From 57c899763db9998f6e5a85ea886e762757385553 Mon Sep 17 00:00:00 2001 From: Nikolay Igotti Date: Tue, 14 Feb 2017 11:03:20 +0300 Subject: [PATCH] Switch runtime to UTF16 strings, sigh. (#236) --- .../kotlin/backend/konan/llvm/LlvmUtils.kt | 4 + .../backend/konan/llvm/RTTIGenerator.kt | 2 +- .../backend/konan/llvm/StaticObjects.kt | 2 +- runtime/src/launcher/cpp/launcher.cpp | 5 +- runtime/src/main/cpp/Exceptions.cpp | 5 +- runtime/src/main/cpp/Memory.cpp | 9 - runtime/src/main/cpp/Memory.h | 5 +- runtime/src/main/cpp/Natives.cpp | 158 --------- runtime/src/main/cpp/Natives.h | 9 + runtime/src/main/cpp/String.cpp | 211 +++++++++++ runtime/src/main/cpp/String.h | 20 ++ runtime/src/main/cpp/ToString.cpp | 34 +- runtime/src/main/cpp/utf8.h | 34 ++ runtime/src/main/cpp/utf8/checked.h | 327 +++++++++++++++++ runtime/src/main/cpp/utf8/core.h | 329 ++++++++++++++++++ runtime/src/main/cpp/utf8/unchecked.h | 228 ++++++++++++ runtime/src/main/kotlin/kotlin/String.kt | 5 - 17 files changed, 1182 insertions(+), 205 deletions(-) create mode 100644 runtime/src/main/cpp/String.cpp create mode 100644 runtime/src/main/cpp/String.h create mode 100644 runtime/src/main/cpp/utf8.h create mode 100644 runtime/src/main/cpp/utf8/checked.h create mode 100644 runtime/src/main/cpp/utf8/core.h create mode 100644 runtime/src/main/cpp/utf8/unchecked.h diff --git a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/LlvmUtils.kt b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/LlvmUtils.kt index ad3270372e7..715aa10b88d 100644 --- a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/LlvmUtils.kt +++ b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/LlvmUtils.kt @@ -76,6 +76,10 @@ internal class Int8(val value: Byte) : ConstValue { override val llvm = LLVMConstInt(LLVMInt8Type(), value.toLong(), 1)!! } +internal class Char16(val value: Char) : ConstValue { + override val llvm = LLVMConstInt(LLVMInt16Type(), value.toLong(), 1)!! +} + internal class Int32(val value: Int) : ConstValue { override val llvm = LLVMConstInt(LLVMInt32Type(), value.toLong(), 1)!! } diff --git a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/RTTIGenerator.kt b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/RTTIGenerator.kt index 53937485183..4770fd86994 100644 --- a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/RTTIGenerator.kt +++ b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/RTTIGenerator.kt @@ -70,7 +70,7 @@ internal class RTTIGenerator(override val context: Context) : ContextUtils { "kotlin.FloatArray" to -4, "kotlin.DoubleArray" to -8, "kotlin.BooleanArray" to -1, - "kotlin.String" to -1 + "kotlin.String" to -2 ) private fun getInstanceSize(classType: LLVMTypeRef?, className: FqName) : Int { diff --git a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/StaticObjects.kt b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/StaticObjects.kt index 97f22d58cf0..cd98a9fec88 100644 --- a/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/StaticObjects.kt +++ b/backend.native/compiler/ir/backend.native/src/org/jetbrains/kotlin/backend/konan/llvm/StaticObjects.kt @@ -28,7 +28,7 @@ internal fun StaticData.createKotlinStringLiteral(value: IrConst) = crea internal fun StaticData.createKotlinStringLiteral(value: String): ConstPointer { val name = "kstr:" + value.globalHashBase64 - val elements = value.toByteArray(Charsets.UTF_8).map(::Int8) + val elements = value.toCharArray().map(::Char16) val objRef = createKotlinArray(KonanPlatform.builtIns.stringType, elements) diff --git a/runtime/src/launcher/cpp/launcher.cpp b/runtime/src/launcher/cpp/launcher.cpp index eaf11f0f927..ea315ec67d4 100644 --- a/runtime/src/launcher/cpp/launcher.cpp +++ b/runtime/src/launcher/cpp/launcher.cpp @@ -2,6 +2,7 @@ #include "Memory.h" #include "Natives.h" +#include "String.h" #include "Types.h" //--- Setup args --------------------------------------------------------------// @@ -11,8 +12,8 @@ OBJ_GETTER(setupArgs, int argc, char** argv) { ObjHeader* result = AllocArrayInstance(theArrayTypeInfo, argc - 1, OBJ_RESULT); ArrayHeader* array = result->array(); for (int index = 1; index < argc; index++) { - AllocStringInstance(argv[index], strlen(argv[index]), - ArrayAddressOfElementAt(array, index - 1)); + CreateStringFromCString( + argv[index], ArrayAddressOfElementAt(array, index - 1)); } return result; } diff --git a/runtime/src/main/cpp/Exceptions.cpp b/runtime/src/main/cpp/Exceptions.cpp index 65fed9011bd..4d5cb749ab0 100644 --- a/runtime/src/main/cpp/Exceptions.cpp +++ b/runtime/src/main/cpp/Exceptions.cpp @@ -6,6 +6,7 @@ #include "Exceptions.h" #include "Memory.h" #include "Natives.h" +#include "String.h" #include "Types.h" class KotlinException { @@ -53,8 +54,8 @@ OBJ_GETTER0(GetCurrentStackTrace) { ObjHeader* result = AllocArrayInstance(theArrayTypeInfo, size, OBJ_RESULT); ArrayHeader* array = result->array(); for (int index = 0; index < size; ++index) { - AllocStringInstance(symbols[index], strlen(symbols[index]), - ArrayAddressOfElementAt(array, index)); + CreateStringFromCString( + symbols[index], ArrayAddressOfElementAt(array, index)); } return result; } diff --git a/runtime/src/main/cpp/Memory.cpp b/runtime/src/main/cpp/Memory.cpp index f4492dcdd1b..79321fcd6a0 100644 --- a/runtime/src/main/cpp/Memory.cpp +++ b/runtime/src/main/cpp/Memory.cpp @@ -534,15 +534,6 @@ OBJ_GETTER(AllocArrayInstance, const