🚀 go-pugleaf

I assume that the code you posted has a cut and paste error, that
anisotropic mode, window size and viewport size are set long before the
mouse move message is received.

If so then try these things:
  while debugging break at the SetMapMode call and examine the
    values of xwin, ywin, width and height to make sure they are
    as you expect
  place the call  SetErrorMode(0);  prior to your call to DPtoLP

If not then try moving the SetMapMode, SetWindowExtEx and SetViewportExtEx
calls to the code associated with processing WM_CREATE

If you have difficulty deciphering the meaning for a GetLastError return
value then you might call this function which will display the text meaning
of the message in a message box.

  -----------------
  void ShowLastError(DWORD err_number)
    {
    TCHAR *err_msg;

    FormatMessage(
        FORMAT_MESSAGE_ALLOCATE_BUFFER |
            FORMAT_MESSAGE_FROM_SYSTEM,
        NULL,
        err_number,
        MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT),
        (LPTSTR)(&err_msg),
        0,
        NULL);

    MessageBox(HWND_DESKTOP, err_msg, "Error", MB_OK);
    LocalFree(err_msg);
    }

  -----------------

.  Ed

> M. Finch wrote in message
> news:3fec8cda$1@newsgroups.borland.com...
>
> I seem to be missing how the DPtoLP function works.  I am using
>
> POINT pt;
>     .
>     .
>   case WM_MOUSEMOVE:
>
>         pt.x = (LONG)LOWORD(lParam);
>         pt.y = (LONG)HIWORD(lParam);
>
>         SetMapMode(hDC,MM_ANISOTROPIC);
>         SetWindowExtEx(hDC, xwin,ywin,NULL);
>         SetViewportExtEx(hDC,width,-height,NULL);
>
> // Mouse moves are stored in pt.x and pt.y in DEVICE
> COORDINATES?? // Using the DPtoLP function should return
> their coordinates in LOGICAL COORDINATES??
>
>         retval = DPtoLP(hDC, &pt, 1);
>         reterr = GetLastError();
>
> I find that I get an error in the retval and I retrieve the error using
> GetLastError() which is 87.  87 is Invalid Parameter as defined
> in System Error Codes on MSDN.
>
> Anyone have any ideas on what is causing the error?
> Should I expect that the coordinates returned are
>     pt.x = pt.x * (xwin / width)
>     pt.y = pt.y * (ywin / height)
>
> which are the device coordinates times  the ratio of logical to
> device coordinates??