首页 > 代码库 > 自己编写的Windows字符串类 封装字符串格式转换和常用操作 方便使用

自己编写的Windows字符串类 封装字符串格式转换和常用操作 方便使用

最近开发的语音识别的项目经常碰到转码的问题,各种宽窄字节转换,ASNI, UTF8, 宽字节 ……  代码写得冗长,繁琐,维护性太差了。决定自己写一个能直接使用各种编码的字符串类,于是实现了一个,功能不多,但是也够用。由于string 和 wstring 在多线程下共享数据会出问题,这里只用最基础的char 和 wchar_t。  

基于Windows平台,代码如下


/*
**************************************************   
		Title: 自定义字符串类
************************************************** 
		Date:2014/12/14
************************************************** 
		author:刘旭  
**************************************************   
		Blog:http://blog.csdn.net/liuxucoder
**************************************************  
*/ 
#ifndef __YKSTRING
#define __YKSTRING

#include <cstring>
#include <new>
#include <tchar.h>
#include <strsafe.h>
#include <Windows.h>


#define MAX_COUNT  1024000

typedef unsigned char BYTE;  

//十进制转十六进制
inline BYTE Int_to_Hex(const BYTE src)  
{  
	return src > 9 ? (src+55) : (src+48);  
} 
//十六进制转十进制
inline BYTE Hex_to_Int(const BYTE &x)  
{  
	return isdigit(x) ? x-'0' : x-'A'+10;  
}  

/*
*@自定义字符串类
*/
class CYkString
{
public:
	/*
	*@目的:		字符串转码,由 Ansi 转为 wchar_t
	*@参数:		const char* szAnsi		转码前 Ansi 字符串
	*@参数:		wchar_t* szWchar			转码后 wchar_t 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Ansi_to_Wchar(const char* szAnsi,wchar_t*& szWchar)
	{
		bool res = false;
		size_t nLen = MultiByteToWideChar(CP_ACP, NULL, szAnsi, -1, NULL, 0);
		wchar_t* strTemp  = NULL;
		nLen = (nLen+1)*sizeof(wchar_t);
		__try{
			if(NULL == (szWchar = new wchar_t[nLen])){
				__leave;
			}		
			if(NULL == (strTemp = new wchar_t[nLen])){
				__leave;
			}
			ZeroMemory(szWchar, nLen);
			ZeroMemory(strTemp, nLen);
			if(0 == MultiByteToWideChar(CP_ACP, NULL, szAnsi, -1, strTemp, nLen)){
				__leave;
			}
			if(S_OK != StringCchCopyW(szWchar, nLen, strTemp)){
				__leave;
			}
			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}
		}
		return res;
	}

	/*
	*@目的:		字符串转码,由 Utf-8 转为 wchar_t
	*@参数:		const char* szUtf8	转码前 Utf-8 字符串
	*@参数:		wchar_t* szWchar		转码后 wchar_t 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Utf8_to_Wchar(const char* szUtf8,wchar_t*& szWchar)
	{
		bool res = false;
		size_t nLen = MultiByteToWideChar(CP_UTF8, NULL, szUtf8, -1, NULL, 0);
		wchar_t* strTemp =NULL;
		nLen = (nLen+1)*sizeof(wchar_t);
		__try{
			if(NULL == (szWchar = new wchar_t[nLen])){
				__leave;
			}
			if(NULL == (strTemp = new wchar_t[nLen])){
				__leave;
			}
			ZeroMemory(szWchar, nLen);
			ZeroMemory(strTemp, nLen);
			if(0 == MultiByteToWideChar(CP_UTF8, NULL, szUtf8, -1, strTemp, nLen)){
				__leave;
			}
			if(S_OK != StringCchCopyW(szWchar, nLen, strTemp)){
				__leave;
			}
			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}
		}
		return res;
	}

