码迷,mamicode.com
首页 > 编程语言 > 详细

C++ writestring 为什么不能写进中文 CStdioFile向无法向文本中写入中文【二】

时间:2014-05-01 01:47:39      阅读:712      评论:0      收藏:0      [点我收藏+]

标签:des   com   http   blog   style   class   code   c   log   t   size   

本地化设置需要具备三个条件:
a. 语言代码 (Language Code)
b. 国家代码 (Country Code)
c. 编码(Encoding)
本地名字可以用下面这些部分来构造:
语言代码_国家代码.编码 比如(zh_CN.UTF-8, en_US等)

locale的别名表见 /usr/lib/X11/locale/locale.alias(以Debian GNU/Linux为例)
setlocale语言字符串参考

本文来自CSDN博客,转载请标明出处:http://blog.csdn.net/ALENTAM/archive/2008/04/11/2281121.aspx

另外还有一种方法就是重新写CStdioFile的派生类CStdioFileEx(网上有)。

//好像C++中没有类能够读些Unicode格式的文本文件,所以我写了下面这个类。用法很简单,大家尝试几下就明白了。

#pragma once

class CStdioFileEx: public CStdioFile
{
public:
CStdioFileEx();
CStdioFileEx( LPCTSTR lpszFileName, UINT nOpenFlags );

virtual BOOL Open( LPCTSTR lpszFileName, UINT nOpenFlags, CFileException* pError = NULL );
virtual BOOL ReadString(CString& rString);
BOOL ReadWideString(CStringW& rString);
BOOL ReadAnsiString(CStringA& rString);
virtual void WriteString(LPCTSTR lpsz);
void WriteWideString(LPCWSTR lpsz);
void WriteAnsiString(LPCSTR lpsz);
bool IsUnicodeFormat() {return m_bIsUnicodeText;}
unsigned long GetCharCount();

// Additional flag to allow Unicode text format writing
enum {modeWriteUnicode = 0x100000};

static bool IsFileUnicode(const CString& sFilePath);

protected:
UINT PreprocessFlags(const CString& sFilePath, UINT& nOpenFlags);

bool m_bIsUnicodeText;
};

//。cpp文件
#include "stdafx.h"
#include "StdioFileEx.h"

//在UCS 编码中有一个叫做"ZERO WIDTH NO-BREAK SPACE"的字符,它的编码是FEFF。而FFFE在UCS中是不存在的字符,
//所以不应该出现在实际传输中。UCS规范建议我们在传输字节流前,先传输字符"ZERO WIDTH NO-BREAK SPACE"。这样
//如果接收者收到FEFF,就表明这个字节流是Big-Endian的;如果收到FFFE,就表明这个字节流Little-Endian的。
//因此字符"ZERO WIDTH NO-BREAK SPACE"又被称作BOM。
//UTF-8不需要BOM来表明字节顺序,但可以用BOM来表明编码方式。字符"ZERO WIDTH NO-BREAK SPACE"的UTF-8编码是
//EF BB BF。所以如果接收者收到以EF BB BF开头的字节流,就知道这是UTF-8编码了。
//Windows就是使用BOM来标记文本文件的编码方式的。
//有些老的浏览器和文本编辑器不支持BOM。
#define UNICODE_BOM 0xFEFF//Unicode "byte order mark" which goes at start of file

CStdioFileEx::CStdioFileEx(): CStdioFile()
{
m_bIsUnicodeText = false;
}

CStdioFileEx::CStdioFileEx(LPCTSTR lpszFileName,UINT nOpenFlags)
:CStdioFile(lpszFileName, PreprocessFlags(lpszFileName, nOpenFlags))
{
}

BOOL CStdioFileEx::Open(LPCTSTR lpszFileName,UINT nOpenFlags,CFileException* pError /*=NULL*/)
{
PreprocessFlags(lpszFileName, nOpenFlags);

return CStdioFile::Open(lpszFileName, nOpenFlags, pError);
}

BOOL CStdioFileEx::ReadString(CString& rString)
{
#ifdef _UNICODE
return ReadWideString(rString);
#else
return ReadAnsiString(rString);
#endif
}