TypeInfo* type_info, uint32_t elements) { RETURN_OBJ(ArrayContainer(type_info, elements).GetPlace()->obj()); } -OBJ_GETTER(AllocStringInstance, const char* data, uint32_t length) { - ArrayHeader* array = ArrayContainer(theStringTypeInfo, length).GetPlace(); - memcpy( - ByteArrayAddressOfElementAt(array, 0), - data, - length); - RETURN_OBJ(array->obj()); -} - OBJ_GETTER(InitInstance, ObjHeader** location, const TypeInfo* type_info, void (*ctor)(ObjHeader*)) { ObjHeader* sentinel = reinterpret_cast(1); diff --git a/runtime/src/main/cpp/Memory.h b/runtime/src/main/cpp/Memory.h index afc9a99ad88..76f74fe0b57 100644 --- a/runtime/src/main/cpp/Memory.h +++ b/runtime/src/main/cpp/Memory.h @@ -321,9 +321,8 @@ void DeinitMemory(); // OBJ_GETTER(AllocInstance, const TypeInfo* type_info) RUNTIME_NOTHROW; OBJ_GETTER(AllocArrayInstance, const TypeInfo* type_info, uint32_t elements) RUNTIME_NOTHROW; -OBJ_GETTER(AllocStringInstance, const char* data, uint32_t length) RUNTIME_NOTHROW; -OBJ_GETTER(InitInstance, - ObjHeader** location, const TypeInfo* type_info, void (*ctor)(ObjHeader*)); +OBJ_GETTER(InitInstance, ObjHeader** location, const TypeInfo* type_info, + void (*ctor)(ObjHeader*)); // // Object reference management. diff --git a/runtime/src/main/cpp/Natives.cpp b/runtime/src/main/cpp/Natives.cpp index 81dc436da95..75c3bb5ad01 100644 --- a/runtime/src/main/cpp/Natives.cpp +++ b/runtime/src/main/cpp/Natives.cpp @@ -3,10 +3,8 @@ #include #include #include -#include #include "Assert.h" -#include "City.h" #include "Exceptions.h" #include "Memory.h" #include "Natives.h" @@ -25,162 +23,6 @@ KInt Kotlin_Any_hashCode(KConstRef thiz) { return reinterpret_cast(thiz); } -// io/Console.kt -void Kotlin_io_Console_print(KString message) { - RuntimeAssert(message->type_info() == theStringTypeInfo, "Must use a string"); - // TODO: system stdout must be aware about UTF-8. - write(STDOUT_FILENO, ByteArrayAddressOfElementAt(message, 0), message->count_); -} - -void Kotlin_io_Console_println(KString message) { - RuntimeAssert(message->type_info() == theStringTypeInfo, "Must use a string"); - // TODO: system stdout must be aware about UTF-8. - write(STDOUT_FILENO, ByteArrayAddressOfElementAt(message, 0), message->count_); - Kotlin_io_Console_println0(); -} - -void Kotlin_io_Console_println0() { - write(STDOUT_FILENO, "\n", 1); -} - -OBJ_GETTER0(Kotlin_io_Console_readLine) { - char data[2048]; - if (!fgets(data, sizeof(data) - 1, stdin)) { - return nullptr; - } - int32_t length = strlen(data); - ArrayHeader* result = ArrayContainer(theStringTypeInfo, length).GetPlace(); - memcpy( - ByteArrayAddressOfElementAt(result, 0), - data, length); - RETURN_OBJ(result->obj()); -} - -// String.kt -KInt Kotlin_String_compareTo(KString thiz, KString other) { - return memcmp(ByteArrayAddressOfElementAt(thiz, 0), - ByteArrayAddressOfElementAt(other, 0), - thiz->count_ < other->count_ ? thiz->count_ : other->count_); -} - -KChar Kotlin_String_get(KString thiz, KInt index) { - // TODO: support full UTF-8. - if (static_cast(index) >= thiz->count_) { - ThrowArrayIndexOutOfBoundsException(); - } - return *ByteArrayAddressOfElementAt(thiz, index); -} - -KInt Kotlin_String_getStringLength(KString thiz) { - return thiz->count_; -} - -OBJ_GETTER(Kotlin_String_fromUtf8Array, KConstRef thiz, KInt start, KInt size) { - const ArrayHeader* array = thiz->array(); - RuntimeAssert(array->type_info() == theByteArrayTypeInfo, "Must use a byte array"); - if (start < 0 || size < 0 || start + size > array->count_) { - ThrowArrayIndexOutOfBoundsException(); - } - - if (size == 0) { - RETURN_RESULT_OF0(TheEmptyString); - } - - // TODO: support full UTF-8. - ArrayHeader* result = ArrayContainer(theStringTypeInfo, size).GetPlace(); - memcpy( - ByteArrayAddressOfElementAt(result, 0), - ByteArrayAddressOfElementAt(array, start), - size); - RETURN_OBJ(result->obj()); -} - -OBJ_GETTER(Kotlin_String_fromCharArray, KConstRef thiz, KInt start, KInt size) { - const ArrayHeader* array = thiz->array(); - RuntimeAssert(array->type_info() == theCharArrayTypeInfo, "Must use a byte array"); - if (start < 0 || size < 0 || start + size > array->count_) { - ThrowArrayIndexOutOfBoundsException(); - } - - if (size == 0) { - RETURN_RESULT_OF0(TheEmptyString); - } - - // TODO: support full UTF-8. - ArrayHeader* result = ArrayContainer(theStringTypeInfo, size).GetPlace(); - for (KInt index = 0; index < size; ++index) { - *ByteArrayAddressOfElementAt(result, index) = - *PrimitiveArrayAddressOfElementAt(array, start + index); - } - RETURN_OBJ(result->obj()); -} - -OBJ_GETTER(Kotlin_String_toCharArray, KString string) { - // TODO: support full UTF-8. - ArrayHeader* result = ArrayContainer( - theCharArrayTypeInfo, string->count_).GetPlace(); - for (int index = 0; index < string->count_; ++index) { - *PrimitiveArrayAddressOfElementAt(result, index) = - *ByteArrayAddressOfElementAt(string, index); - } - RETURN_OBJ(result->obj()); -} - - -OBJ_GETTER(Kotlin_String_plusImpl, KString thiz, KString other) { - // TODO: support UTF-8 - RuntimeAssert(thiz != nullptr, "this cannot be null"); - RuntimeAssert(other != nullptr, "other cannot be null"); - RuntimeAssert(thiz->type_info() == theStringTypeInfo, "Must be a string"); - RuntimeAssert(other->type_info() == theStringTypeInfo, "Must be a string"); - uint32_t result_length = thiz->count_ + other->count_; - ArrayHeader* result = ArrayContainer( - theStringTypeInfo, result_length).GetPlace(); - memcpy( - ByteArrayAddressOfElementAt(result, 0), - ByteArrayAddressOfElementAt(thiz, 0), - thiz->count_); - memcpy( - ByteArrayAddressOfElementAt(result, thiz->count_), - ByteArrayAddressOfElementAt(other, 0), - other->count_); - RETURN_OBJ(result->obj()); -} - -KBoolean Kotlin_String_equals(KString thiz, KConstRef other) { - if (other == nullptr || other->type_info() != theStringTypeInfo) return false; - // Important, due to literal internalization. - KString otherString = other->array(); - if (thiz == otherString) return true; - return thiz->count_ == otherString->count_ && - memcmp(ByteArrayAddressOfElementAt(thiz, 0), - ByteArrayAddressOfElementAt(otherString, 0), - thiz->count_) == 0; -} - -KInt Kotlin_String_hashCode(KString thiz) { - // TODO: consider caching strings hashes. - // TODO: maybe use some simpler hashing algorithm? - return CityHash64(ByteArrayAddressOfElementAt(thiz, 0), thiz->count_); -} - -OBJ_GETTER(Kotlin_String_subSequence, KString thiz, KInt startIndex, KInt endIndex) { - if (startIndex < 0 || endIndex >= thiz->count_ || startIndex > endIndex) { - // TODO: is it correct exception? - ThrowArrayIndexOutOfBoundsException(); - } - if (startIndex == endIndex) { - RETURN_RESULT_OF0(TheEmptyString); - } - // TODO: support UTF-8. - KInt length = endIndex - startIndex; - ArrayHeader* result = ArrayContainer(theStringTypeInfo, length).GetPlace(); - memcpy(ByteArrayAddressOfElementAt(result, 0), - ByteArrayAddressOfElementAt(thiz, startIndex), - length); - RETURN_OBJ(result->obj()); -} - OBJ_GETTER0(Kotlin_getCurrentStackTrace) { RETURN_RESULT_OF0(GetCurrentStackTrace); } diff --git a/runtime/src/main/cpp/Natives.h b/runtime/src/main/cpp/Natives.h index 17b57d0918d..ea90bd600ca 100644 --- a/runtime/src/main/cpp/Natives.h +++ b/runtime/src/main/cpp/Natives.h @@ -25,6 +25,15 @@ inline const KByte* ByteArrayAddressOfElementAt( return reinterpret_cast(obj + 1) + index; } +inline KChar* CharArrayAddressOfElementAt(ArrayHeader* obj, KInt index) { + return reinterpret_cast(obj + 1) + index; +} + +inline const KChar* CharArrayAddressOfElementAt( + const ArrayHeader* obj, KInt index) { + return reinterpret_cast(obj + 1) + index; +} + // Consider aligning of base to sizeof(T). template inline T* PrimitiveArrayAddressOfElementAt(ArrayHeader* obj, KInt index) { diff --git a/runtime/src/main/cpp/String.cpp b/runtime/src/main/cpp/String.cpp new file mode 100644 index 00000000000..f08bb1ff7fa --- /dev/null +++ b/runtime/src/main/cpp/String.cpp @@ -0,0 +1,211 @@ +#include +#include +#include +#include +#include +#include + +#include +#include + +#include "Assert.h" +#include "City.h" +#include "Exceptions.h" +#include "Memory.h" +#include "Natives.h" +#include "String.h" +#include "Types.h" + +#include "utf8.h" + +namespace { + +// Container with STL-like iterator over raw data. +template +class Raw { + public: + Raw(T* start, uint32_t count) : start_(start), end_(start + count) {} + + T* begin() { return start_; } + T* end() { return end_; } + + private: + T* start_; + T* end_; +}; + + +OBJ_GETTER(utf8ToUtf16, Raw rawString) { + uint32_t charCount = utf8::distance(rawString.begin(), rawString.end()); + ArrayHeader* result = AllocArrayInstance( + theStringTypeInfo, charCount, OBJ_RESULT)->array(); + Raw rawResult(CharArrayAddressOfElementAt(result, 0), charCount); + auto convertResult = + utf8::utf8to16(rawString.begin(), rawString.end(), rawResult.begin()); + RuntimeAssert(rawResult.end() == convertResult, "Must properly fit"); + RETURN_OBJ(result->obj()); +} + +} // namespace + +extern "C" { + +OBJ_GETTER(CreateStringFromCString, const char* cstring) { + RETURN_RESULT_OF(utf8ToUtf16, Raw(cstring, strlen(cstring))); +} + +OBJ_GETTER(CreateStringFromUtf8, const char* utf8, uint32_t lengthBytes) { + RETURN_RESULT_OF(utf8ToUtf16, Raw(utf8, lengthBytes)); +} + +// String.kt +KInt Kotlin_String_compareTo(KString thiz, KString other) { + return memcmp( + CharArrayAddressOfElementAt(thiz, 0), + CharArrayAddressOfElementAt(other, 0), + (thiz->count_ < other->count_ ? thiz->count_ : other->count_) * sizeof(KChar)); +} + +KChar Kotlin_String_get(KString thiz, KInt index) { + if (static_cast(index) >= thiz->count_) { + ThrowArrayIndexOutOfBoundsException(); + } + return *CharArrayAddressOfElementAt(thiz, index); +} + +KInt Kotlin_String_getStringLength(KString thiz) { + return thiz->count_; +} + +OBJ_GETTER(Kotlin_String_fromUtf8Array, KConstRef thiz, KInt start, KInt size) { + const ArrayHeader* array = thiz->array(); + RuntimeAssert(array->type_info() == theByteArrayTypeInfo, "Must use a byte array"); + if (start < 0 || size < 0 || + start + size > array->count_ || start + size < start) { + ThrowArrayIndexOutOfBoundsException(); + } + if (size == 0) { + RETURN_RESULT_OF0(TheEmptyString); + } + Raw rawString( + reinterpret_cast( + ByteArrayAddressOfElementAt(array, start)), size); + RETURN_RESULT_OF(utf8ToUtf16, rawString); +} + +OBJ_GETTER(Kotlin_String_fromCharArray, KConstRef thiz, KInt start, KInt size) { + const ArrayHeader* array = thiz->array(); + RuntimeAssert(array->type_info() == theCharArrayTypeInfo, "Must use a char array"); + if (start < 0 || size < 0 || + start + size > array->count_ || start + size < start) { + ThrowArrayIndexOutOfBoundsException(); + } + + if (size == 0) { + RETURN_RESULT_OF0(TheEmptyString); + } + + ArrayHeader* result = AllocArrayInstance( + theStringTypeInfo, size, OBJ_RESULT)->array(); + memcpy(CharArrayAddressOfElementAt(result, 0), + CharArrayAddressOfElementAt(array, start), + size * sizeof(KChar)); + RETURN_OBJ(result->obj()); +} + +OBJ_GETTER(Kotlin_String_toCharArray, KString string) { + ArrayHeader* result = AllocArrayInstance( + theCharArrayTypeInfo, string->count_, OBJ_RESULT)->array(); + memcpy(CharArrayAddressOfElementAt(result, 0), + CharArrayAddressOfElementAt(string, 0), + string->count_ * sizeof(KChar)); + RETURN_OBJ(result->obj()); +} + +OBJ_GETTER(Kotlin_String_plusImpl, KString thiz, KString other) { + RuntimeAssert(thiz != nullptr, "this cannot be null"); + RuntimeAssert(other != nullptr, "other cannot be null"); + RuntimeAssert(thiz->type_info() == theStringTypeInfo, "Must be a string"); + RuntimeAssert(other->type_info() == theStringTypeInfo, "Must be a string"); + KInt result_length = thiz->count_ + other->count_; + if (result_length < thiz->count_ || result_length < other->count_) { + ThrowArrayIndexOutOfBoundsException(); + } + ArrayHeader* result = AllocArrayInstance( + theStringTypeInfo, result_length, OBJ_RESULT)->array(); + memcpy( + CharArrayAddressOfElementAt(result, 0), + CharArrayAddressOfElementAt(thiz, 0), + thiz->count_ * sizeof(KChar)); + memcpy( + CharArrayAddressOfElementAt(result, thiz->count_), + CharArrayAddressOfElementAt(other, 0), + other->count_ * sizeof(KChar)); + RETURN_OBJ(result->obj()); +} + +KBoolean Kotlin_String_equals(KString thiz, KConstRef other) { + if (other == nullptr || other->type_info() != theStringTypeInfo) return false; + // Important, due to literal internalization. + KString otherString = other->array(); + if (thiz == otherString) return true; + return thiz->count_ == otherString->count_ && + memcmp(CharArrayAddressOfElementAt(thiz, 0), + CharArrayAddressOfElementAt(otherString, 0), + thiz->count_ * sizeof(KChar)) == 0; +} + +KInt Kotlin_String_hashCode(KString thiz) { + // TODO: consider caching strings hashes. + // TODO: maybe use some simpler hashing algorithm? + return CityHash64( + CharArrayAddressOfElementAt(thiz, 0), thiz->count_ * sizeof(KChar)); +} + +OBJ_GETTER(Kotlin_String_subSequence, KString thiz, KInt startIndex, KInt endIndex) { + if (startIndex < 0 || endIndex > thiz->count_ || startIndex > endIndex) { + // TODO: is it correct exception? + ThrowArrayIndexOutOfBoundsException(); + } + if (startIndex == endIndex) { + RETURN_RESULT_OF0(TheEmptyString); + } + KInt length = endIndex - startIndex; + ArrayHeader* result = AllocArrayInstance( + theStringTypeInfo, length, OBJ_RESULT)->array(); + memcpy(CharArrayAddressOfElementAt(result, 0), + CharArrayAddressOfElementAt(thiz, startIndex), + length * sizeof(KChar)); + RETURN_OBJ(result->obj()); +} + +// io/Console.kt +void Kotlin_io_Console_print(KString message) { + RuntimeAssert(message->type_info() == theStringTypeInfo, "Must use a string"); + // TODO: system stdout must be aware about UTF-8. + Raw utf16( + CharArrayAddressOfElementAt(message, 0), message->count_); + std::string utf8; + utf8::utf16to8(utf16.begin(), utf16.end(), back_inserter(utf8)); + write(STDOUT_FILENO, utf8.c_str(), utf8.size()); +} + +void Kotlin_io_Console_println(KString message) { + Kotlin_io_Console_print(message); + Kotlin_io_Console_println0(); +} + +void Kotlin_io_Console_println0() { + write(STDOUT_FILENO, "\n", 1); +} + +OBJ_GETTER0(Kotlin_io_Console_readLine) { + char data[4096]; + if (!fgets(data, sizeof(data) - 1, stdin)) { + return nullptr; + } + RETURN_RESULT_OF(CreateStringFromCString, data); +} + + +} // extern "C" diff --git a/runtime/src/main/cpp/String.h b/runtime/src/main/cpp/String.h new file mode 100644 index 00000000000..93c2006ecba --- /dev/null +++ b/runtime/src/main/cpp/String.h @@ -0,0 +1,20 @@ +#ifndef RUNTIME_STRING_H +#define RUNTIME_STRING_H + +#include "Common.h" +#include "Memory.h" +#include "Types.h" +#include "TypeInfo.h" + +#ifdef __cplusplus +extern "C" { +#endif + +OBJ_GETTER(CreateStringFromCString, const char* cstring); +OBJ_GETTER(CreateStringFromUtf8, const char* utf8, uint32_t size); + +#ifdef __cplusplus +} +#endif + +#endif // RUNTIME_STRING_H diff --git a/runtime/src/main/cpp/ToString.cpp b/runtime/src/main/cpp/ToString.cpp index 313e02bfc47..d7aa605b875 100644 --- a/runtime/src/main/cpp/ToString.cpp +++ b/runtime/src/main/cpp/ToString.cpp @@ -5,23 +5,9 @@ #include "Exceptions.h" #include "Memory.h" #include "Natives.h" +#include "String.h" #include "Types.h" -namespace { - -OBJ_GETTER(makeString, const char* cstring) { - uint32_t length = strlen(cstring); - ArrayHeader* result = AllocArrayInstance( - theStringTypeInfo, length, OBJ_RESULT)->array(); - memcpy( - ByteArrayAddressOfElementAt(result, 0), - cstring, - length); - RETURN_OBJ(result->obj()); -} - -} // namespace - extern "C" { OBJ_GETTER(Kotlin_Any_toString, KConstRef thiz) { @@ -29,55 +15,55 @@ OBJ_GETTER(Kotlin_Any_toString, KConstRef thiz) { snprintf(cstring, sizeof(cstring), "%s %p type %p", IsArray(thiz) ? "array" : "object", thiz, thiz->type_info_); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Byte_toString, KByte value) { char cstring[8]; snprintf(cstring, sizeof(cstring), "%d", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Char_toString, KChar value) { char cstring[5]; // TODO: support UTF-8. snprintf(cstring, sizeof(cstring), "%c", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Short_toString, KShort