 JSHandle<TaggedArray> JSSegmenter::GetAvailableLocales(JSThread *thread)
43 {
44 std::vector<std::string> availableStringLocales = intl::LocaleHelper::GetAvailableLocales(thread, nullptr, nullptr);
45 JSHandle<TaggedArray> availableLocales = JSLocale::ConstructLocaleList(thread, availableStringLocales);
46 return availableLocales;
47 }
48
InitializeIcuBreakIterator(JSThread * thread,const JSHandle<JSSegmenter> & segmenter,const icu::Locale & icuLocale,GranularityOption granularity)49 void JSSegmenter::InitializeIcuBreakIterator(JSThread *thread, const JSHandle<JSSegmenter> &segmenter,
50 const icu::Locale &icuLocale, GranularityOption granularity)
51 {
52 UErrorCode status = U_ZERO_ERROR;
53 std::unique_ptr<icu::BreakIterator> icuBreakIterator;
54
55 switch (granularity) {
56 case GranularityOption::GRAPHEME:
57 icuBreakIterator.reset(icu::BreakIterator::createCharacterInstance(icuLocale, status));
58 break;
59 case GranularityOption::WORD:
60 icuBreakIterator.reset(icu::BreakIterator::createWordInstance(icuLocale, status));
61 break;
62 case GranularityOption::SENTENCE:
63 icuBreakIterator.reset(icu::BreakIterator::createSentenceInstance(icuLocale, status));
64 break;
65 default:
66 LOG_ECMA(FATAL) << "this branch is unreachable";
67 UNREACHABLE();
68 }
69 if (U_FAILURE(status) || icuBreakIterator == nullptr) {
70 if (status == UErrorCode::U_MISSING_RESOURCE_ERROR) {
71 THROW_ERROR(thread, ErrorType::REFERENCE_ERROR, "can not find icu data resources");
72 }
73 THROW_ERROR(thread, ErrorType::RANGE_ERROR, "create icu::BreakIterator failed");
74 }
75
76 SetIcuBreakIterator(thread, segmenter, icuBreakIterator.release(), JSSegmenter::FreeIcuBreakIterator);
77 }
78
InitializeSegmenter(JSThread * thread,const JSHandle<JSSegmenter> & segmenter,const JSHandle<JSTaggedValue> & locales,const JSHandle<JSTaggedValue> & options)79 JSHandle<JSSegmenter> JSSegmenter::InitializeSegmenter(JSThread *thread,
80 const JSHandle<JSSegmenter> &segmenter,
81 const JSHandle<JSTaggedValue> &locales,
82 const JSHandle<JSTaggedValue> &options)
83 {
84 ObjectFactory *factory = thread->GetEcmaVM()->GetFactory();
85 auto globalConst = thread->GlobalConstants();
86 // 4. Let requestedLocales be ? CanonicalizeLocaleList(locales).
87 JSHandle<TaggedArray> requestedLocales = intl::LocaleHelper::CanonicalizeLocaleList(thread, locales);
88 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
89
90 // 5. Let options be ? GetOptionsObject(options).
91 JSHandle<JSObject> optionsObject;
92 if (options->IsUndefined()) {
93 optionsObject = factory->CreateNullJSObject();
94 } else if (!options->IsJSObject()) {
95 THROW_TYPE_ERROR_AND_RETURN(thread, "options is not Object", segmenter);
96 } else {
97 optionsObject = JSTaggedValue::ToObject(thread, options);
98 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
99 }
100
101 // 6. Let opt be a new Record.
102 // 7. Let matcher be ? GetOption(options, "localeMatcher", string, « "lookup", "best fit" », "best fit").
103 JSHandle<JSTaggedValue> property = globalConst->GetHandledLocaleMatcherString();
104 auto matcher = JSLocale::GetOptionOfString<LocaleMatcherOption>(
105 thread, optionsObject, property, {LocaleMatcherOption::LOOKUP, LocaleMatcherOption::BEST_FIT},
106 {"lookup", "best fit"}, LocaleMatcherOption::BEST_FIT);
107 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
108
109 // 9. Let localeData be %Segmenter%.[[LocaleData]].
