pagerankdlg.cpp

google搜索的核心技术

// pagerankDlg.cpp : implementation file
//

#include "stdafx.h"
#include "pagerank.h"
#include "pagerankDlg.h"

#include "AdjacencyWDigraph.h"
#ifdef _DEBUG
#define new DEBUG_NEW
#undef THIS_FILE
static char THIS_FILE[] = __FILE__;
#endif

/////////////////////////////////////////////////////////////////////////////
// CAboutDlg dialog used for App About


class CAboutDlg : public CDialog
{
public:
	CAboutDlg();

// Dialog Data
	//{{AFX_DATA(CAboutDlg)
	enum { IDD = IDD_ABOUTBOX };
	//}}AFX_DATA

	// ClassWizard generated virtual function overrides
	//{{AFX_VIRTUAL(CAboutDlg)
	protected:
	virtual void DoDataExchange(CDataExchange* pDX);    // DDX/DDV support
	//}}AFX_VIRTUAL

// Implementation
protected:
	//{{AFX_MSG(CAboutDlg)
	//}}AFX_MSG
	DECLARE_MESSAGE_MAP()
};

CAboutDlg::CAboutDlg() : CDialog(CAboutDlg::IDD)
{
	//{{AFX_DATA_INIT(CAboutDlg)
	//}}AFX_DATA_INIT
}

void CAboutDlg::DoDataExchange(CDataExchange* pDX)
{
	CDialog::DoDataExchange(pDX);
	//{{AFX_DATA_MAP(CAboutDlg)
	//}}AFX_DATA_MAP
}

BEGIN_MESSAGE_MAP(CAboutDlg, CDialog)
	//{{AFX_MSG_MAP(CAboutDlg)
		// No message handlers
	//}}AFX_MSG_MAP
END_MESSAGE_MAP()

/////////////////////////////////////////////////////////////////////////////
// CPagerankDlg dialog


int CALLBACK ListCompare(LPARAM lParam1, LPARAM lParam2, LPARAM lParamSort)
{
	CListCtrl * pV=(CListCtrl*)lParamSort;

	CString szComp1,szComp2;
	szComp1 = pV->GetItemText(lParam1,3);
	szComp2 = pV->GetItemText(lParam2,3);
//	int iCompRes;
	double dComp1,dComp2;
	dComp1 = atof(szComp1);
	dComp2 = atof(szComp2);
	double result = dComp1 - dComp2;
	double temp = 0.0000001f;

	
	return dComp1>dComp2? 1 : -1;
	
}
CPagerankDlg::CPagerankDlg(CWnd* pParent /*=NULL*/)
	: CDialog(CPagerankDlg::IDD, pParent)
{
	//{{AFX_DATA_INIT(CPagerankDlg)
	m_Key = _T("");
	m_prNum = 10;
	m_prInit = 1.0f;
	m_xs = 0.5f;
	//}}AFX_DATA_INIT
	// Note that LoadIcon does not require a subsequent DestroyIcon in Win32
	m_hIcon = AfxGetApp()->LoadIcon(IDR_MAINFRAME);
}

void CPagerankDlg::DoDataExchange(CDataExchange* pDX)
{
	CDialog::DoDataExchange(pDX);
	//{{AFX_DATA_MAP(CPagerankDlg)
	DDX_Control(pDX, IDC_RECORD_LIST, m_Record_List);
	DDX_Text(pDX, IDC_KEY_EDIT, m_Key);
	DDX_Text(pDX, IDC_EDIT_NUM, m_prNum);
	DDV_MinMaxInt(pDX, m_prNum, 5, 100);
	DDX_Text(pDX, IDC_EDIT_INIT, m_prInit);
	DDV_MinMaxFloat(pDX, m_prInit, 0.f, 1.f);
	DDX_Text(pDX, IDC_EDIT_XS, m_xs);
	DDV_MinMaxFloat(pDX, m_xs, 0.f, 1.f);
	//}}AFX_DATA_MAP
}

BEGIN_MESSAGE_MAP(CPagerankDlg, CDialog)
	//{{AFX_MSG_MAP(CPagerankDlg)
	ON_WM_SYSCOMMAND()
	ON_WM_PAINT()
	ON_WM_QUERYDRAGICON()
	ON_BN_CLICKED(IDC_SEARCH_BUTTON, OnSearchButton)
	ON_BN_CLICKED(IDC_PAGERANK_BUTTON, OnPagerankButton)
	ON_BN_CLICKED(IDC_SORT_BUTTON, OnSortButton)
	ON_BN_CLICKED(IDC_CLEAR_BUTTON, OnClearButton)
	ON_NOTIFY(LVN_COLUMNCLICK, IDC_RECORD_LIST, OnColumnclickRecordList)
	//}}AFX_MSG_MAP
END_MESSAGE_MAP()

/////////////////////////////////////////////////////////////////////////////
// CPagerankDlg message handlers

BOOL CPagerankDlg::OnInitDialog()
{
	CDialog::OnInitDialog();

	// Add "About..." menu item to system menu.

