 |
Multimedia »
General Graphics »
General
Intermediate
Rendering GIF, JPEG, Icon, or Bitmap Files with OleLoadPictureBy Wes RogersEasy way to do basic rendering of image files into a view window |
VC6, Windows, Visual Studio, MFC, Dev
|
|
Advanced Search Add to IE Search |
|
|
|
||||||||||||||||
This article describes a simple way to do basic rendering of various types of image files into a view window.
It uses a rendering method described in the MSDN article Q218972, LOADPIC.EXE.
I ran across this sample application purely by accident. After reviewing it, I re-wrote the guts of the functionality for use in an MFC document-view environment.
The re-write consists of three functions:
DrawImage(), which draws the image contained in an image file.
GetImageSize(), which returns the height and width of the image in the image file.
LoadPictureFile(), which loads the image from disk and readies it for display via OleLoadPicture().
BOOL DrawImage(CDC* pDC, //Device context CString csFile, //Pathname to image file CPoint ptCenter, //Center point for image HWND hWnd, //Handle to view window CRect rectImage); //Display area rectangle
DrawImage() will render the image whose pathname
is passed in in csFile into the area of
the rectangle in rectImage.
rectImage must be in device units.
Important: These functions assume a DC mapping
mode other than MM_TEXT. That is, negative Y is downward.
If you are going to use MM_TEXT, you will need to change the
signs of the height variables in the Render() function call.
BOOL GetImageSize(CDC* pDC, //Device context LPCTSTR szFile, //Pathname to image file int* pnHeight, //Height to return int* pnWidth); //Width to return
GetImageSize() returns the height and width of
the image in szFile. These values are in device units for the DC
passed in to the function.
It is good to use this function to fetch the true size of the image, in case you want to let the user reset the image display.
BOOL LoadPictureFile(LPCTSTR szFile, //Image file pathname LPPICTURE* pgpPicture); //Ptr to PICTURE for image
LoadPictureFile() loads the image bytes in szFile into pgpPicture.
However, I am using these functions in an MFC application with no special includes.
MSDN does not mention any need for OleLoadPicture.
BOOL DrawImage(CDC* pDC,
CString csFile,
CPoint ptCenter,
HWND hWnd,
CRect rectImage)
{
if (pDC == NULL || csFile.IsEmpty() || hWnd == NULL)
return FALSE;
LPPICTURE gpPicture = NULL;
if (LoadPictureFile((LPCTSTR)csFile,&gpPicture))
{
//Get width and height of picture
long hmWidth = 0;
long hmHeight = 0;
gpPicture->get_Width (&hmWidth);
gpPicture->get_Height(&hmHeight);
//Use to get height and width for display
CRect rectI = rectImage;
rectI.NormalizeRect();
int nWidth = rectI.Width();
int nHeight= rectI.Height();
CPoint ptUL(ptCenter.x-(nWidth /2),
ptCenter.y+(nHeight/2));
RECT rc;
GetClientRect(hWnd, &rc);
HRESULT hrP = NULL;
hrP =
gpPicture->Render(pDC->m_hDC,
ptUL.x,
ptUL.y,
nWidth,
-nHeight,
0,
hmHeight,
hmWidth,
-hmHeight,
&rc);
gpPicture->Release();
if (SUCCEEDED(hrP))
return TRUE;
}
return FALSE;
}
BOOL GetImageSize(CDC* pDC,
LPCTSTR szFile,
int* pnHeight,
int* pnWidth)
{
LPPICTURE gpPicture = NULL;
if (!LoadPictureFile(szFile,&gpPicture))
return FALSE;
//Get width and height of picture
long hmWidth = 0;
long hmHeight = 0;
gpPicture->get_Width (&hmWidth);
gpPicture->get_Height(&hmHeight);
int nPixX = pDC->GetDeviceCaps(LOGPIXELSX);
int nPixY = pDC->GetDeviceCaps(LOGPIXELSY);
//For printers, pixels per inch is really DPI, so be careful!
if (pDC->IsPrinting())
{
nPixX = 96;
nPixY = 96;
}
//Convert to device units
*pnWidth = MulDiv(hmWidth, nPixX, HIMETRIC_INCH);
*pnHeight = MulDiv(hmHeight, nPixY, HIMETRIC_INCH);
return TRUE;
}
BOOL LoadPictureFile(LPCTSTR szFile,
LPPICTURE* pgpPicture)
{
// open file
HANDLE hFile = CreateFile(szFile,
GENERIC_READ,
0,
NULL,
OPEN_EXISTING,
0,
NULL);
if (hFile == INVALID_HANDLE_VALUE)
{
AfxMessageBox ("Could not read file");
return FALSE;
}
// get file size
DWORD dwFileSize = GetFileSize(hFile, NULL);
if (dwFileSize == (DWORD)-1)
{
CloseHandle(hFile);
AfxMessageBox ("File seems to be empty");
return FALSE;
}
LPVOID pvData = NULL;
// alloc memory based on file size
HGLOBAL hGlobal = GlobalAlloc(GMEM_MOVEABLE, dwFileSize);
if (hGlobal == NULL)
{
CloseHandle(hFile);
AfxMessageBox ("Could not allocate memory for image");
return FALSE;
}
pvData = GlobalLock(hGlobal);
if (pvData == NULL)
{
GlobalUnlock(hGlobal);
CloseHandle(hFile);
AfxMessageBox ("Could not lock memory for image");
return FALSE;
}
DWORD dwBytesRead = 0;
// read file and store in global memory
BOOL bRead = ReadFile(hFile,
pvData,
dwFileSize,
&dwBytesRead,
NULL);
GlobalUnlock(hGlobal);
CloseHandle(hFile);
if (!bRead)
{
AfxMessageBox ("Could not read file");
return FALSE;
}
LPSTREAM pstm = NULL;
// create IStream* from global memory
HRESULT hr = CreateStreamOnHGlobal(hGlobal,
TRUE,
&pstm);
if (!(SUCCEEDED(hr)))
{
AfxMessageBox ("CreateStreamOnHGlobal() failed");
if (pstm != NULL)
pstm->Release();
return FALSE;
}
else if (pstm == NULL)
{
AfxMessageBox ("CreateStreamOnHGlobal() failed");
return FALSE;
}
// Create IPicture from image file
if (*pgpPicture)
(*pgpPicture)->Release();
hr = ::OleLoadPicture(pstm,
dwFileSize,
FALSE,
IID_IPicture,
(LPVOID *)&(*pgpPicture));
if (!(SUCCEEDED(hr)))
{
pstm->Release();
AfxMessageBox("Could not load image (hr failure)");
return FALSE;
}
else if (*pgpPicture == NULL)
{
pstm->Release();
AfxMessageBox("Could not load image (pgpPicture failure)");
return FALSE;
}
pstm->Release();
return TRUE; //Made it ...!
