第3讲：导入表的定位和读取操作

标签：

// ImportFun.cpp : 完整代码如下
//

#include "stdafx.h"
#include <stdio.h>
#include <windows.h>

//用来实现RVA到磁盘文件偏移的转换
DWORD RVAToOffset(LPVOID lpBase,DWORD VirtualAddress)
{
IMAGE_DOS_HEADER *dosHeader;
IMAGE_NT_HEADERS *ntHeader;
IMAGE_SECTION_HEADER *SectionHeader;
int NumOfSections;//Section 的数量
//定位到PE head
dosHeader=(IMAGE_DOS_HEADER*)lpBase;
ntHeader=(IMAGE_NT_HEADERS*)((BYTE*)lpBase+dosHeader->e_lfanew);
NumOfSections=ntHeader->FileHeader.NumberOfSections;
for (int i=0;i<NumOfSections;i++)
{
SectionHeader=(IMAGE_SECTION_HEADER*)((BYTE*)lpBase+dosHeader->e_lfanew+sizeof(IMAGE_NT_HEADERS))+i;
if(VirtualAddress>SectionHeader->VirtualAddress&&VirtualAddress<SectionHeader->VirtualAddress+SectionHeader->SizeOfRawData)
{
DWORD AposRAV=VirtualAddress-SectionHeader->VirtualAddress;
DWORD Offset=SectionHeader->PointerToRawData+AposRAV;
return Offset;
}
}
return 0;
}

int main(int argc, char* argv[])
{
//打开文件
HANDLE hFile=CreateFile("c:\\calc.exe",GENERIC_READ,FILE_SHARE_READ,NULL,OPEN_EXISTING,FILE_ATTRIBUTE_NORMAL,NULL);
if(hFile==INVALID_HANDLE_VALUE)
{
printf("CreateFile Failed\n");
return 0;
}
//创建内存映射文件的内核对象
HANDLE hMap=CreateFileMapping(hFile,NULL,PAGE_READONLY,NULL,NULL,NULL);
if(hMap==INVALID_HANDLE_VALUE)
{
printf("CreateFileMapping Failed\n");
return 0;
}
//把文件映射入内存
LPVOID lpBase=MapViewOfFile(hMap,FILE_MAP_READ,0,0,0);//返回内存文件映射句柄，lpBase
if(lpBase==NULL)
{
printf("MapViewOfFile Failed\n");
return 0;
}

IMAGE_DOS_HEADER *dosHeader;
IMAGE_NT_HEADERS *ntHeader;
IMAGE_IMPORT_BY_NAME *ImportName;
//lpBase由MapViewOfFile函数返回
dosHeader=(IMAGE_DOS_HEADER*)lpBase;
//检测是否是有效的PE文件
if (dosHeader->e_magic!=IMAGE_DOS_SIGNATURE)
{
printf("This is not a windows file\n");
return 0;
}
//定位到PE header
ntHeader=(IMAGE_NT_HEADERS*)((BYTE*)lpBase+dosHeader->e_lfanew);//e_lfanew成员定位到PE header
//判断是否是一个有效的win32文件
if(ntHeader->Signature!=IMAGE_NT_SIGNATURE)
{
printf("This is not a win32 file\n");
return 0;
}
//定位到导入表
IMAGE_IMPORT_DESCRIPTOR *ImportDec=(IMAGE_IMPORT_DESCRIPTOR*)((BYTE*)lpBase+RVAToOffset(lpBase,ntHeader->OptionalHeader.DataDirectory[IMAGE_DIRECTORY_ENTRY_IMPORT].VirtualAddress));
while(ImportDec->FirstThunk)
{
//得到DLL文件名
char *pDllName=(char*)((BYTE*)lpBase+RVAToOffset(lpBase,ImportDec->Name));
printf("\nDLL文件名：%s\n",pDllName);
//通过OriginalFirstThunk定位到PIMAGE_THUNK_DATA结构数组
PIMAGE_THUNK_DATA pThunk=(PIMAGE_THUNK_DATA)((BYTE*)lpBase+RVAToOffset(lpBase,ImportDec->OriginalFirstThunk));
while(pThunk->u1.Function)
{
//判断函数是用函数名导入的还是序号导入的
if(pThunk->u1.Ordinal& IMAGE_ORDINAL_FLAG32)//高位为1
{
//输出序号
printf("从此DLL模块导出的函数的序号：%x\n",pThunk->u1.Ordinal&0xFFFF);
}
else//高位为0
{
//得到IMAGE_IMPORT_BY_NAME结构中的函数名
ImportName=(IMAGE_IMPORT_BY_NAME*)((BYTE*)lpBase+RVAToOffset(lpBase,(DWORD)pThunk->u1.AddressOfData));
printf("从此DLL模块导出的函数的函数名：%s\n",ImportName->Name);
}
pThunk++;
}
ImportDec++;
}
UnmapViewOfFile(lpBase);
CloseHandle(hMap);
CloseHandle(hFile);
return 0;
}

第3讲：导入表的定位和读取操作

标签：

原文地址：http://www.cnblogs.com/mayingkun/p/4769544.html