hyzboy/CMCore
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
CMCore/inc/hgl/Endian.h

314 lines
14 KiB
C
Raw Permalink Normal View History

first commit
2019-08-19 19:19:58 +08:00
#ifndef HGL_ENDIAN_INCLUDE
#define HGL_ENDIAN_INCLUDE
#include<hgl/platform/Platform.h> // 平台定义
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
#include<hgl/TypeFunc.h>
first commit
2019-08-19 19:19:58 +08:00
namespace hgl
{
namespace endian
{
/**
* Windows代码页枚举
* 全部Windows所支持代码页请参见 http://msdn.microsoft.com/en-us/library/dd317756
add iso-8859-1 to 15 codepages.
2021-01-18 15:30:39 +08:00
* https://docs.microsoft.com/en-us/windows/win32/intl/code-page-identifiers
first commit
2019-08-19 19:19:58 +08:00
*/
added prefix or postfix
2023-06-06 15:07:25 +08:00
enum class CharCodePage:uint16 ///代码页枚举
first commit
2019-08-19 19:19:58 +08:00
{
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
NONE=0, ///<起始定义，无意义
first commit
2019-08-19 19:19:58 +08:00
added get_utf16_length/to_utf16 functions.
2022-06-14 15:55:42 +08:00
IBM437 =437, ///<OEM United States (dos)
first commit
2019-08-19 19:19:58 +08:00
//中文
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
GBK =936, ///<中国GBK标准中文
Big5 =950, ///<中国台湾Big5标准繁体中文
GB2312 =20936, ///<中国GB2312标准简体中文
GB18030 =54936, ///<中国GB18030-2000标准中文
first commit
2019-08-19 19:19:58 +08:00
//日文
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
ShiftJIS =932, ///<日文ShiftJIS
used uint16 instead of CharCodePage at CodePage.h/Endian.h
2022-06-13 18:15:31 +08:00
EUC_JP =20932, ///<日文JIS 0208-1990 and 0212-1990
ISO2022JP =50220, ///<ISO 2022 Japanese with no halfwidth Katakana; Japanese (JIS)
csISO2022JP =50221, ///<ISO 2022 Japanese with halfwidth Katakana; Japanese (JIS-Allow 1 byte Kana)
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
JISX =50222, ///<日文JIS X/ISO 2022
first commit
2019-08-19 19:19:58 +08:00
//韩文
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
Korean =949, ///<韩文
first commit
2019-08-19 19:19:58 +08:00
//苹果编码
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
MacJanpan =10001, ///<日文
MacTraditionalChinese =10002, ///<繁体中文
MacSimplifiedChinese =10008, ///<简体中文
first commit
2019-08-19 19:19:58 +08:00
add iso-8859-1 to 15 codepages.
2021-01-18 15:30:39 +08:00
//ISO
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
ISO_8859_1 =28591, ///<ISO 8859-1 Latin 1; Western European 西欧语系(阿尔巴尼亚语,西班牙加泰罗尼亚语,丹麦语,荷兰语,英语,Faeroese语,芬兰语,法语,德语,加里西亚语,爱尔兰语,冰岛语,意大利语,挪威语,葡萄牙语,瑞士语.)这同时适用于美国英语.
ISO_8859_2 =28592, ///<ISO 8859-2 Central European; Central European (ISO) 斯拉夫/中欧语系(捷克语,德语,匈牙利语,波兰语,罗马尼亚语,克罗地亚语,斯洛伐克语,斯洛文尼亚语)
ISO_8859_3 =28593, ///<ISO 8859-3 Latin 3 世界语,加里西亚语,马耳他语,土耳其语
ISO_8859_4 =28594, ///<ISO 8859-4 Baltic 爱莎尼亚语,拉脱维亚语,立陶宛语
ISO_8859_5 =28595, ///<ISO 8859-5 Cyrillic 斯拉夫语系(保加利亚语,Byelorussian语,马其顿语,俄语,塞尔维亚语,乌克兰语)
ISO_8859_6 =28596, ///<ISO 8859-6 Arabic 阿拉伯语
ISO_8859_7 =28597, ///<ISO 8859-7 Greek 现代希腊语
ISO_8859_8 =28598, ///<ISO 8859-8 Hebrew; Hebrew (ISO-Visual) 希伯来语
ISO_8859_9 =28599, ///<ISO 8859-9 Turkish Latin 5 字符集, (去掉了 Latin 1中不经常使用的一些冰岛语字符而代以土耳其语字符)
ISO_8859_13 =28603, ///<ISO 8859-13 Estonian
ISO_8859_15 =28605, ///<ISO 8859-15 Latin 9 Latin 9 字符集, 是Latin 1字符集的更新版本,去掉一些不常用的字符,增加了对爱莎尼亚语的支持,修正了法语和芬兰语部份,增加了欧元字符)
add iso-8859-1 to 15 codepages.
