win32 读写UTF-8格式的文件的方法

news2025/7/27 20:57:59

1，写入数据

最开始是在写入数据前先写入三个字节

BYTE btHead[] = { 0xEF,0xBB,0xBF };
::WriteFile(hFile, btHead, 3, &dwWrite, 0);
::WriteFile(hFile, str, lstrlen(str)*sizeof(TCHAR), &dwWrite, 0);

这样写入后文件样式为：
在这里插入图片描述
格式是UTF-8，但后面有个BOM，而且汉字也是乱码，
最后我又测试了下用c语言写入UTF-8数据，

//打开文件，按指定编码格式打开
FILE* fp=fopen("UTF_8Test.txt", "wt+,ccs=UTF-8");

//写入数据
TCHAR str[] = TEXT("测试文件,1234,abcd,ABCD");
fwrite(str, sizeof(TCHAR), lstrlen(str), fp);

//关闭
fclose(fp);

汉字不是乱码了，但是格式仍然是UTF-8 BOM

这就是不我想要的效果。

最后查找一翻资料后，
1。先把要保存的字符串转化为UTF-8格式的char字符串
2。在写入char字符串到文件中

//lpBuffer是传来的宽字符串的指针
//nBufferLeng是字符串的字符个数（用lstrlen求出的长度）

//求字符串转为UTF-8的字符长度
int size = ::WideCharToMultiByte(CP_UTF8, 0, (LPTSTR)lpBuffer, nBufferLeng, NULL, 0, NULL,
	NULL);
	
//再转为char字符串
char* pTem = new char[size+1];
memset(pTem, 0, size);
::WideCharToMultiByte(CP_UTF8, 0, (LPTSTR)lpBuffer, nBufferLeng, pTem, size, nullptr, nullptr);
pTem[size] = '\0';

//最后写入char字符串
bResult = ::WriteFile(m_hFile, pTem, size, &dwWrite, 0);
delete[] pTem;

这样写入文件的效果：

在这里插入图片描述
这样汉字就不会出现乱码了，格式也是UTF-8，没有后面的BOM了

2。读取数据

数据的读取和写入一样，先读取数据到char数组中，再转为UTF-8格式的宽字符串中

//获得文件字节长度
int n = GetFileSize();

//写入char数组中
char* pChar = new char[n+1];
bResult = ::ReadFile(m_hFile, pChar, n, &dwRead, 0);
pChar[n] = '\0';

//计算字符串转UTF-8的字符长度
int len = MultiByteToWideChar(CP_UTF8, 0, pChar, strlen(pChar), nullptr, 0);

//转为TCHAR字符数组
TCHAR* pWChar = new TCHAR[len+1];
memset(pWChar, 0, len);
MultiByteToWideChar(CP_UTF8, 0, pChar, strlen(pChar), pWChar, len);
pWChar[len] = '\0';

//拷贝到函数外面的字符指针
lstrcpy((LPTSTR)lpBuffer, pWChar);

//清除内存
delete[] pWChar;
delete[] pChar;

读取数据的效果：
在这里插入图片描述

4。完整代码片段

QFile.h文件

//文件操作类
// 对Ascii，Unicode,UTF_8,Binary
// 四种文件格式的读写
class QFile
{
public:
	QFile();
	~QFile();
public:
	enum OpenModule	//文件打开模式
	{
		Read = GENERIC_READ,	//只读模式
		Write = GENERIC_WRITE,	//只写模式
		WriteRead = GENERIC_WRITE | GENERIC_READ //读写模式
	};

	enum FileEncoded //文件编码
	{
		enAnsii=0,	//Ansi编码
		enUnicode,	//Unicode编码
		enUTF_8,	//UTF——8编码
		enBinary	//二进制编码
	};

public:
	//以指定模式打开文件
	BOOL Open(const TCHAR* pFileName, OpenModule openModule);
	void Close();
	BOOL FileExists(const TCHAR* pFileName);//检查文件是否存在
	BOOL CreateNewFile(const TCHAR* pFileName); //创建新文件
	//以指定编码格式写入数据
	int WriteFile(void* lpBuffer, int nBufferLeng, FileEncoded encodedType);
	//以指定编码格式读取数据
	int ReadFile(void* lpBuffer, int nBufferLeng, FileEncoded encodedType);
	//获得文件的字节长度
	int GetFileSize()const;
public:
	HANDLE m_hFile;
};