	// IDM_ABOUTBOX must be in the system command range.
	ASSERT((IDM_ABOUTBOX & 0xFFF0) == IDM_ABOUTBOX);
	ASSERT(IDM_ABOUTBOX < 0xF000);

	CMenu* pSysMenu = GetSystemMenu(FALSE);
	if (pSysMenu != NULL)
	{
		CString strAboutMenu;
		strAboutMenu.LoadString(IDS_ABOUTBOX);
		if (!strAboutMenu.IsEmpty())
		{
			pSysMenu->AppendMenu(MF_SEPARATOR);
			pSysMenu->AppendMenu(MF_STRING, IDM_ABOUTBOX, strAboutMenu);
		}
	}

	// Set the icon for this dialog.  The framework does this automatically
	//  when the application's main window is not a dialog
	SetIcon(m_hIcon, TRUE);			// Set big icon
	SetIcon(m_hIcon, FALSE);		// Set small icon
	
	// TODO: Add extra initialization here
	m_Record_List.SetExtendedStyle(LVS_EX_FULLROWSELECT|LVS_EX_GRIDLINES);
	LV_COLUMN h;
	h.mask=LVCF_FMT|LVCF_TEXT|LVCF_WIDTH;
	h.fmt=LVCFMT_LEFT;
	h.cx=50;
	h.pszText="ID";
	m_Record_List.InsertColumn(0,&h);
	h.cx=195;
	h.pszText="WebPage";
	m_Record_List.InsertColumn(1,&h);
	h.cx=80;
	h.pszText="Links";
	m_Record_List.InsertColumn(2,&h);
	h.cx=80;
	h.pszText="pr value";
	m_Record_List.InsertColumn(3,&h);



    ::CoInitialize(NULL);
	m_pCon.CreateInstance(__uuidof(Connection));
	try{
		m_pCon->Open("Provider=Microsoft.Jet.OLEDB.4.0;\
			Data Source=search.mdb","","",NULL);
	}catch(_com_error e)
	{
		AfxMessageBox("数据库连接失败，确认数据库search.mdb是否在当前路径下!");
		exit(0);
		
	}
	return TRUE;  // return TRUE  unless you set the focus to a control
}

void CPagerankDlg::OnSysCommand(UINT nID, LPARAM lParam)
{
	if ((nID & 0xFFF0) == IDM_ABOUTBOX)
	{
		CAboutDlg dlgAbout;
		dlgAbout.DoModal();
	}
	else
	{
		CDialog::OnSysCommand(nID, lParam);
	}
}

// If you add a minimize button to your dialog, you will need the code below
//  to draw the icon.  For MFC applications using the document/view model,
//  this is automatically done for you by the framework.

void CPagerankDlg::OnPaint() 
{
	if (IsIconic())
	{
		CPaintDC dc(this); // device context for painting

		SendMessage(WM_ICONERASEBKGND, (WPARAM) dc.GetSafeHdc(), 0);

		// Center icon in client rectangle
		int cxIcon = GetSystemMetrics(SM_CXICON);
		int cyIcon = GetSystemMetrics(SM_CYICON);
		CRect rect;
		GetClientRect(&rect);
		int x = (rect.Width() - cxIcon + 1) / 2;
		int y = (rect.Height() - cyIcon + 1) / 2;

		// Draw the icon
		dc.DrawIcon(x, y, m_hIcon);
	}
	else
	{
		CDialog::OnPaint();
	}
}

// The system calls this to obtain the cursor to display while the user drags
//  the minimized window.
HCURSOR CPagerankDlg::OnQueryDragIcon()
{
	return (HCURSOR) m_hIcon;
}

void CPagerankDlg::OnSearchButton() 
{
	// TODO: Add your control notification handler code here
	variant_t vFieldValue;
	CString strFieldValue;
	_RecordsetPtr m_pRs;
	m_pRs.CreateInstance(__uuidof(Recordset));
	while(m_Record_List.GetItemCount()>0)
	{
		m_Record_List.DeleteItem(0);
	}
	