2021-01-18 15:30:39 +08:00
first commit
2019-08-19 19:19:58 +08:00
//unicode
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
UTF7 =65000, ///<utf-7
UTF8 =65001, ///<utf-8
first commit
2019-08-19 19:19:58 +08:00
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
UTF16LE =1200,
UTF16BE =1201,
UTF32LE =12000,
UTF32BE =12001,
first commit
2019-08-19 19:19:58 +08:00
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
END ///<结束定义，无意义
first commit
2019-08-19 19:19:58 +08:00
};//enum CharCodePage
/**
* 字节序类型枚举
*/
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
enum class ByteOrderMask
first commit
2019-08-19 19:19:58 +08:00
{
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
NONE=0,
UTF8,
UTF16LE,
UTF16BE,
UTF32LE,
UTF32BE,
ENUM_CLASS_RANGE(UTF8,UTF32BE)
first commit
2019-08-19 19:19:58 +08:00
};
use "enum class" instead of "enum" at "CharCodePage".
2021-07-27 11:32:39 +08:00
constexpr uint CHAR_SET_NAME_MAX_LENGTH=32; ///<字符集名称长度
using CharSetName=char[CHAR_SET_NAME_MAX_LENGTH]; ///<字符集名称类型定义
first commit
2019-08-19 19:19:58 +08:00
support "u8char" type
2020-07-07 19:14:42 +08:00
template<int,char> const CharSetName &GetCurCharSet(); ///<取得当前程序编码字符集
first commit
2019-08-19 19:19:58 +08:00
constexpr CharSetName utf8_charset="utf8";
constexpr CharSetName utf16le_charset="utf-16le";
constexpr CharSetName utf16be_charset="utf-16be";
constexpr CharSetName utf32le_charset="utf-32le";
constexpr CharSetName utf32be_charset="utf-32be";
template<> inline const CharSetName &GetCurCharSet<2,HGL_LITTLE_ENDIAN >(){return utf16le_charset;}
template<> inline const CharSetName &GetCurCharSet<2,HGL_BIG_ENDIAN >(){return utf16be_charset;}
template<> inline const CharSetName &GetCurCharSet<4,HGL_LITTLE_ENDIAN >(){return utf32le_charset;}
template<> inline const CharSetName &GetCurCharSet<4,HGL_BIG_ENDIAN >(){return utf32be_charset;}
template<typename T> inline const CharSetName &GetCharSet()
{
return GetCurCharSet<sizeof(T),HGL_ENDIAN>();
}
/**
* 字节序文件头数据结构
*/
struct BOMFileHeader
{
int size; ///<字节序文件头长度
support "u8char" type
2020-07-07 19:14:42 +08:00
uint8 data[4]; ///<字节序数据
first commit
2019-08-19 19:19:58 +08:00
ByteOrderMask bom; ///<字节序枚举
const CharSetName *char_set;///<字符集名称
used uint16 instead of CharCodePage at CodePage.h/Endian.h
2022-06-13 18:15:31 +08:00
uint16 code_page; ///<代码页
first commit
2019-08-19 19:19:58 +08:00
};
/**
* 字节序文件头定义
*/
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
constexpr BOMFileHeader BOMData[size_t(ByteOrderMask::RANGE_SIZE)]=
first commit
2019-08-19 19:19:58 +08:00
{
used uint16 instead of CharCodePage at CodePage.h/Endian.h
2022-06-13 18:15:31 +08:00
{3,{0xEF,0xBB,0xBF} ,ByteOrderMask::UTF8, &utf8_charset ,(uint16)CharCodePage::UTF8 },
{2,{0xFF,0xFE} ,ByteOrderMask::UTF16LE,&utf16le_charset ,(uint16)CharCodePage::UTF16LE },