value) { char cstring[8]; snprintf(cstring, sizeof(cstring), "%d", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Int_toString, KInt value) { char cstring[16]; snprintf(cstring, sizeof(cstring), "%d", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Long_toString, KLong value) { char cstring[32]; snprintf(cstring, sizeof(cstring), "%lld", static_cast(value)); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } // TODO: use David Gay's dtoa() here instead. It's *very* big and ugly. OBJ_GETTER(Kotlin_Float_toString, KFloat value) { char cstring[32]; snprintf(cstring, sizeof(cstring), "%G", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Double_toString, KDouble value) { char cstring[32]; snprintf(cstring, sizeof(cstring), "%G", value); - RETURN_RESULT_OF(makeString, cstring); + RETURN_RESULT_OF(CreateStringFromCString, cstring); } OBJ_GETTER(Kotlin_Boolean_toString, KBoolean value) { - RETURN_RESULT_OF(makeString, value ? "true" : "false"); + RETURN_RESULT_OF(CreateStringFromCString, value ? "true" : "false"); } } // extern "C" diff --git a/runtime/src/main/cpp/utf8.h b/runtime/src/main/cpp/utf8.h new file mode 100644 index 00000000000..4e4451403f6 --- /dev/null +++ b/runtime/src/main/cpp/utf8.h @@ -0,0 +1,34 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "utf8/checked.h" +#include "utf8/unchecked.h" + +#endif // header guard diff --git a/runtime/src/main/cpp/utf8/checked.h b/runtime/src/main/cpp/utf8/checked.h new file mode 100644 index 00000000000..13311551381 --- /dev/null +++ b/runtime/src/main/cpp/utf8/checked.h @@ -0,0 +1,327 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" +#include + +namespace utf8 +{ + // Base for the exceptions that may be thrown from the library + class exception : public ::std::exception { + }; + + // Exceptions that may be thrown from the library functions. + class invalid_code_point : public exception { + uint32_t cp; + public: + invalid_code_point(uint32_t cp) : cp(cp) {} + virtual const char* what() const throw() { return "Invalid code point"; } + uint32_t code_point() const {return cp;} + }; + + class invalid_utf8 : public exception { + uint8_t u8; + public: + invalid_utf8 (uint8_t u) : u8(u) {} + virtual const char* what() const throw() { return "Invalid UTF-8"; } + uint8_t utf8_octet() const {return u8;} + }; + + class invalid_utf16 : public exception { + uint16_t u16; + public: + invalid_utf16 (uint16_t u) : u16(u) {} + virtual const char* what() const throw() { return "Invalid UTF-16"; } + uint16_t utf16_word() const {return u16;} + }; + + class not_enough_room : public exception { + public: + virtual const char* what() const throw() { return "Not enough space"; } + }; + + /// The library API - functions intended to be called by the users + + template + octet_iterator append(uint32_t cp, octet_iterator result) + { + if (!utf8::internal::is_code_point_valid(cp)) + throw invalid_code_point(cp); + + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f) | 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + template + output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement) + { + while (start != end) { + octet_iterator sequence_start = start; + internal::utf_error err_code = utf8::internal::validate_next(start, end); + switch (err_code) { + case internal::UTF8_OK : + for (octet_iterator it = sequence_start; it != start; ++it) + *out++ = *it; + break; + case internal::NOT_ENOUGH_ROOM: + throw not_enough_room(); + case internal::INVALID_LEAD: + out = utf8::append (replacement, out); + ++start; + break; + case internal::INCOMPLETE_SEQUENCE: + case internal::OVERLONG_SEQUENCE: + case internal::INVALID_CODE_POINT: + out = utf8::append (replacement, out); + ++start; + // just one replacement mark for the sequence + while (start != end && utf8::internal::is_trail(*start)) + ++start; + break; + } + } + return out; + } + + template + inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out) + { + static const uint32_t replacement_marker = utf8::internal::mask16(0xfffd); + return utf8::replace_invalid(start, end, out, replacement_marker); + } + + template + uint32_t next(octet_iterator& it, octet_iterator end) + { + uint32_t cp = 0; + internal::utf_error err_code = utf8::internal::validate_next(it, end, cp); + switch (err_code) { + case internal::UTF8_OK : + break; + case internal::NOT_ENOUGH_ROOM : + throw not_enough_room(); + case internal::INVALID_LEAD : + case internal::INCOMPLETE_SEQUENCE : + case internal::OVERLONG_SEQUENCE : + throw invalid_utf8(*it); + case internal::INVALID_CODE_POINT : + throw invalid_code_point(cp); + } + return cp; + } + + template + uint32_t peek_next(octet_iterator it, octet_iterator end) + { + return utf8::next(it, end); + } + + template + uint32_t prior(octet_iterator& it, octet_iterator start) + { + // can't do much if it == start + if (it == start) + throw not_enough_room(); + + octet_iterator end = it; + // Go back until we hit either a lead octet or start + while (utf8::internal::is_trail(*(--it))) + if (it == start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + return utf8::peek_next(it, end); + } + + /// Deprecated in versions that include "prior" + template + uint32_t previous(octet_iterator& it, octet_iterator pass_start) + { + octet_iterator end = it; + while (utf8::internal::is_trail(*(--it))) + if (it == pass_start) + throw