    UpdateData(true);
	bool search=false;
	CString strSql;
	strSql.Format("%s",m_Key);
    strSql="select * from pagerank where WebPage like '%"+strSql+"%'";
	BSTR bsSql=strSql.AllocSysString();
	try{
		
		m_pRs->Open((CComVariant)bsSql,m_pCon.GetInterfacePtr(),adOpenDynamic,adLockOptimistic,adCmdText);
        if(!m_pRs->FirstOfFiles)
			m_pRs->MoveFirst();
		else
		{
			AfxMessageBox("没有找到满足条件的网页");
			return;
		}
		int i=0;
		while(VARIANT_FALSE==m_pRs->EndOfFile)
		{
			
			_variant_t vFieldValue;
			CString strID,strWebPage,strPR,strLinks;		
			vFieldValue=m_pRs->GetCollect("id");
			strID=(char *)_bstr_t(vFieldValue);
			m_Record_List.InsertItem(i,strID);
			m_Record_List.SetItemData(i,i);
			vFieldValue=m_pRs->GetCollect("webpage");
			strWebPage=(char *)_bstr_t(vFieldValue);
			m_Record_List.SetItemText(i,1,strWebPage);
			
			vFieldValue=m_pRs->GetCollect("links");
			if(vFieldValue.vt != VT_NULL){
			strLinks=(char *)_bstr_t(vFieldValue);
			m_Record_List.SetItemText(i,2,strLinks);
			}

			i++;
			m_pRs->MoveNext();
			
		}
		GetDlgItem(IDC_CLEAR_BUTTON)->EnableWindow();
		GetDlgItem(IDC_PAGERANK_BUTTON)->EnableWindow();
		GetDlgItem(IDC_SEARCH_BUTTON)->EnableWindow();
		GetDlgItem(IDC_SORT_BUTTON)->EnableWindow(FALSE);
		m_pRs->Close();
		m_pRs = NULL;
		//添加
		this->pagerank_value();
		this->sort_value();

	}catch(_com_error * e){
		AfxMessageBox(e->ErrorMessage());
		if(m_pRs != NULL && m_pRs->State){
			m_pRs->Close();
			m_pRs = NULL;
		}
		return;
	}
	
}

void CPagerankDlg::OnPagerankButton() 
{
	// TODO: Add your control notification handler code here
	int nCount = m_Record_List.GetItemCount();
	int ** relationArray;
	relationArray = new int * [nCount];
	for(int i = 0; i < nCount; i++){
		relationArray[i] = new int [nCount];
	}
	for(i=0; i < nCount; i++){
		CString links;
		links = this->m_Record_List.GetItemText(i,2);
		for(int j = 0; j < nCount; j++){
			CString id;
			id = this->m_Record_List.GetItemText(j,0);
			int index = links.FindOneOf(id);
			if(index != -1) relationArray[i][j] = 1;
			else relationArray[i][j] = 0;

		}
	}
	CString strout;
	for(i = 0;i < nCount; i++){
		for(int j=0; j<nCount;j++){
			if(relationArray[i][j] == 0){
				strout += "0 ";
			}else{
				strout += "1 ";
			}
		}
		strout+="\n";
	}

	MessageBox(strout,"网页间的关系矩阵");
	UpdateData();
	AdjacencyWDigraph<int> graph(nCount, 0, relationArray, m_prNum, m_prInit ,m_xs);
	float *pr = graph.calculate();
	CString prOut;
	for( i = 0;i < nCount; i++){
		CString temp;
		CString p;
		p.Format("P%s=",m_Record_List.GetItemText(i,0));
		temp.Format("%f\n",pr[i]);
		prOut += p + temp;
	}
  	GetDlgItem(IDC_SORT_BUTTON)->EnableWindow();
 	MessageBox(prOut,"PageRank value");
	for( i = 0;i < nCount; i++){
		CString temp;
		temp.Format("%f",pr[i]);
		m_Record_List.SetItemText(i,3,temp);
	}
	SetDlgItemText(IDC_SORT_BUTTON,"网页排序");
	delete [] pr;
	



	//afxDump << links;


