axmol/cocos/base/ccUTF8.cpp

503 lines
14 KiB
C++
Raw Normal View History

/****************************************************************************
Copyright (c) 2014 cocos2d-x.org
Copyright (c) 2014-2016 Chukong Technologies Inc.
Copyright (c) 2017-2018 Xiamen Yaji Software Co., Ltd.
http://www.cocos2d-x.org
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
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 AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
****************************************************************************/
2014-08-29 15:39:52 +08:00
#include "base/ccUTF8.h"
Squashed commit of the following: commit a794d107ad85667e3d754f0b6251fc864dfbf288 Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 14:33:49 2014 -0700 Yeah... everything compiles on win32 and wp8 commit 4740be6e4a0d16f742c27996e7ab2c100adc76af Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 13:58:38 2014 -0700 CCIME moved to base and compiles on Android commit ff3e1bf1eb27a01019f4e1b56d1aebbe2d385f72 Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 13:02:57 2014 -0700 compiles Ok for Windows Phone 8 commit 8160a4eb2ecdc61b5bd1cf56b90d2da6f11e3ebd Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 12:25:31 2014 -0700 fixes for Windows Phone 8 commit 418197649efc93032aee0adc205e502101cdb53d Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 11:15:13 2014 -0700 Compiles on Win32 commit 08813ed7cf8ac1079ffadeb1ce78ea9e833e1a33 Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 10:08:31 2014 -0700 Compiles on linux! commit 118896521e5b335a5257090b6863f1fb2a2002fe Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 09:30:42 2014 -0700 moves cocos/2d/platform -> cocos/platform commit 4fe9319d7717b0c1bccb2db0156eeb86255a89e0 Merge: bd68ec2 511295e Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Fri May 16 08:24:41 2014 -0700 Merge remote-tracking branch 'cocos2d/v3' into files commit bd68ec2f0e3a826d8b2f4b60564ba65ce766bc56 Author: Ricardo Quesada <ricardoquesada@gmail.com> Date: Thu May 15 19:36:23 2014 -0700 files in the correct directory
2014-05-17 05:36:00 +08:00
#include "platform/CCCommon.h"
2014-04-27 01:35:57 +08:00
#include "base/CCConsole.h"
#include "ConvertUTF.h"
#include <limits>
NS_CC_BEGIN
namespace StringUtils {
/*--- This a C++ universal sprintf in the future.
** @pitfall: The behavior of vsnprintf between VS2013 and VS2015/2017 is different
** VS2013 or Unix-Like System will return -1 when buffer not enough, but VS2015/2017 will return the actural needed length for buffer at this station
** The _vsnprintf behavior is compatible API which always return -1 when buffer isn't enough at VS2013/2015/2017
** Yes, The vsnprintf is more efficient implemented by MSVC 19.0 or later, AND it's also standard-compliant, see reference: http://www.cplusplus.com/reference/cstdio/vsnprintf/
*/
std::string format(const char* format, ...)
{
#define CC_VSNPRINTF_BUFFER_LENGTH 512
va_list args;
2020-05-06 16:28:34 +08:00
std::string buf(CC_VSNPRINTF_BUFFER_LENGTH, '\0');
va_start(args, format);
2020-05-06 16:28:34 +08:00
int nret = vsnprintf(&buf.front(), buf.length() + 1, format, args);
va_end(args);
2020-05-06 16:28:34 +08:00
if (nret >= 0)
{
if ((unsigned int)nret < buf.length())
{
buf.resize(nret);
}
2020-05-06 16:28:34 +08:00
else if ((unsigned int)nret > buf.length())
{ // handle return required length when buffer insufficient
buf.resize(nret);
va_start(args, format);
2020-05-06 16:28:34 +08:00
nret = vsnprintf(&buf.front(), buf.length() + 1, format, args);
va_end(args);
}
// else equals, do nothing.