{2,{0xFE,0xFF} ,ByteOrderMask::UTF16BE,&utf16be_charset ,(uint16)CharCodePage::UTF16BE },
{4,{0xFF,0xFE,0x00,0x00},ByteOrderMask::UTF32LE,&utf32le_charset ,(uint16)CharCodePage::UTF32LE },
{4,{0x00,0x00,0xFE,0xFF},ByteOrderMask::UTF32BE,&utf32be_charset ,(uint16)CharCodePage::UTF32BE }
first commit
2019-08-19 19:19:58 +08:00
};
splited ASyncEvent
2022-07-07 21:32:35 +08:00
inline const BOMFileHeader *GetBOM(const ByteOrderMask &bom)
{
return RangeCheck(bom)?BOMData+uint(bom)-uint(ByteOrderMask::BEGIN_RANGE):nullptr;
}
增加CheckBOM(void *)函数
2020-06-12 13:15:15 +08:00
inline ByteOrderMask CheckBOM(const void *data)
{
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
const BOMFileHeader *bom=BOMData;
for(int i=int(ByteOrderMask::BEGIN_RANGE);i<int(ByteOrderMask::END_RANGE);i++)
增加CheckBOM(void *)函数
2020-06-12 13:15:15 +08:00
{
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
if(memcmp(data,bom->data,bom->size)==0)
增加CheckBOM(void *)函数
2020-06-12 13:15:15 +08:00
return (ByteOrderMask)i;
}
use "enum class" in ByteOrderMask
2020-09-11 18:51:36 +08:00
return ByteOrderMask::NONE;
增加CheckBOM(void *)函数
2020-06-12 13:15:15 +08:00
}
first commit
2019-08-19 19:19:58 +08:00
template<typename T>
inline T EndianSwap(const T value)
{
union
{
T v;
uint8 bytes[sizeof(T)];
}a,b;
a.v=value;
for(uint i=0;i<sizeof(T);i++)
b.bytes[i]=a.bytes[sizeof(T)-1-i];
return b.v;
}
template<> inline int8 EndianSwap< int8>(const int8 value){return value;}
template<> inline uint8 EndianSwap<uint8>(const uint8 value){return value;}
template<> inline uint16 EndianSwap(const uint16 value)
{
return ((value&0xFF)<<8)
|((value&0xFF00)>>8);
}
template<> inline uint32 EndianSwap(const uint32 value)
{
return ((value&0xFF)<<24)
|((value&0xFF00)<<8)
|((value&0xFF0000)>>8)
|((value&0xFF000000)>>24);
}
template<> inline uint64 EndianSwap(const uint64 value)
{
return ((value&0xFF)<<56)
|((value&0xFF00)<<40)
|((value&0xFF0000)<<24)
|((value&0xFF000000)<<8)
|((value&0xFF00000000)>>8)
|((value&0xFF0000000000)>>24)
|((value&0xFF000000000000)>>40)
|((value&0xFF00000000000000)>>56);
}
template<> inline u16char EndianSwap(const u16char value){return EndianSwap(uint16(value));}
template<typename T>
inline void EndianSwap(T *value,const int64 count)
{
for(int64 i=0;i<count;i++)
{
*value=EndianSwap(*value);
++value;
}
}
template<typename T>
inline void EndianSwap(T *dst,const T *src,const int64 count)
{
for(uint i=0;i<count;i++)
{
*dst=EndianSwap(*src);
++dst;
++src;
}
}
template<typename D,typename S>
inline void EndianSwap(D *dst,const S *src,const int64 count)
{
for(uint i=0;i<count;i++)
{
*dst=EndianSwap(D(*src)); //必须在ENDIAN SWAP前转换类型，否则在32转16位时，会将0000转出来。
++dst;
++src;
}
}
#if HGL_ENDIAN == HGL_BIG_ENDIAN
constexpr uint HGL_BOM_UTF16LE =0xfffe;
constexpr uint HGL_BOM_UTF16BE =0xfeff;
constexpr uint HGL_BOM_UTF32LE =0xfffe0000;
constexpr uint HGL_BOM_UTF32BE =0x0000feff;