	/*
	*@目的:		字符串转码,由 wchar_t 转为 Ansi
	*@参数:		const wchar_t* szWchar	转码前 wchar_t 字符串
	*@参数:		char* szAnsi					转码后 Ansi 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Wchar_to_Ansi(const wchar_t* szWchar,char*& szAnsi)
	{
		bool res = false;
		size_t nLen = 0;
		char *strTemp = NULL;
		StringCchLengthW(szWchar, MAX_COUNT, &nLen);
		nLen = (nLen+1)*sizeof(wchar_t);
		__try{
			if(NULL == (szAnsi = new char[nLen])){
				__leave;
			}
			if(NULL == (strTemp = new char[nLen])){
				__leave;
			}
			ZeroMemory(szAnsi, nLen);
			ZeroMemory(strTemp, nLen);
			if( 0 == WideCharToMultiByte(CP_ACP, 0, szWchar, -1, strTemp, nLen, NULL, NULL)){
				__leave;
			}
			if(S_OK  != StringCchCopyA(szAnsi, nLen, strTemp)){
				__leave;
			}
			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}
		}

		return res;
	}

	/*
	*@目的:		字符串转码,由 wchar_t 转为 Utf-8
	*@参数:		const wchar_t* szWchar	转码前 wchar_t 字符串
	*@参数:		char* szUtf8						转码后 Utf-8 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Wchar_to_Utf8(const wchar_t* szWchar,char*& szUtf8)
	{
		bool res = false;
		size_t nLen = 0;
		char *strTemp = NULL;
		StringCchLengthW(szWchar, MAX_COUNT, &nLen);
		nLen = (nLen+1)*2*sizeof(wchar_t);
		__try{
			if(NULL == (szUtf8 = new char[nLen])){
				__leave;
			}
			if(NULL == (strTemp = new char[nLen])){
				__leave;
			}
			ZeroMemory(szUtf8, nLen);
			ZeroMemory(strTemp, nLen);
			if( 0 == WideCharToMultiByte(CP_UTF8, 0, szWchar, -1, strTemp, nLen, NULL, NULL)){
				__leave;
			}
			if(S_OK  != StringCchCopyA(szUtf8, nLen, strTemp)){
				__leave;
			}
			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}
		}

		return  res;
	}

	/*
	*@目的:		字符串转码,由 Ansi 转为 Utf-8
	*@参数:		const char * szAnsi  转码前 Ansi 字符串
	*@参数:		char* szUtf8				 转码后 Utf-8 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Ansi_to_Utf8(const char* szAnsi,char*& szUtf8)
	{
		bool res = false;
		size_t nLen = 0;
		wchar_t* strTemp  = NULL;
		StringCchLengthA(szAnsi, MAX_COUNT, &nLen);
		__try{
			if(NULL == (strTemp = new wchar_t[nLen+1])){
				__leave;
			}
			if(false == Ansi_to_Wchar(szAnsi, strTemp)){
				__leave;
			}
			if(false == Wchar_to_Utf8(strTemp, szUtf8)){
				__leave;
			}

			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}	
		}
		return res;
	}


	/*
	*@目的:		字符串转码,由 Utf-8 转为 Ansi
	*@参数:		const char * szUtf8  转码前 Utf-8 字符串
	*@参数:		char* szAnsi			转码后 Ansi 字符串
	*@返回值: bool  成功返回true,失败返回false
	*/
	bool Utf8_to_Ansi(const char* szUtf8,char*& szAnsi)
	{
		bool res = false;
		size_t nLen = 0;
		wchar_t* strTemp  = NULL;
		StringCchLengthA(szUtf8, MAX_COUNT, &nLen);
		__try{

			if(NULL == (strTemp = new wchar_t[nLen+1])){
				__leave;
			}
			if(false == (Utf8_to_Wchar(szUtf8, strTemp))){
				__leave;
			}
			if(false ==Wchar_to_Ansi(strTemp, szAnsi)){
				__leave;
			}

			res = true;
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
			}	
		}
		return res;