BOOL CStdioFileEx::ReadWideString(CStringW& rString)
{
_ASSERTE(m_pStream);
rString = L""; // empty string without deallocating

if(m_bIsUnicodeText)
{
// If at position 0, discard byte-order mark before reading
if(GetPosition() == 0)
{
wchar_t bom;
Read(&bom, sizeof(wchar_t));
}
const int nMaxSize = 128;
LPWSTR lpsz = rString.GetBuffer(nMaxSize);
LPWSTR lpszResult;
int nLen = 0;
for (;;)
{
lpszResult = fgetws(lpsz, nMaxSize+1, m_pStream);
rString.ReleaseBuffer();

// handle error/eof case
if (lpszResult == NULL && !feof(m_pStream))
{
Afx_clearerr_s(m_pStream);
AfxThrowFileException(CFileException::genericException, _doserrno,
m_strFileName);
}

// if string is read completely or EOF
if (lpszResult == NULL ||
(nLen = (int)lstrlenW(lpsz)) < nMaxSize ||
lpsz[nLen-1] == ‘\n‘)
break;

nLen = rString.GetLength();
lpsz = rString.GetBuffer(nMaxSize + nLen) + nLen;
}
//remove crlf if exist.
nLen = rString.GetLength();
if (nLen > 1 && rString.Mid(nLen-2) == L"\r\n")
{
rString.GetBufferSetLength(nLen-2);
}
return rString.GetLength() > 0;
}
else
{
CStringA ansiString;
BOOL bRetval = ReadAnsiString(ansiString);
//setlocale(LC_ALL, "chs_chn.936");//no need
rString = ansiString;
return bRetval;
}
}

BOOL CStdioFileEx::ReadAnsiString(CStringA& rString)
{
_ASSERTE(m_pStream);
rString = ""; // empty string without deallocating

if(!m_bIsUnicodeText)
{
const int nMaxSize = 128;
LPSTR lpsz = rString.GetBuffer(nMaxSize);
LPSTR lpszResult;
int nLen = 0;
for (;;)
{
lpszResult = fgets(lpsz, nMaxSize+1, m_pStream);
rString.ReleaseBuffer();

// handle error/eof case
if (lpszResult == NULL && !feof(m_pStream))
{
Afx_clearerr_s(m_pStream);
AfxThrowFileException(CFileException::genericException, _doserrno,
m_strFileName);
}

// if string is read completely or EOF
if (lpszResult == NULL ||
(nLen = (int)lstrlenA(lpsz)) < nMaxSize ||
lpsz[nLen-1] == ‘\n‘)
break;

nLen = rString.GetLength();
lpsz = rString.GetBuffer(nMaxSize + nLen) + nLen;
}
//remove crlf if exist.
nLen = rString.GetLength();
if (nLen > 1 && rString.Mid(nLen-2) == "\r\n")
{
rString.GetBufferSetLength(nLen-2);
}
return rString.GetLength() > 0;
}
else
{
CStringW wideString;
BOOL bRetval = ReadWideString(wideString);
//setlocale(LC_ALL, "chs_chn.936");//no need
rString = wideString;
return bRetval;
}
}