}
The picture above shows use of these functions in a map-making program I have been working on.
Here is how the program uses them:
First a handler for the "Map" menu allows a user to navigate to the desired image file and open it. This simply saves off the file path for use later, and sets the program into "rendering images" mode: Wherever the user clicks, a stroke "image" object will be created to display the image.
The CStroke class handles this (remember Scribble?).
Once the program is in "rendering images" mode, the view class
handles the left mouse button click.
Here is the code block from OnLButtonDown():
void CGameMaprView::OnLButtonDown(UINT nFlags, CPoint point) { CGameMaprDoc* pDoc = GetDocument(); CRect rectTemp; CClientDC dc(this); OnPrepareDC(&dc); dc.DPtoLP(&point); ... //--------------------------------------- else if (pDoc->DrawType() == DRAW_IMAGE_FILE) { m_pStrokeCur = pDoc->NewStroke(FALSE, &dc, point.x, point.y, pDoc->BitmapResourceName()); //Pathname to image file if (m_pStrokeCur != NULL) { m_pStrokeCur->DrawStroke(&dc,FALSE,this); //Invalidate for re-display pDoc->RedrawArea(m_pStrokeCur->m_rectBounding); } } ... return; }
Back in the document class, NewStroke() creates a new stroke,
and gets the original image size for the initial bounding box:
CStroke* CGameMaprDoc::NewStroke(BOOL bPasting,
CDC *pDC,
int nX,
int nY,
CString csBmp) //Image file in this case...
{
...
else if (m_eDrawType == DRAW_IMAGE_FILE)
{
eType = S_TYPE_IMAGE_FILE;
}
pStrokeItem = new CStroke( ... );
...
//Do bitmap-related processing
else if (m_eDrawType == DRAW_IMAGE_FILE &&
!csBmp.IsEmpty())
{
int nH = 0;
int nW = 0;
//Set the original size of the image
if (GetImageSize(pDC,(LPCTSTR)csBmp,&nH,&nW))
{
pStrokeItem->m_csBitmapResource = csBmp;
pStrokeItem->m_rectBounding.SetRect(nX-(nW/2),
nY+(nH/2),
nX+(nW/2),
nY-(nH/2));
}
else
{
AfxMessageBox("Could not create image object");
delete pStrokeItem;
return NULL;
}
}
...
}
CStroke class, the DrawStroke() function handles drawing
of the image:
BOOL CStroke::DrawStroke(CDC* pDC,
BOOL bDirectDisplay,
CGameMaprView* pView)
{
...
//Render an image file-----------------------
else if (StrokeType() == S_TYPE_IMAGE_FILE)
{
DrawImage(pDC,
m_csBitmapResource,
m_rectBounding.CenterPoint(),
pView->m_hWnd,
m_rectBounding);
}
...
}
The stroke class also has an "update properties" method that allows a user to specify a different height and width for the image.
The biggest one is the aspect ratio:
LOGPIXELSX and LOGPIXELSY, to a printer's Device Context (DC)
mean the number of Dots Per Inch (DPI).
On a 600 DPI printer, these values will
be returned as 600.
If you are not careful, this will have "interesting" results when you print!
The second issue is transparency. Hewlett-Packard printers,
for example,
do not seem to recognize the BitBlt ROP codes properly,
making it impossible to render an image transparently by the normally-shown methods.
(This may have been fixed on later models - HP never responded when I tried to contact them about it.)
This limitation also exists in the above functions:
You will not be able to print a transparent image
(such as the knight in the image sample above) using OleLoadImage().
Non-transparent printing seems to work fine.
To do transparent printing, you would need to turn the image into a bitmap, and then print it using regions of color rather than bit-blitting with the "true mask" method.
LOADPIC states that it will open Enhanced metafiles and Windows metafiles (EMF and WMF formats).
I successfully loaded two EMF's; however, when I tried to load c:\windows\system\spbanner.wmf, I got:
"Could not load image (hr failure)". Word 97 would not load this file properly either.
If desired, a reader could add further analysis code to list the detailed cause of such failures.
I have found these functions very useful; it is my hope that other readers will also find them useful for rendering various types of images.
General 
News 
Question 
Answer 
Joke 
Rant 
Admin
|
PermaLink |
Privacy |
Terms of Use
Last Updated: 26 Sep 2000 Editor: Chris Maunder |
Copyright 2000 by Wes Rogers Everything else Copyright © CodeProject, 1999-2009 Web15 | Advertise on the Code Project |
i have the same question