	}

public:

	CYkString(size_t len = 256){
		Last_Style = Style_Char;
		IsAnsi = true;
		m_char = NULL;
		m_wchar = NULL;
		try{
			m_char = new char[len];
			m_wchar = new wchar_t[len];
			ZeroMemory(m_char, len);
			ZeroMemory(m_wchar, len);
		}
		catch(std::bad_alloc){
			if(m_char){
				delete []m_char;
			}
			if(m_wchar){
				delete []m_wchar;
			}
			throw;	//此处抛出异常,停止构造函数,释放资源
		}
	};

	~CYkString(void){
		if(NULL != m_char){
			delete []m_char;
			m_char = NULL;
		}
		if(NULL != m_wchar){
			delete []m_wchar;
			m_wchar = NULL;
		}
	};

	/*
	*@目的: 刷新字符串,是每种字符串储存内容一致
	*@参数: 无
	*@返回值: bool 成功返回 true  失败返回 false
	*/
	bool FlushString(){
		if(Style_Char == Last_Style){
			if(false == IsAnsi){
				return Utf8_to_Wchar(m_char, m_wchar);
			}else{
				return Ansi_to_Wchar(m_char, m_wchar);
			}
		}
		else if(Style_WChar == Last_Style){
			if(false == IsAnsi){
				return 	Wchar_to_Utf8(m_wchar, m_char);
			}
			else{
				return 	Wchar_to_Ansi(m_wchar, m_char);
			}
		}
	}

	/*
	*@目的:		使用字符串的char*格式
	*@参数:		无
	*@返回值: char* char*字符串地址
	*/
	char*  asChar_str(){
		FlushString();
		Last_Style = Style_Char;
		return m_char;
	}

	/*
	*@目的:		使用字符串的wchar_t*格式
	*@参数:		无
	*@返回值: wchar_t* wchar_t*字符串地址
	*/
	wchar_t* asWchar_str(){
		FlushString();
		Last_Style = Style_WChar;
		return m_wchar;
	}

	/*
	*@目的:		返回字符串的长度
	*@参数:		无
	*@返回值:	size_t  字符串的长度
	*/
	size_t GetLength(){
		FlushString();
		size_t nLen = 0;
		StringCchLengthW(m_wchar, MAX_COUNT, &nLen);
		return nLen;
	}

	/*
	*@目的:		判断字符串是否为空
	*@参数:		无
	*@返回值:	bool 空则返回true,非空则返回false
	*/
	bool IsEmpty(){
		FlushString();
		return GetLength() ? false : true;
	}

	/*
	*@目的:		将字符串转换为Ansi编码形式
	*@参数:		无
	*@返回值: char* 转码后的字符串
	*/
	char* ToAnsi(){
		FlushString();
		if(false == IsAnsi){
			IsAnsi = true;
			if(false == Utf8_to_Ansi(m_char, m_char)){
				m_char = NULL;
			}
		}
		return m_char;
	};

	/*
	*@目的:		将字符串转换为Utf-8编码形式
	*@参数:		无
	*@返回值:	char* 转码后的字符串 
	*/
	char* ToUtf_8(){
		FlushString();
		if(true == IsAnsi){
			IsAnsi = false;
			if(false == Ansi_to_Utf8(m_char, m_char)){
				m_char = NULL;
			}
		}
		return m_char;
	};

