This  page is in Pre Alpha. That means no grammer checks,
no editting and speeling mistakes galore! The explanations and the final version
will come soon. Check it out at your own risk!


a.bat
cl.exe /nologo  /I "\inetsdk\include" /c HTMLPars.cpp
link.exe ole32.lib oleaut32.lib urlmon.lib ..\..\lib\uuid.lib user32.lib /subsystem:windows HTMLPars.obj

htmlpars.cpp

#include <windows.h>
#include "..\..\include\mshtml.h"
#include <urlmon.h>
#include <stdio.h>
void abc(char *p){FILE *fp=fopen("z.txt","a+");fprintf(fp,"%s\n",p);fclose(fp);}
void abc1(unsigned short *p){FILE *fp=fopen("z.txt","a+");char *p1= (char *)p;while ( *p1){fprintf(fp,"%c",*p1);p1++;p1++;}fprintf(fp,"\n",*p1);fclose(fp);}
int _stdcall WinMain(HINSTANCE ii,HINSTANCE j,char *k,int l)
{
IUnknown* m_punkMSHTML;
OleInitialize(NULL);
CoCreateInstance( CLSID_HTMLDocument,0,CLSCTX_INPROC_SERVER, IID_IUnknown,
  (void **)&m_punkMSHTML );
OLECHAR  szwName[256];
MultiByteToWideChar(CP_ACP, 0, "http://www.microsoft.com", -1, szwName, 256);    
IMoniker* pIMoniker;
CreateURLMoniker(NULL, (LPWSTR)szwName, &pIMoniker);
IPersistMoniker* pPMk;
m_punkMSHTML->QueryInterface(IID_IPersistMoniker,(void **)&pPMk);
pPMk->Load(FALSE, pIMoniker, 0, STGM_READ);
::MessageBox(0,"hi","hi",0);
IHTMLDocument2* pHTMLDocument2;
m_punkMSHTML->QueryInterface(IID_IHTMLDocument2, (void **)&pHTMLDocument2);
IHTMLElementCollection* pColl;
pHTMLDocument2->get_all( &pColl );
LONG celem;
pColl->get_length( &celem );
for ( int i=0; i< celem; i++ )
{
VARIANT varIndex;
varIndex.vt = VT_UINT;
varIndex.lVal = i;
VARIANT var2;
VariantInit( &var2 );
IDispatch* pDisp; 
pColl->item( varIndex, var2, &pDisp );
IHTMLElement* pElem;
pDisp->QueryInterface( IID_IHTMLElement, (void **)&pElem );
BSTR bstr;
pElem->get_tagName(&bstr);
abc1(bstr);
IHTMLImgElement* pImgElem;
HRESULT hr = pDisp->QueryInterface( IID_IHTMLImgElement, (void **)&pImgElem );
if ( hr == S_OK )
{
pImgElem->get_href(&bstr);
abc1(bstr);
}
}
}