}
2020-05-06 16:28:34 +08:00
else
{ // handle return -1 when buffer insufficient
/*
vs2013/older & glibc <= 2.0.6, they would return -1 when the output was truncated.
see: http://man7.org/linux/man-pages/man3/vsnprintf.3.html
*/
#if (defined(__linux__) && ((__GLIBC__ < 2) || ((__GLIBC__ == 2) && (__GLIBC_MINOR__ < 1)))) || \
(defined(_MSC_VER) && _MSC_VER < 1900)
enum : size_t
{
enlarge_limits = (1 << 20), // limits the buffer cost memory less than 2MB
};
do
{
buf.resize(buf.length() << 1);
va_start(args, format);
2020-05-06 16:28:34 +08:00
nret = vsnprintf(&buf.front(), buf.length() + 1, format, args);
va_end(args);
2020-05-06 16:28:34 +08:00
} while (nret < 0 && buf.size() <= enlarge_limits);
if (nret > 0)
buf.resize(nret);
else
buf = "strfmt: an error is encountered!";
#else
/* other standard implementation
see: http://www.cplusplus.com/reference/cstdio/vsnprintf/
*/
buf = "strfmt: an error is encountered!";
#endif
}
2020-05-06 16:28:34 +08:00
return buf;
}
/*
* @str: the string to search through.
* @c: the character to not look for.
*
* Return value: the index of the last character that is not c.
* */
unsigned int getIndexOfLastNotChar16(const std::vector<char16_t>& str, char16_t c)
{
2013-12-06 16:32:06 +08:00
int len = static_cast<int>(str.size());
int i = len - 1;
for (; i >= 0; --i)
if (str[i] != c) return i;
return i;
}
/*
* @str: the string to trim
* @index: the index to start trimming from.
*
* Trims str st str=[0, index) after the operation.
*
* Return value: the trimmed string.
* */
static void trimUTF16VectorFromIndex(std::vector<char16_t>& str, int index)
{
int size = static_cast<int>(str.size());
if (index >= size || index < 0)
return;
str.erase(str.begin() + index, str.begin() + size);
}
/*
* @str: the string to trim
* @index: the index to start trimming from.
*
* Trims str st str=[0, index) after the operation.
*
* Return value: the trimmed string.
* */
static void trimUTF32VectorFromIndex(std::vector<char32_t>& str, int index)
{
int size = static_cast<int>(str.size());
if (index >= size || index < 0)
return;
str.erase(str.begin() + index, str.begin() + size);
}
/*
* @ch is the unicode character whitespace?
*
* Reference: http://en.wikipedia.org/wiki/Whitespace_character#Unicode
*
* Return value: weather the character is a whitespace character.
* */
bool isUnicodeSpace(char32_t ch)
{
return (ch >= 0x0009 && ch <= 0x000D) || ch == 0x0020 || ch == 0x0085 || ch == 0x00A0 || ch == 0x1680
|| (ch >= 0x2000 && ch <= 0x200A) || ch == 0x2028 || ch == 0x2029 || ch == 0x202F
|| ch == 0x205F || ch == 0x3000;
}
bool isCJKUnicode(char32_t ch)
2014-02-08 14:15:42 +08:00
{
return (ch >= 0x4E00 && ch <= 0x9FBF) // CJK Unified Ideographs
|| (ch >= 0x2E80 && ch <= 0x2FDF) // CJK Radicals Supplement & Kangxi Radicals
|| (ch >= 0x2FF0 && ch <= 0x30FF) // Ideographic Description Characters, CJK Symbols and Punctuation & Japanese
|| (ch >= 0x3100 && ch <= 0x31BF) // Korean
|| (ch >= 0xAC00 && ch <= 0xD7AF) // Hangul Syllables
|| (ch >= 0xF900 && ch <= 0xFAFF) // CJK Compatibility Ideographs
|| (ch >= 0xFE30 && ch <= 0xFE4F) // CJK Compatibility Forms
|| (ch >= 0x31C0 && ch <= 0x4DFF) // Other extensions
|| (ch >= 0x1f004 && ch <= 0x1f682);// Emoji
2014-02-08 14:15:42 +08:00
}
bool isUnicodeNonBreaking(char32_t ch)
{
return ch == 0x00A0 // Non-Breaking Space
|| ch == 0x202F // Narrow Non-Breaking Space
|| ch == 0x2007 // Figure Space
|| ch == 0x2060; // Word Joiner
}
void trimUTF16Vector(std::vector<char16_t>& str)
{
int len = static_cast<int>(str.size());
if ( len <= 0 )
return;
int last_index = len - 1;
// Only start trimming if the last character is whitespace..
