tinyxml读取中文时显示乱码的解决方法

tinyxml解析器是一个很好用的，乱码是由于GB2312与UTF8之间转换不当造成的,tinyxml在处理UTF8本身没有问题。

这里是网上找的一个经测试可以用的转换方法： printf("%s", U("中文汉字"));  即可

代码如下：

#include <Windows.h>  
  
#define U   (CUtf8String)  
#define W   (CGb2312String)  
  
class CUtf8String  
{  
public:  
    inline CUtf8String(const char* gb2312)  
    {  
        m_bIsConst = true;  
#ifdef TEST_TYPE_MAP  
        if (m[gb2312])  
        {  
            m_utf8 = m[gb2312];  
            return ;  
        }  
#endif  
        int buffLen = 0;  
        WCHAR wbuff[5120];  
        MultiByteToWideChar(CP_ACP, 0, gb2312, -1, wbuff, 5120);  
        buffLen = WideCharToMultiByte(CP_UTF8, 0, wbuff, -1, NULL, 0, 0, 0);  
        m_utf8 = new char[buffLen+1];  
        WideCharToMultiByte(CP_UTF8, 0, wbuff, -1, (LPSTR)m_utf8, buffLen, 0, 0);  
#ifdef TEST_TYPE_MAP  
        m[gb2312] = m_utf8;  
#endif  
    }  
  
    inline CUtf8String(char* gb2312)  
    {  
        m_bIsConst = false;  
        int buffLen = 0;  
        WCHAR wbuff[5120];  
        MultiByteToWideChar(CP_ACP, 0, gb2312, -1, wbuff, 5120);  
        buffLen = WideCharToMultiByte(CP_UTF8, 0, wbuff, -1, NULL, 0, 0, 0);  
        m_utf8 = new char[buffLen+1];  
        WideCharToMultiByte(CP_UTF8, 0, wbuff, -1, (LPSTR)m_utf8, buffLen, 0, 0);  
    }  
  
    inline ~CUtf8String()  
    {  
#ifndef TEST_TYPE_MAP  
        if (m_utf8)  
        {  
            delete m_utf8;  
            m_utf8 = 0;  
        }  
#else  
        if (!m_bIsConst)  
        {  
            if (m_utf8)  
            {  
                delete m_utf8;  
                m_utf8 = 0;  
            }  
        }  
#endif  
    }  
  
    inline operator char*()  
    {  
        return (char*)m_utf8;  
    }  
private:  
    const char* m_utf8;  
    bool m_bIsConst;  
#ifdef TEST_TYPE_MAP  
    static strmap m;  
#endif  
};  
  
class CGb2312String  
{  
public:  
    inline CGb2312String(const char* utf8)  
    {  
#ifdef TEST_TYPE_MAP  
        if (m[utf8])  
        {  
            m_gb2312 = 0;  
            m_gb2312 = m[utf8];  
        }  
#endif  
        int buffLen = 0;  
        WCHAR wbuff[5120];  
        MultiByteToWideChar(CP_UTF8, 0, utf8, -1, wbuff, 5120);  
        buffLen = WideCharToMultiByte(CP_ACP, 0, wbuff, -1, NULL, 0, 0, 0);  
        m_gb2312 = new char[buffLen+1];  
        WideCharToMultiByte(CP_ACP, 0, wbuff, -1, (LPSTR)m_gb2312, buffLen, 0, 0);  
#ifdef TEST_TYPE_MAP  
        m[utf8] = m_gb2312;  
#endif  
    }  
  
    inline CGb2312String(char* utf8)  
    {  
#ifdef TEST_TYPE_MAP  
        if (m[utf8])  
        {  
            m_gb2312 = 0;  
            m_gb2312 = m[utf8];  
        }  
#endif  
        int buffLen = 0;  
        WCHAR wbuff[5120];  
        MultiByteToWideChar(CP_UTF8, 0, utf8, -1, wbuff, 5120);  
        buffLen = WideCharToMultiByte(CP_ACP, 0, wbuff, -1, NULL, 0, 0, 0);  
        m_gb2312 = new char[buffLen+1];  
        WideCharToMultiByte(CP_ACP, 0, wbuff, -1, (LPSTR)m_gb2312, buffLen, 0, 0);  
#ifdef TEST_TYPE_MAP  
        m[utf8] = m_gb2312;  
#endif  
    }  
  
    inline ~CGb2312String()  
    {  
#ifndef TEST_TYPE_MAP  
        if (m_gb2312)  
        {  
            delete m_gb2312;  
            m_gb2312 = 0;  
        }  
#endif  
    }  
  
    inline operator char*()  
    {  
        return (char*)m_gb2312;  
    }  
private:  
    const char* m_gb2312;  
#ifdef TEST_TYPE_MAP  
    static strmap m;  
#endif  
};  

载自： http://blog.csdn.net/xiaoxiaoyusheng2012/article/details/45315955