invalid_utf8(*it); // error - no lead byte in the sequence + octet_iterator temp = it; + return utf8::next(temp, end); + } + + template + void advance (octet_iterator& it, distance_type n, octet_iterator end) + { + for (distance_type i = 0; i < n; ++i) + utf8::next(it, end); + } + + template + typename std::iterator_traits::difference_type + distance (octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::next(first, last); + return dist; + } + + template + octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + uint32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + if (start != end) { + uint32_t trail_surrogate = utf8::internal::mask16(*start++); + if (utf8::internal::is_trail_surrogate(trail_surrogate)) + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + else + throw invalid_utf16(static_cast(trail_surrogate)); + } + else + throw invalid_utf16(static_cast(cp)); + + } + // Lone trail surrogate + else if (utf8::internal::is_trail_surrogate(cp)) + throw invalid_utf16(static_cast(cp)); + + result = utf8::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start != end) { + uint32_t cp = utf8::next(start, end); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start != end) + (*result++) = utf8::next(start, end); + + return result; + } + + // The iterator class + template + class iterator : public std::iterator { + octet_iterator it; + octet_iterator range_start; + octet_iterator range_end; + public: + iterator () {} + explicit iterator (const octet_iterator& octet_it, + const octet_iterator& range_start, + const octet_iterator& range_end) : + it(octet_it), range_start(range_start), range_end(range_end) + { + if (it < range_start || it > range_end) + throw std::out_of_range("Invalid utf-8 iterator position"); + } + // the default "big three" are OK + octet_iterator base () const { return it; } + uint32_t operator * () const + { + octet_iterator temp = it; + return utf8::next(temp, range_end); + } + bool operator == (const iterator& rhs) const + { + if (range_start != rhs.range_start || range_end != rhs.range_end) + throw std::logic_error("Comparing utf-8 iterators defined with different ranges"); + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + utf8::next(it, range_end); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + utf8::next(it, range_end); + return temp; + } + iterator& operator -- () + { + utf8::prior(it, range_start); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::prior(it, range_start); + return temp; + } + }; // class iterator + +} // namespace utf8 + +#endif //header guard + + diff --git a/runtime/src/main/cpp/utf8/core.h b/runtime/src/main/cpp/utf8/core.h new file mode 100644 index 00000000000..693d388c078 --- /dev/null +++ b/runtime/src/main/cpp/utf8/core.h @@ -0,0 +1,329 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include + +namespace utf8 +{ + // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers + // You may need to change them to match your system. + // These typedefs have the same names as ones from cstdint, or boost/cstdint + typedef unsigned char uint8_t; + typedef unsigned short uint16_t; + typedef unsigned int uint32_t; + +// Helper code - not intended to be directly called by the library users. May be changed at any time +namespace internal +{ + // Unicode constants + // Leading (high) surrogates: 0xd800 - 0xdbff + // Trailing (low) surrogates: 0xdc00 - 0xdfff + const uint16_t LEAD_SURROGATE_MIN = 0xd800u; + const uint16_t LEAD_SURROGATE_MAX = 0xdbffu; + const uint16_t TRAIL_SURROGATE_MIN = 0xdc00u; + const uint16_t TRAIL_SURROGATE_MAX = 0xdfffu; + const uint16_t LEAD_OFFSET = LEAD_SURROGATE_MIN - (0x10000 >> 10); + const uint32_t SURROGATE_OFFSET = 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN; + + // Maximum valid value for a Unicode code point + const uint32_t CODE_POINT_MAX = 0x0010ffffu; + + template + inline uint8_t mask8(octet_type oc) + { + return static_cast(0xff & oc); + } + template + inline uint16_t mask16(u16_type oc) + { + return static_cast(0xffff & oc); + } + template + inline bool is_trail(octet_type oc) + { + return ((utf8::internal::mask8(oc) >> 6) == 0x2); + } + + template + inline bool is_lead_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= LEAD_SURROGATE_MAX); + } + + template + inline bool is_trail_surrogate(u16 cp) + { + return (cp >= TRAIL_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_surrogate(u16 cp) + { + return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX); + } + + template + inline bool is_code_point_valid(u32 cp) + { + return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp)); + } + + template + inline typename std::iterator_traits::difference_type + sequence_length(octet_iterator lead_it) + { + uint8_t lead = utf8::internal::mask8(*lead_it); + if (lead < 0x80) + return 1; + else if ((lead >> 5) == 0x6) + return 2; + else if ((lead >> 4) == 0xe) + return 3; + else if ((lead >> 3) == 0x1e) + return 4; + else + return 0; + } + + template + inline bool is_overlong_sequence(uint32_t cp, octet_difference_type length) + { + if (cp < 0x80) { + if (length != 1) + return true; + } + else if (cp < 0x800) { + if (length != 2) + return true; + } + else if (cp < 0x10000) { + if (length != 3) + return true; + } + + return false; + } + + enum utf_error {UTF8_OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT}; + + /// Helper for get_sequence_x + template + utf_error increase_safely(octet_iterator& it, octet_iterator end) + { + if (++it == end) + return NOT_ENOUGH_ROOM; + + if (!utf8::internal::is_trail(*it)) + return INCOMPLETE_SEQUENCE; + + return UTF8_OK; + } + + #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;} + + /// get_sequence_x functions decode utf-8 sequences of the length x + template + utf_error get_sequence_1(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + return UTF8_OK; + } + + template + utf_error get_sequence_2(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 6) & 0x7ff) + ((*it) & 0x3f); + + return UTF8_OK; + } + + template + utf_error get_sequence_3(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (*it) & 0x3f; + + return UTF8_OK; + } + + template + utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + if (it == end) + return NOT_ENOUGH_ROOM; + + code_point = utf8::internal::mask8(*it); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point = ((code_point << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (utf8::internal::mask8(*it) << 6) & 0xfff; + + UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end) + + code_point += (*it) & 0x3f; + + return UTF8_OK; + } + + #undef UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR + + template + utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t& code_point) + { + // Save the original value of it so we can go back in case of failure + // Of course, it does not make much sense with i.e. stream iterators + octet_iterator original_it = it; + + uint32_t cp = 0; + // Determine the sequence length based on the lead octet + typedef typename std::iterator_traits::difference_type octet_difference_type; + const octet_difference_type length = utf8::internal::sequence_length(it); + + // Get trail octets and calculate the code point + utf_error err = UTF8_OK; + switch (length) { + case 0: + return INVALID_LEAD; + case 1: + err = utf8::internal::get_sequence_1(it, end, cp); + break; + case 2: + err = utf8::internal::get_sequence_2(it, end, cp); + break; + case 3: + err = utf8::internal::get_sequence_3(it, end, cp); + break; + case 4: + err = utf8::internal::get_sequence_4(it, end, cp); + break; + } + + if (err == UTF8_OK) { + // Decoding succeeded. Now, security checks... + if (utf8::internal::is_code_point_valid(cp)) { + if (!utf8::internal::is_overlong_sequence(cp, length)){ + // Passed! Return here. + code_point = cp; + ++it; + return UTF8_OK; + } + else + err = OVERLONG_SEQUENCE; + } + else + err = INVALID_CODE_POINT; + } + + // Failure branch - restore the original value of the iterator + it = original_it; + return err; + } + + template + inline utf_error validate_next(octet_iterator& it, octet_iterator end) { + uint32_t ignored; + return utf8::internal::validate_next(it, end, ignored); + } + +} // namespace internal + + /// The library API - functions intended to be called by the users + + // Byte order mark + const uint8_t bom[] = {0xef, 0xbb, 0xbf}; + + template + octet_iterator find_invalid(octet_iterator start, octet_iterator end) + { + octet_iterator result = start; + while (result != end) { + utf8::internal::utf_error err_code = utf8::internal::validate_next(result, end); + if (err_code != internal::UTF8_OK) + return result; + } + return result; + } + + template + inline bool is_valid(octet_iterator start, octet_iterator end) + { + return (utf8::find_invalid(start, end) == end); + } + + template + inline bool starts_with_bom (octet_iterator it, octet_iterator end) + { + return ( + ((it != end) && (utf8::internal::mask8(*it++)) == bom[0]) && + ((it != end) && (utf8::internal::mask8(*it++)) == bom[1]) && + ((it != end) && (utf8::internal::mask8(*it)) == bom[2]) + ); + } + + //Deprecated in release 2.3 + template + inline bool is_bom (octet_iterator it) + { + return ( + (utf8::internal::mask8(*it++)) == bom[0] && + (utf8::internal::mask8(*it++)) == bom[1] && + (utf8::internal::mask8(*it)) == bom[2] + ); + } +} // namespace utf8 + +#endif // header guard + + diff --git a/runtime/src/main/cpp/utf8/unchecked.h b/runtime/src/main/cpp/utf8/unchecked.h new file mode 100644 index 00000000000..cb2427166b1 --- /dev/null +++ b/runtime/src/main/cpp/utf8/unchecked.h @@ -0,0 +1,228 @@ +// Copyright 2006 Nemanja Trifunovic + +/* +Permission is hereby granted, free of charge, to any person or organization +obtaining a copy of the software and accompanying documentation covered by +this license (the "Software") to use, reproduce, display, distribute, +execute, and transmit the Software, and to prepare derivative works of the +Software, and to permit third-parties to whom the Software is furnished to +do so, all subject to the following: + +The copyright notices in the Software and this entire statement, including +the above license grant, this restriction and the following disclaimer, +must be included in all copies of the Software, in whole or in part, and +all derivative works of the Software, unless such copies or derivative +works are solely in the form of machine-executable object code generated by +a source language processor. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT +SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE +FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE, +ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +DEALINGS IN THE SOFTWARE. +*/ + + +#ifndef UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 +#define UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731 + +#include "core.h" + +namespace utf8 +{ + namespace unchecked + { + template + octet_iterator append(uint32_t cp, octet_iterator result) + { + if (cp < 0x80) // one octet + *(result++) = static_cast(cp); + else if (cp < 0x800) { // two octets + *(result++) = static_cast((cp >> 6) | 0xc0); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else if (cp < 0x10000) { // three octets + *(result++) = static_cast((cp >> 12) | 0xe0); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + else { // four octets + *(result++) = static_cast((cp >> 18) | 0xf0); + *(result++) = static_cast(((cp >> 12) & 0x3f)| 0x80); + *(result++) = static_cast(((cp >> 6) & 0x3f) | 0x80); + *(result++) = static_cast((cp & 0x3f) | 0x80); + } + return result; + } + + template + uint32_t next(octet_iterator& it) + { + uint32_t cp = utf8::internal::mask8(*it); + typename std::iterator_traits::difference_type length = utf8::internal::sequence_length(it); + switch (length) { + case 1: + break; + case 2: + it++; + cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f); + break; + case 3: + ++it; + cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff); + ++it; + cp += (*it) & 0x3f; + break; + case 4: + ++it; + cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff); + ++it; + cp += (utf8::internal::mask8(*it) << 6) & 0xfff; + ++it; + cp += (*it) & 0x3f; + break; + } + ++it; + return cp; + } + + template + uint32_t peek_next(octet_iterator it) + { + return utf8::unchecked::next(it); + } + + template + uint32_t prior(octet_iterator& it) + { + while (utf8::internal::is_trail(*(--it))) ; + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + + // Deprecated in versions that include prior, but only for the sake of consistency (see utf8::previous) + template + inline uint32_t previous(octet_iterator& it) + { + return utf8::unchecked::prior(it); + } + + template + void advance (octet_iterator& it, distance_type n) + { + for (distance_type i = 0; i < n; ++i) + utf8::unchecked::next(it); + } + + template + typename std::iterator_traits::difference_type + distance (octet_iterator first, octet_iterator last) + { + typename std::iterator_traits::difference_type dist; + for (dist = 0; first < last; ++dist) + utf8::unchecked::next(first); + return dist; + } + + template + octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result) + { + while (start != end) { + uint32_t cp = utf8::internal::mask16(*start++); + // Take care of surrogate pairs first + if (utf8::internal::is_lead_surrogate(cp)) { + uint32_t trail_surrogate = utf8::internal::mask16(*start++); + cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET; + } + result = utf8::unchecked::append(cp, result); + } + return result; + } + + template + u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result) + { + while (start < end) { + uint32_t cp = utf8::unchecked::next(start); + if (cp > 0xffff) { //make a surrogate pair + *result++ = static_cast((cp >> 10) + internal::LEAD_OFFSET); + *result++ = static_cast((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN); + } + else + *result++ = static_cast(cp); + } + return result; + } + + template + octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result) + { + while (start != end) + result = utf8::unchecked::append(*(start++), result); + + return result; + } + + template + u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result) + { + while (start < end) + (*result++) = utf8::unchecked::next(start); + + return result; + } + + // The iterator class + template + class iterator : public std::iterator { + octet_iterator it; + public: + iterator () {} + explicit iterator (const octet_iterator& octet_it): it(octet_it) {} + // the default "big three" are OK + octet_iterator base () const { return it; } + uint32_t operator * () const + { + octet_iterator temp = it; + return utf8::unchecked::next(temp); + } + bool operator == (const iterator& rhs) const + { + return (it == rhs.it); + } + bool operator != (const iterator& rhs) const + { + return !(operator == (rhs)); + } + iterator& operator ++ () + { + ::std::advance(it, utf8::internal::sequence_length(it)); + return *this; + } + iterator operator ++ (int) + { + iterator temp = *this; + ::std::advance(it, utf8::internal::sequence_length(it)); + return temp; + } + iterator& operator -- () + { + utf8::unchecked::prior(it); + return *this; + } + iterator operator -- (int) + { + iterator temp = *this; + utf8::unchecked::prior(it); + return temp; + } + }; // class iterator + + } // namespace utf8::unchecked +} // namespace utf8 + + +#endif // header guard + diff --git a/runtime/src/main/kotlin/kotlin/String.kt b/runtime/src/main/kotlin/kotlin/String.kt index 0b1d9e953f3..4a10582eac0 100644 --- a/runtime/src/main/kotlin/kotlin/String.kt +++ b/runtime/src/main/kotlin/kotlin/String.kt @@ -1,10 +1,6 @@ package kotlin -// TODO: enable, once global variables implemented. -// @SymbolName("theEmptyString") -// val theEmptyString = "" - @ExportTypeInfo("theStringTypeInfo") public final class String : Comparable, CharSequence { @SymbolName("Kotlin_String_hashCode") @@ -21,7 +17,6 @@ public final class String : Comparable, CharSequence { public override val length: Int get() = getStringLength() - // Can be O(N). @SymbolName("Kotlin_String_get") external override public fun get(index: Int): Char