	/*
	*@目的:		将Utf-8格式字符串用Urlcode编码
	*@参数:		无
	*@返回值: char* 编码后的字符串
	*/
	char* UrlEncode_Utf8()
	{
		bool IsAnsiTemp = IsAnsi;
		ToUtf_8();
		char* result = NULL;  
		size_t nLen = 0;
		StringCchLengthA(m_char, MAX_COUNT, &nLen);
		size_t pos = 0;    //结果字符串的长度标记  
		try{
			result = new char[(nLen+1)*sizeof(wchar_t)];
			ZeroMemory(result, sizeof(result));
		}catch(std::bad_alloc){
			delete []result;
			result = NULL;
			return result;
		}

		for(size_t i = 0; i < nLen; i++){  
			if(isalnum((BYTE)m_char[i]) || /*判断是否为字母或者数字  必须进行类型转换*/  
				':' == m_char[i]	|| 
				'/' == m_char[i]	|| 
				'_' == m_char[i]   ||     
				'.' == m_char[i]	||     
				'~' == m_char[i]	||  
				'?' == m_char[i]	||   
				'&' == m_char[i]	||   
				'=' == m_char[i]   ){ 
					result[pos++] = m_char[i];   //保持不变  
			}  
			else if(' ' == m_char[i]){//如果是空格  
				result[pos++] = '+';  
			}  
			else {//如果是其他字符  
				BYTE temp = Int_to_Hex((BYTE) m_char[i]);  
				result[pos++] = '%';  
				result[pos++] = Int_to_Hex((BYTE)m_char[i] >> 4);    
				result[pos++] = Int_to_Hex((BYTE)m_char[i] % 16);    
			}  
		}

		IsAnsi = IsAnsiTemp ? ToAnsi() : ToUtf_8();	//恢复原来的编码

		result[pos++] = '\0';
		return result;  
	}

	/*
	*@目的:		将Ansi格式字符串用Urlcode编码
	*@参数:		无
	*@返回值: char* 编码后的字符串
	*/
	char* UrlEncode_Ansi()
	{
		bool IsAnsiTemp = IsAnsi;
		ToAnsi();

		char* result = NULL;  
		size_t nLen = 0;
		StringCchLengthA(m_char, MAX_COUNT, &nLen);
		size_t pos = 0;    //结果字符串的长度标记  
		try{
			result = new char[(nLen+1)*sizeof(wchar_t)];
			ZeroMemory(result, sizeof(result));
		}catch(std::bad_alloc){
			delete []result;
			result = NULL;
			return result;
		}

		for(size_t i = 0; i < nLen; i++){  
			if(isalnum((BYTE)m_char[i]) || /*判断是否为字母或者数字  必须进行类型转换*/  
				':' == m_char[i]	|| 
				'/' == m_char[i]	|| 
				'_' == m_char[i]   ||     
				'.' == m_char[i]	||     
				'~' == m_char[i]	||  
				'?' == m_char[i]	||   
				'&' == m_char[i]	||   
				'=' == m_char[i]   ){ 
					result[pos++] = m_char[i];   //保持不变  
			}  
			else if(' ' == m_char[i]){//如果是空格  
				result[pos++] = '+';  
			}  
			else {//如果是其他字符  
				BYTE temp = Int_to_Hex((BYTE) m_char[i]);  
				result[pos++] = '%';  
				result[pos++] = Int_to_Hex((BYTE)m_char[i] >> 4);    
				result[pos++] = Int_to_Hex((BYTE)m_char[i] % 16);    
			}  
		}

		IsAnsi = IsAnsiTemp ? ToAnsi() : ToUtf_8();	//恢复原来的编码

		result[pos++] = '\0';
		return result;  
	}

	/*
	*@目的:		将字符串用Urlcode解码
	*@参数:		无
	*@返回值: char* 解码后的字符串(格式取决于字符串本身的格式)
	*/
	char* UrlDecode()
	{
		FlushString();
		char* result = NULL;  
		size_t nLen = 0;
		StringCchLengthA(m_char, MAX_COUNT, &nLen);
		size_t pos = 0;    //结果字符串的长度标记  
		try{
			result = new char[(nLen+1)*sizeof(wchar_t)];
			ZeroMemory(result, sizeof(result));
		}catch(std::bad_alloc){
			delete []result;
			result = NULL;
			return result;
		}