// Purpose: Writes string to file either in Unicode or multibyte, depending on whether the caller specified the
// CStdioFileEx::modeWriteUnicode flag. Override of base class function.
void CStdioFileEx::WriteString(LPCTSTR lpsz)
{
#ifdef _UNICODE
WriteWideString(lpsz);
#else
WriteAnsiString(lpsz);
#endif
}

void CStdioFileEx::WriteWideString(LPCWSTR lpsz)
{
ASSERT(lpsz != NULL);

if (lpsz == NULL)
{
AfxThrowInvalidArgException();
}
if(m_bIsUnicodeText)
{
ASSERT(m_pStream != NULL);
// If writing Unicode and at the start of the file, need to write byte mark
if(GetPosition() == 0)
{
wchar_t cBOM = (wchar_t)UNICODE_BOM;
CFile::Write(&cBOM, sizeof(wchar_t));
}
if (fputws(lpsz, m_pStream) == _TEOF)
AfxThrowFileException(CFileException::diskFull, _doserrno, m_strFileName);
}
else
{
USES_CONVERSION;
WriteAnsiString(CW2A(lpsz));
}
}

void CStdioFileEx::WriteAnsiString(LPCSTR lpsz)
{
ASSERT(lpsz != NULL);

if (lpsz == NULL)
{
AfxThrowInvalidArgException();
}
if(!m_bIsUnicodeText)
{
ASSERT(m_pStream != NULL);
if (fputs(lpsz, m_pStream) == _TEOF)
AfxThrowFileException(CFileException::diskFull, _doserrno, m_strFileName);
}
else
{
USES_CONVERSION;
WriteWideString(CA2W(lpsz));
}
}

UINT CStdioFileEx::PreprocessFlags(const CString& sFilePath, UINT& nOpenFlags)
{
m_bIsUnicodeText = false;

// If we have writeUnicode we must have write or writeRead as well
if (nOpenFlags & CStdioFileEx::modeWriteUnicode)
{
ASSERT(nOpenFlags & CFile::modeWrite || nOpenFlags & CFile::modeReadWrite);
m_bIsUnicodeText = true;
}
// If reading in text mode and not creating...
else if (nOpenFlags & CFile::typeText && !(nOpenFlags & CFile::modeCreate) && !(nOpenFlags & CFile::modeWrite ))
{
m_bIsUnicodeText = IsFileUnicode(sFilePath);
}

//如果要读写Unicode格式的文本文件, 必须切换到typeBinary方式, 因为这会影响fputws/fgetws的工作方式(具体情况参考MSDN)。
if (m_bIsUnicodeText)
{
nOpenFlags &= ~(CFile::typeText);
nOpenFlags |= CFile::typeBinary;
}

return nOpenFlags;
}

// Purpose: Determines whether a file is Unicode by reading the first character and detecting
// whether it‘s the Unicode byte marker.
bool CStdioFileEx::IsFileUnicode(const CString& sFilePath)
{
CFile file;
wchar_t cFirstChar;
CFileException exFile;

bool bIsUnicode = false;
// Open file in binary mode and read first character
if (file.Open(sFilePath, CFile::typeBinary | CFile::modeRead, &exFile))
{
// If byte is Unicode byte-order marker, let‘s say it‘s Unicode
if (file.Read(&cFirstChar, sizeof(wchar_t)) > 0 && cFirstChar == (wchar_t)UNICODE_BOM)
{
bIsUnicode = true;
}

file.Close();
}
else
{
// Handle error here if you like
}

return bIsUnicode;
}

unsigned long CStdioFileEx::GetCharCount()
{
int nCharSize;
unsigned long nByteCount, nCharCount = 0;

if (m_pStream)
{
// Get size of chars in file
nCharSize = m_bIsUnicodeText ? sizeof(wchar_t): sizeof(char);

// If Unicode, remove byte order mark from count
nByteCount = (unsigned long)GetLength();

if (m_bIsUnicodeText)
{
nByteCount = nByteCount - sizeof(wchar_t);
}

// Calc chars
nCharCount = (nByteCount / nCharSize);
}

return nCharCount;
}

C++ writestring 为什么不能写进中文 CStdioFile向无法向文本中写入中文【二】,码迷,mamicode.com

C++ writestring 为什么不能写进中文 CStdioFile向无法向文本中写入中文【二】

标签:des   com   http   blog   style   class   code   c   log   t   size   

原文地址:http://www.cnblogs.com/XiHua/p/3701110.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!