if (isUnicodeSpace(str[last_index]))
{
for (int i = last_index - 1; i >= 0; --i)
{
if (isUnicodeSpace(str[i]))
last_index = i;
else
break;
}
trimUTF16VectorFromIndex(str, last_index);
}
}
void trimUTF32Vector(std::vector<char32_t>& str)
{
int len = static_cast<int>(str.size());
if ( len <= 0 )
return;
int last_index = len - 1;
// Only start trimming if the last character is whitespace..
if (isUnicodeSpace(str[last_index]))
{
for (int i = last_index - 1; i >= 0; --i)
{
if (isUnicodeSpace(str[i]))
last_index = i;
else
break;
}
trimUTF32VectorFromIndex(str, last_index);
}
}
template <typename T>
struct ConvertTrait {
typedef T ArgType;
};
template <>
struct ConvertTrait<char> {
typedef UTF8 ArgType;
};
template <>
struct ConvertTrait<char16_t> {
typedef UTF16 ArgType;
};
template <>
struct ConvertTrait<char32_t> {
typedef UTF32 ArgType;
};
template <typename From, typename To, typename FromTrait = ConvertTrait<From>, typename ToTrait = ConvertTrait<To>>
bool utfConvert(
const std::basic_string<From>& from, std::basic_string<To>& to,
ConversionResult(*cvtfunc)(const typename FromTrait::ArgType**, const typename FromTrait::ArgType*,
typename ToTrait::ArgType**, typename ToTrait::ArgType*,
ConversionFlags)
)
{
static_assert(sizeof(From) == sizeof(typename FromTrait::ArgType), "Error size mismatched");
static_assert(sizeof(To) == sizeof(typename ToTrait::ArgType), "Error size mismatched");
if (from.empty())
{
to.clear();
return true;
}
// See: http://unicode.org/faq/utf_bom.html#gen6
static const int most_bytes_per_character = 4;
const size_t maxNumberOfChars = from.length(); // all UTFs at most one element represents one character.
const size_t numberOfOut = maxNumberOfChars * most_bytes_per_character / sizeof(To);
std::basic_string<To> working(numberOfOut, 0);
auto inbeg = reinterpret_cast<const typename FromTrait::ArgType*>(&from[0]);
auto inend = inbeg + from.length();
auto outbeg = reinterpret_cast<typename ToTrait::ArgType*>(&working[0]);
auto outend = outbeg + working.length();
auto r = cvtfunc(&inbeg, inend, &outbeg, outend, strictConversion);
if (r != conversionOK)
return false;
working.resize(reinterpret_cast<To*>(outbeg) - &working[0]);
to = std::move(working);
return true;
};
bool UTF8ToUTF16(const std::string& utf8, std::u16string& outUtf16)
{
return utfConvert(utf8, outUtf16, ConvertUTF8toUTF16);
}
bool UTF8ToUTF32(const std::string& utf8, std::u32string& outUtf32)
{
return utfConvert(utf8, outUtf32, ConvertUTF8toUTF32);
}
bool UTF16ToUTF8(const std::u16string& utf16, std::string& outUtf8)
{
return utfConvert(utf16, outUtf8, ConvertUTF16toUTF8);
}
bool UTF16ToUTF32(const std::u16string& utf16, std::u32string& outUtf32)
{
return utfConvert(utf16, outUtf32, ConvertUTF16toUTF32);
}
bool UTF32ToUTF8(const std::u32string& utf32, std::string& outUtf8)
{
return utfConvert(utf32, outUtf8, ConvertUTF32toUTF8);
}
bool UTF32ToUTF16(const std::u32string& utf32, std::u16string& outUtf16)
{
return utfConvert(utf32, outUtf16, ConvertUTF32toUTF16);
}
#if (CC_TARGET_PLATFORM == CC_PLATFORM_ANDROID)