		for(size_t i = 0; i < nLen; i++){  
			if('%' == m_char[i]){ //判断是否为汉字
				BYTE cha;
				cha = Hex_to_Int(m_char[i+1])<<4;
				cha |= Hex_to_Int(m_char[i+2]);
				result[pos++] = (char)cha;   
				i += 2;
			}  
			else if('+' == m_char[i]){//如果是空格  
				result[pos++] = ' ';  
			}  
			else {//如果是其他字符  
				result[pos++] = m_char[i]; //保持不变  
			}  
		}

		result[pos++] = '\0';
		return result;  
	}

	/*
	*@目的:		将字符串格式化(char*)
	*@参数:		char* Format  格式化字符串
	*@参数:		...  参数列表
	*@返回值: int 格式化参数的个数, -1表示失败
	*/
	int Format(char* Format, ...)
	{
		Last_Style = Style_Char;
		va_list argList = NULL;
		size_t nLen = 0;
		va_start(argList, Format);
		nLen = _vsprintf_p(NULL, 0, Format, argList);
		try{
			m_char = new char[nLen+1];
			_vsprintf_p(m_char, nLen + 1, Format, argList);
			return FlushString();
		}
		catch(std::bad_alloc){
			delete []m_char;
			m_char = NULL;
			return -1;
		}
	}

	/*
	*@目的:		将字符串格式化(wchar_t*)
	*@参数:		wchar_t* Format  格式化字符串
	*@参数:		...  参数列表
	*@返回值: int 格式化参数的个数, -1表示失败
	*/
	int Format(wchar_t* Format, ...)
	{
		Last_Style = Style_WChar;
		va_list argList = NULL;
		size_t nLen = 0;
		va_start(argList, Format);
		nLen = _vswprintf_p(NULL, 0, Format, argList);
		try{
			m_wchar = new wchar_t[nLen+1];
			_vswprintf_p(m_wchar, nLen + 1, Format, argList);
			return FlushString();
		}
		catch(std::bad_alloc){
			delete []m_wchar;
			m_wchar = NULL;
			return -1;
		}
	}

public:
	//==运算符重载
	bool operator == (const char* obj){
		return 0 == strncmp(m_char, obj, MAX_COUNT) ? 			true : false;
	}

	bool operator == (const wchar_t* obj){
		return CSTR_EQUAL == CompareStringOrdinal(m_wchar, -1, obj, -1, false) ? 			true : false;
	}

	bool operator == (CYkString &obj){
		return CSTR_EQUAL == CompareStringOrdinal(m_wchar, -1, obj.asWchar_str(), -1, false) ? 			true : false;
	}

	//=运算符重载
	CYkString& operator = (const char* obj)
	{
		Last_Style = Style_Char;
		if(*this == obj){
			return *this;
		}
		size_t nLen1 = GetLength()*sizeof(wchar_t);
		size_t nLen2 = 0;
		StringCchLengthA(obj, MAX_COUNT, &nLen2);
		nLen2 += 1;

		if(nLen1 >= nLen2){
			StringCchCopyA(m_char, nLen1, obj);
		}
		else{
			char *strTemp = NULL;
			__try{
				if(NULL == (strTemp = new char[nLen2])){
					__leave;
				}
				if(S_OK != StringCchCopyA(strTemp, nLen2, obj)){
					__leave;
				}
				delete m_char;
				m_char = NULL;
				if(NULL == (m_char = new char[nLen2])){
					__leave;
				}
				if(S_OK != StringCchCopyA(m_char, nLen2, strTemp)){
					__leave;
				}
			}
			__finally{
				if(NULL != strTemp){
					delete []strTemp;
					strTemp = NULL;