}

void CPagerankDlg::OnSortButton() 
{
	// TODO: Add your control notification handler code here
	_variant_t vFieldValue;
	CString strFieldValue;
	_RecordsetPtr m_pRs;	 
	m_Record_List.SortItems( ListCompare, (DWORD)&m_Record_List );
	CString capture;
	GetDlgItemText(IDC_SORT_BUTTON,capture);
	if(capture == "网页排序"){
		capture = "还原";
	}else{
		capture = "网页排序";
	}
	SetDlgItemText(IDC_SORT_BUTTON,capture);
	
}

void CPagerankDlg::OnClearButton() 
{
	// TODO: Add your control notification handler code here
   while(m_Record_List.GetItemCount()>0)
   {
	   m_Record_List.DeleteItem(0);
   }
   GetDlgItem(IDC_CLEAR_BUTTON)->EnableWindow(FALSE);
   GetDlgItem(IDC_PAGERANK_BUTTON)->EnableWindow(FALSE);
   GetDlgItem(IDC_SORT_BUTTON)->EnableWindow(FALSE);
   UpdateData(TRUE);
}

/* 
* 类似java字符串的split()方法。
* 使用一个完整的串delimit（而不是其中的某个字符）来分割src串,没有trim选项，
* 即严格分割。num用来确定最多分割为多少个串,如果是0（默认），则按照delimit
* 分割，若为1，则返回源串。
*/

void CPagerankDlg::Split(const CString& src, CString delimit, 
		CStringList* pOutList, int num, CString nullSubst){
	ASSERT( !src.IsEmpty() && !delimit.IsEmpty() );
	if(num==1) 
	{
		pOutList->AddTail(src);
		return;
	}
	
	int deliLen = delimit.GetLength();
	long index = -1, lastSearchPosition = 0, cnt = 0;
	
	while( (index=src.Find(delimit, lastSearchPosition))!=-1 )
	{
		if(index==lastSearchPosition)
			pOutList->AddTail(nullSubst);
		else
			pOutList->AddTail(src.Mid(lastSearchPosition, index-lastSearchPosition));
		lastSearchPosition = index + deliLen;
		
		if(num)
		{
			++cnt;
			if(cnt+1==num) break;
		}
	}
	CString lastOne = src.Mid(lastSearchPosition);
	pOutList->AddTail( lastOne.IsEmpty()? nullSubst:lastOne);
}


void CPagerankDlg::OnColumnclickRecordList(NMHDR* pNMHDR, LRESULT* pResult) 
{
	NM_LISTVIEW* pNMListView = (NM_LISTVIEW*)pNMHDR;
	// TODO: Add your control notification handler code here
	m_Record_List.SortItems( ListCompare, (DWORD)&m_Record_List );
	*pResult = 0;
}

void CPagerankDlg::pagerank_value()
{
	int nCount = m_Record_List.GetItemCount();
	int ** relationArray;
	relationArray = new int * [nCount];
	for(int i = 0; i < nCount; i++){
		relationArray[i] = new int [nCount];
	}
	//计算链接begin
	for(i=0; i < nCount; i++){
		CString links;
		links = this->m_Record_List.GetItemText(i,2);
		for(int j = 0; j < nCount; j++){
			CString id;
			id = this->m_Record_List.GetItemText(j,0);
			int index = links.FindOneOf(id);
			if(index != -1) relationArray[i][j] = 1;
			else relationArray[i][j] = 0;

		}
	}
	CString strout;
	for(i = 0;i < nCount; i++){
		for(int j=0; j<nCount;j++){
			if(relationArray[i][j] == 0){
				strout += "0 ";
			}else{
				strout += "1 ";
			}
		}
		strout+="\n";
	}
	//计算链接end

	
	UpdateData();
	AdjacencyWDigraph<int> graph(nCount, 0, relationArray, m_prNum, m_prInit,m_xs);
	float *pr = graph.calculate();
	
	CString prOut;
	for( i = 0;i < nCount; i++){
		CString temp;
		CString p;
		p.Format("P%s=",m_Record_List.GetItemText(i,0));
		temp.Format("%f\n",pr[i]);
		prOut += p + temp;
	}
  	GetDlgItem(IDC_SORT_BUTTON)->EnableWindow();
 
	
	for( i = 0;i < nCount; i++){
		CString temp;
		temp.Format("%f",pr[i]);
		m_Record_List.SetItemText(i,3,temp);
	}

	
	delete [] pr;
}

void CPagerankDlg::sort_value()
{
	_variant_t vFieldValue;
	CString strFieldValue;
	_RecordsetPtr m_pRs;	 
	m_Record_List.SortItems( ListCompare, (DWORD)&m_Record_List );
}