110 // 10. Let r be ResolveLocale(%Segmenter%.[[AvailableLocales]], requestedLocales, opt,
111 // %Segmenter%.[[RelevantExtensionKeys]], localeData).
112 JSHandle<TaggedArray> availableLocales;
113 if (requestedLocales->GetLength() == 0) {
114 availableLocales = factory->EmptyArray();
115 } else {
116 availableLocales = JSSegmenter::GetAvailableLocales(thread);
117 }
118 std::set<std::string> relevantExtensionKeys {""};
119 ResolvedLocale r =
120 JSLocale::ResolveLocale(thread, availableLocales, requestedLocales, matcher, relevantExtensionKeys);
121 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
122
123 // 11. Set segmenter.[[Locale]] to r.[[locale]].
124 icu::Locale icuLocale = r.localeData;
125 JSHandle<EcmaString> localeStr = intl::LocaleHelper::ToLanguageTag(thread, icuLocale);
126 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
127 segmenter->SetLocale(thread, localeStr.GetTaggedValue());
128
129 // 12. Let granularity be ? GetOption(options, "granularity", string, « "grapheme", "word", "sentence" »,
130 // "grapheme").
131 property = globalConst->GetHandledGranularityString();
132 auto granularity = JSLocale::GetOptionOfString<GranularityOption>(thread, optionsObject, property,
133 {GranularityOption::GRAPHEME,
134 GranularityOption::WORD,
135 GranularityOption::SENTENCE},
136 {"grapheme", "word", "sentence"},
137 GranularityOption::GRAPHEME);
138 RETURN_HANDLE_IF_ABRUPT_COMPLETION(JSSegmenter, thread);
139
140 // 13. Set segmenter.[[SegmenterGranularity]] to granularity.
141 segmenter->SetGranularity(granularity);
142 InitializeIcuBreakIterator(thread, segmenter, icuLocale, granularity);
143 // 14. Return segmenter.
144 return segmenter;
145 }
146
GranularityOptionToEcmaString(JSThread * thread,GranularityOption granularity)147 JSHandle<JSTaggedValue> JSSegmenter::GranularityOptionToEcmaString(JSThread *thread, GranularityOption granularity)
148 {
149 JSMutableHandle<JSTaggedValue> result(thread, JSTaggedValue::Undefined());
150 auto globalConst = thread->GlobalConstants();
151 switch (granularity) {
152 case GranularityOption::GRAPHEME:
153 result.Update(globalConst->GetHandledGraphemeString().GetTaggedValue());
154 break;
155 case GranularityOption::WORD:
156 result.Update(globalConst->GetHandledWordString().GetTaggedValue());
157 break;
158 case GranularityOption::SENTENCE:
159 result.Update(globalConst->GetHandledSentenceString().GetTaggedValue());
160 break;
161 default:
162 LOG_ECMA(FATAL) << "this branch is unreachable";
163 UNREACHABLE();
164 }
165 return result;
166 }
167
ResolvedOptions(JSThread * thread,const JSHandle<JSSegmenter> & segmenter,const JSHandle<JSObject> & options)168 void JSSegmenter::ResolvedOptions(JSThread *thread, const JSHandle<JSSegmenter> &segmenter,
169 const JSHandle<JSObject> &options)
170 {
171 auto globalConst = thread->GlobalConstants();
172
173 // [[Locale]]
174 JSHandle<JSTaggedValue> propertyKey = globalConst->GetHandledLocaleString();
175 JSHandle<JSTaggedValue> locale(thread, segmenter->GetLocale());
176 JSObject::CreateDataPropertyOrThrow(thread, options, propertyKey, locale);
177 RETURN_IF_ABRUPT_COMPLETION(thread);
178
179 // [[SegmenterGranularity]]
180 GranularityOption granularity = segmenter->GetGranularity();
181 propertyKey = globalConst->GetHandledGranularityString();
182 JSHandle<JSTaggedValue> granularityString = GranularityOptionToEcmaString(thread, granularity);
183 JSObject::CreateDataPropertyOrThrow(thread, options, propertyKey, granularityString);
184 RETURN_IF_ABRUPT_COMPLETION(thread);
185 }
186 } // namespace panda::ecmascript
187