					FlushString();
				}
			}
		}
		return *this;
	}

	CYkString& operator = (const wchar_t* obj)
	{
		Last_Style = Style_WChar;
		if(*this == obj){
			return *this;
		}
		size_t nLen1 = GetLength()*sizeof(wchar_t);
		size_t nLen2 = 0;
		StringCchLengthW(obj, MAX_COUNT, &nLen2);
		nLen2 *= sizeof(wchar_t);
		if(nLen1 >= nLen2){
			StringCchCopyW(m_wchar, nLen1, obj);
		}
		else{
			wchar_t *strTemp = NULL;
			__try{
				if(NULL == (strTemp = new wchar_t[nLen2])){
					__leave;
				}
				if(S_OK != StringCchCopyW(strTemp, nLen2, obj)){
					__leave;
				}
				delete []m_wchar;
				m_wchar = NULL;
				if(NULL == (m_wchar = new wchar_t[nLen2])){
					__leave;
				}
				if(S_OK != StringCchCopyW(m_wchar, nLen2, strTemp)){
					__leave;
				}
			}
			__finally{
				if(NULL != strTemp){
					delete []strTemp;
					strTemp = NULL;

					FlushString();
				}
			}
		}
		return *this;
	}

	CYkString& operator = (CYkString &obj)
	{
		return *this = obj.asWchar_str();
	}

	//+=运算符重载
	CYkString& operator += (const char* obj) 
	{
		Last_Style = Style_Char;
		size_t nLen1 = GetLength()*sizeof(wchar_t) + 1;
		size_t nLen2 = 0;
		StringCchLengthA(obj, MAX_COUNT, &nLen2);
		nLen2 += 1;
		char *strTemp = NULL;
		__try{
			if(NULL == (strTemp = new char[nLen1])){
				__leave;
			}
			if(S_OK != StringCchCopyA(strTemp, nLen1, m_char)){
				__leave;
			}
			delete []m_wchar;
			m_wchar = NULL;
			if(NULL == (m_char = new char[nLen1+nLen2])){
				__leave;
			}
			if(S_OK != StringCchCopyA(m_char, nLen1, strTemp)){
				__leave;
			}
			if(S_OK != StringCchCatA(m_char, nLen1+nLen2, obj)){
				__leave;
			}
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
				FlushString();
			}
		}

		return *this;
	}

	CYkString& operator += (const wchar_t* obj) 
	{
		Last_Style = Style_WChar;
		size_t nLen1 = GetLength()*sizeof(wchar_t);
		size_t nLen2 = 0;
		StringCchLengthW(obj, MAX_COUNT, &nLen2);
		nLen2 *= sizeof(wchar_t);
		wchar_t *strTemp = NULL;
		__try{
			if(NULL == (strTemp = new wchar_t[nLen1])){
				__leave;
			}
			if(S_OK != StringCchCopyW(strTemp, nLen1, m_wchar)){
				__leave;
			}
			delete []m_wchar;
			m_wchar = NULL;
			if(NULL == (m_wchar = new wchar_t[nLen1+nLen2])){
				__leave;
			}
			if(S_OK != StringCchCopyW(m_wchar, nLen1, strTemp)){
				__leave;
			}
			if(S_OK != StringCchCatW(m_wchar, nLen1+nLen2, obj)){
				__leave;
			}
		}
		__finally{
			if(NULL != strTemp){
				delete []strTemp;
				strTemp = NULL;
				FlushString();
			}
		}

		return *this;
	}

	CYkString& operator += (CYkString &obj) 
	{
		return *this += obj.asWchar_str();
	}

protected:
	char* m_char;			//内部字符串 char*   窄字节类型
	wchar_t* m_wchar;	//内部字符串 wchar* 宽字节类型

private:
	enum LASTCHANGE{
		Style_Char		= 1,
		Style_WChar		= 2
	};//最后改变的数据类型

	bool IsAnsi;		//字符串是否是Ansi编码格式
	int Last_Style;	//最后改变的数据类型标记
};


#endif // !__YKSTRING


自己编写的Windows字符串类 封装字符串格式转换和常用操作 方便使用