std::string getStringUTFCharsJNI(JNIEnv* env, jstring srcjStr, bool* ret)
{
std::string utf8Str;
if(srcjStr != nullptr && env != nullptr)
{
const unsigned short * unicodeChar = ( const unsigned short *)env->GetStringChars(srcjStr, nullptr);
size_t unicodeCharLength = env->GetStringLength(srcjStr);
const std::u16string unicodeStr((const char16_t *)unicodeChar, unicodeCharLength);
bool flag = UTF16ToUTF8(unicodeStr, utf8Str);
if (ret)
{
*ret = flag;
}
if (!flag)
{
utf8Str = "";
}
env->ReleaseStringChars(srcjStr, unicodeChar);
}
else
{
if (ret)
{
*ret = false;
}
utf8Str = "";
}
return utf8Str;
}
2016-05-17 12:17:56 +08:00
jstring newStringUTFJNI(JNIEnv* env, const std::string& utf8Str, bool* ret)
{
std::u16string utf16Str;
bool flag = cocos2d::StringUtils::UTF8ToUTF16(utf8Str, utf16Str);
if (ret)
{
*ret = flag;
}
if(!flag)
{
utf16Str.clear();
}
jstring stringText = env->NewString((const jchar*)utf16Str.data(), utf16Str.length());
return stringText;
}
#endif
std::vector<char16_t> getChar16VectorFromUTF16String(const std::u16string& utf16)
{
return std::vector<char16_t>(utf16.begin(), utf16.end());
}
long getCharacterCountInUTF8String(const std::string& utf8)
{
return getUTF8StringLength((const UTF8*)utf8.c_str());
}
StringUTF8::StringUTF8()
{
}
StringUTF8::StringUTF8(const std::string& newStr)
{
replace(newStr);
}
StringUTF8::~StringUTF8()
{
}
std::size_t StringUTF8::length() const
{
return _str.size();
}
void StringUTF8::replace(const std::string& newStr)
{
_str.clear();
if (!newStr.empty())
{
UTF8* sequenceUtf8 = (UTF8*)newStr.c_str();
int lengthString = getUTF8StringLength(sequenceUtf8);
if (lengthString == 0)
{
CCLOG("Bad utf-8 set string: %s", newStr.c_str());
return;
}
while (*sequenceUtf8)
{
std::size_t lengthChar = getNumBytesForUTF8(*sequenceUtf8);
CharUTF8 charUTF8;
charUTF8._char.append((char*)sequenceUtf8, lengthChar);
sequenceUtf8 += lengthChar;
_str.push_back(charUTF8);
}
}
}
std::string StringUTF8::getAsCharSequence() const
{
return getAsCharSequence(0, std::numeric_limits<std::size_t>::max());
}
std::string StringUTF8::getAsCharSequence(std::size_t pos) const
{
return getAsCharSequence(pos, std::numeric_limits<std::size_t>::max());
}
std::string StringUTF8::getAsCharSequence(std::size_t pos, std::size_t len) const
{
std::string charSequence;
std::size_t maxLen = _str.size() - pos;
if (len > maxLen)
{
len = maxLen;
}
std::size_t endPos = len + pos;
while (pos < endPos)
{
charSequence.append(_str[pos++]._char);
}
return charSequence;
}
bool StringUTF8::deleteChar(std::size_t pos)
{
if (pos < _str.size())
{
_str.erase(_str.begin() + pos);
return true;
}
else
{
return false;
}
}
bool StringUTF8::insert(std::size_t pos, const std::string& insertStr)
{
StringUTF8 utf8(insertStr);
return insert(pos, utf8);
}
bool StringUTF8::insert(std::size_t pos, const StringUTF8& insertStr)
{
if (pos <= _str.size())
{
_str.insert(_str.begin() + pos, insertStr._str.begin(), insertStr._str.end());
return true;
}
else
{
return false;
}
}
} //namespace StringUtils {
NS_CC_END