QFile.cpp文件

//写入数据
int QFile::WriteFile(void* lpBuffer, int nBufferLeng, FileEncoded encodedType)
{
	if (lpBuffer == nullptr)
		return -1;

	if (m_hFile == INVALID_HANDLE_VALUE)
		return -1;

	DWORD dwWrite = 0;
	BOOL bResult = FALSE;

	//编码UTF-8的写入
	if (encodedType == enUTF_8)
	{
		int size = ::WideCharToMultiByte(CP_UTF8, 0, (LPTSTR)lpBuffer, nBufferLeng, NULL, 0, NULL,
			NULL);
		char* pTem = new char[size];
		memset(pTem, 0, size);
		::WideCharToMultiByte(CP_UTF8, 0, (LPTSTR)lpBuffer, nBufferLeng, pTem, size, nullptr, nullptr);
		pTem[size] = '\0';
		bResult = ::WriteFile(m_hFile, pTem, size, &dwWrite, 0);
		return bResult ? dwWrite : -1;
	}
	

	//Unicode编码先写入两个字节，再写入数据
	if (encodedType == enUnicode)
	{
		BYTE btHead[] = { 0xFF,0xFE };
		bResult = ::WriteFile(m_hFile, btHead, 2, &dwWrite, 0);
	}

	//其他的都是直接写入数据
	bResult = ::WriteFile(m_hFile, lpBuffer, nBufferLeng, &dwWrite, 0);

	return bResult ? dwWrite : -1;
}

//读取数据
int QFile::ReadFile(void* lpBuffer, int nBufferLeng, FileEncoded encodedType)
{
	if (lpBuffer == nullptr)
		return -1;

	if (m_hFile == INVALID_HANDLE_VALUE)
		return -1;

	DWORD dwRead = 0;
	BOOL bResult = FALSE;

	//编码格式UTF-8的读取
	if (encodedType == enUTF_8)
	{
		char* pChar = new char[nBufferLeng +1];
		bResult = ::ReadFile(m_hFile, pChar, nBufferLeng, &dwRead, 0);
		pChar[nBufferLeng] = '\0';

		int len = MultiByteToWideChar(CP_UTF8, 0, pChar, strlen(pChar), nullptr, 0);

		TCHAR* pWChar = new TCHAR[len+1];
		memset(pWChar, 0, len);
		MultiByteToWideChar(CP_UTF8, 0, pChar, strlen(pChar), pWChar, len);
		pWChar[len] = '\0';

		lstrcpy((LPTSTR)lpBuffer, pWChar);
		delete[] pWChar;
		delete[] pChar;

		return bResult ? dwRead : -1;
	}

	//编码Unicode，先跳过前两个字节再读取数据
	if (encodedType == enUnicode)
	{
		SetFilePointer(m_hFile, 2, 0, FILE_BEGIN);
	}
	bResult = ::ReadFile(m_hFile, lpBuffer, nBufferLeng, &dwRead, 0);
	
	return bResult ? dwRead : -1;
}

//…………

5。使用

//1.写入数据
QFile f;
f.Open(Utf_8Txt, QFile::Write);

TCHAR str[] = TEXT("测试数据,1234,ABCD");
f.WriteFile(str, lstrlen(str), QFile::enUTF_8);

f.Close();

//2.读取数据
QFile f;
f.Open(Utf_8Txt, QFile::Read);

TCHAR str[20] = { 0 };
int nLeng = f.GetFileSize();

f.ReadFile(str, nLeng, QFile::enUTF_8);

f.Close();

MessageBox(nullptr, str, TEXT("数据的读取"), 0);