#define LittleToCurrentEndian EndianSwap
#define BigToCurrentEndian ToBigEndian
template<typename T> T ToBigEndian(T value){return value;}
template<typename T> inline void ToBigEndian(T *value,const int64 count){}
template<typename D,typename S> inline void ToBigEndian(D *dst,const S *src,const int64 count){hgl_cpy(dst,src,count);}
template<typename T> T ToLittleEndian(T value){return EndianSwap(value);}
template<typename T> inline void ToLittleEndian(T *value,const int64 count){EndianSwap<T>(value,count);}
template<typename D,typename S> inline void ToLittleEndian(D *dst,const S *src,const int64 count){EndianSwap<D,S>(dst,src,count);}
#else
constexpr uint HGL_BOM_UTF16LE =0xfeff;
constexpr uint HGL_BOM_UTF16BE =0xfffe;
constexpr uint HGL_BOM_UTF32LE =0x0000feff;
constexpr uint HGL_BOM_UTF32BE =0xfffe0000;
#define LittleToCurrentEndian ToLittleEndian
#define BigToCurrentEndian EndianSwap
template<typename T> T ToBigEndian(T value){return EndianSwap(value);}
template<typename T> inline void ToBigEndian(T *value,const int64 count){EndianSwap<T>(value,count);}
template<typename D,typename S> inline void ToBigEndian(D *dst,const S *src,const int64 count){EndianSwap<D,S>(dst,src,count);}
template<typename T> T ToLittleEndian(T value){return value;}
template<typename T> inline void ToLittleEndian(T *,const int64){}
redefine hgl_cpy/hgl_move .... function name.
2020-09-19 18:31:07 +08:00
template<typename D,typename S> inline void ToLittleEndian(D *dst,const S *src,const int64 count){typeconv(dst,src,count);}
first commit
2019-08-19 19:19:58 +08:00
#endif//HGL_BIG_ENDIAN
template<char> struct UTF16CharConvert;
template<> struct UTF16CharConvert<HGL_LITTLE_ENDIAN>
{
#if HGL_ENDIAN == HGL_BIG_ENDIAN
static void convert(u16char *str,const int length)
{
EndianSwap<u16char>(str,length);
}
#else
static void convert(const u16char *,const int){}
#endif//HGL_ENDIAN == HGL_LITTLE_ENDIAN
static void convert(u16char *out_str,const u16char *in_str,const int length)
{
#if HGL_ENDIAN == HGL_LITTLE_ENDIAN
memcpy(out_str,in_str,length*sizeof(u16char));
#else
EndianSwap<u16char>(in_str,length);
#endif//HGL_ENDIAN == HGL_LITTLE_ENDIAN
}
};//template<> struct UTF16CharConvert<HGL_LITTLE_ENDIAN>
template<> struct UTF16CharConvert<HGL_BIG_ENDIAN>
{
static void convert(u16char *str,const int length)
{
#if HGL_ENDIAN == HGL_LITTLE_ENDIAN
EndianSwap<u16char>(str,length);
#endif//HGL_ENDIAN == HGL_LITTLE_ENDIAN
}
static void convert(u16char *out_str,const u16char *in_str,const int length)
{
#if HGL_ENDIAN == HGL_LITTLE_ENDIAN
memcpy(out_str,in_str,length*sizeof(u16char));
#else
EndianSwap<u16char>(in_str,length);
#endif//HGL_ENDIAN == HGL_LITTLE_ENDIAN
}
};//template<> struct UTF16ToWideChar<HGL_BIG_ENDIAN>
}//namespace endian
using namespace endian;
}//namespace hgl
#endif//HGL_ENDIAN_INCLUDE
Reference in New Issue Copy Permalink