文件倒序示例应用程序关键代码（使用内存映射文件）——摘自Windows 核心编程

BOOL FileReverse(PCTSTR pszPathname, PBOOL pfIsTextUnicode) {

   *pfIsTextUnicode = FALSE;  // Assume text is Unicode

   // Open the file for reading and writing.
   HANDLE hFile = CreateFile(pszPathname, GENERIC_WRITE | GENERIC_READ, 0,
      NULL, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);

   if (hFile == INVALID_HANDLE_VALUE) {
      chMB("File could not be opened.");
      return(FALSE);
   }

   // Get the size of the file (I assume the whole file can be mapped).
   DWORD dwFileSize = GetFileSize(hFile, NULL);

   // Create the file-mapping object. The file-mapping object is 1 character
   // bigger than the file size so that a zero character can be placed at the
   // end of the file to terminate the string (file). Because I don't yet know
   // if the file contains ANSI or Unicode characters, I assume worst case
   // and add the size of a WCHAR instead of CHAR.
   HANDLE hFileMap = CreateFileMapping(hFile, NULL, PAGE_READWRITE,
      0, dwFileSize + sizeof(WCHAR), NULL);

   if (hFileMap == NULL) {
      chMB("File map could not be opened.");
      CloseHandle(hFile);
      return(FALSE);
   }

   // Get the address where the first byte of the file is mapped into memory.
   PVOID pvFile = MapViewOfFile(hFileMap, FILE_MAP_WRITE, 0, 0, 0);

   if (pvFile == NULL) {
      chMB("Could not map view of file.");
      CloseHandle(hFileMap);
      CloseHandle(hFile);
      return(FALSE);
   }

   // Does the buffer contain ANSI or Unicode?
   int iUnicodeTestFlags = -1;   // Try all tests
   *pfIsTextUnicode = IsTextUnicode(pvFile, dwFileSize, &iUnicodeTestFlags);

   if (!*pfIsTextUnicode) {
      // For all the file manipulations below, we explicitly use ANSI
      // functions because we are processing an ANSI file.

      // Put a zero character at the very end of the file.
      PSTR pchANSI = (PSTR) pvFile;
      pchANSI[dwFileSize / sizeof(CHAR)] = 0;

      // Reverse the contents of the file.
      _strrev(pchANSI);

      // Convert all "/n/r" combinations back to "/r/n" to
      // preserve the normal end-of-line sequence.
      pchANSI = strchr(pchANSI, '/n'); // Find first '/n'.

      while (pchANSI != NULL) {
         // We have found an occurrence....
         *pchANSI++ = '/r';   // Change '/n' to '/r'.
         *pchANSI++ = '/n';   // Change '/r' to '/n'.
         pchANSI = strchr(pchANSI, '/n'); // Find the next occurrence.
      }

   } else {
      // For all the file manipulations below, we explicitly use Unicode
      // functions because we are processing a Unicode file.

      // Put a zero character at the very end of the file.
      PWSTR pchUnicode = (PWSTR) pvFile;
      pchUnicode[dwFileSize / sizeof(WCHAR)] = 0;

      if ((iUnicodeTestFlags & IS_TEXT_UNICODE_SIGNATURE) != 0) {
         // If the first character is the Unicode BOM (byte-order-mark),
         // 0xFEFF, keep this character at the beginning of the file.
         pchUnicode++;
      }

      // Reverse the contents of the file.
      _wcsrev(pchUnicode);

      // Convert all "/n/r" combinations back to "/r/n" to
      // preserve the normal end-of-line sequence.
      pchUnicode = wcschr(pchUnicode, L'/n'); // Find first '/n'.

      while (pchUnicode != NULL) {
         // We have found an occurrence....
         *pchUnicode++ = L'/r';   // Change '/n' to '/r'.
         *pchUnicode++ = L'/n';   // Change '/r' to '/n'.
         pchUnicode = wcschr(pchUnicode, L'/n'); // Find the next occurrence.
      }
   }

   // Clean up everything before exiting.
   UnmapViewOfFile(pvFile);
   CloseHandle(hFileMap);

   // Remove trailing zero character added earlier.
   SetFilePointer(hFile, dwFileSize, NULL, FILE_BEGIN);
   SetEndOfFile(hFile);
   CloseHandle(hFile);

   return(TRUE);
}