求对txt文件进行删除行操作代码

goodlardy 2010-6-2 12:23 楼主

本人想把一个几万行的txt文件内的，某些行删除，手工太慢。想把带有某些字符（比如一个单词）的行全部删除。操作以后保存退出。

请问用到什么方法？

回复评论（20）

沙发 jianglongbiao

循环读取文件内容，每次读1行或10行，行数随便，然后字符串查找，找到了就删除，把剩下的追加到临时文件里，最后替换原来的文件。

点赞 2010-6-2 12:34

板凳 longtoon

用脚本语言处理吧，两三行代码搞定。
比如：perl

点赞 2010-6-2 12:35

4楼 liulirong1985

fgets()一行一行读文件，一行一行处理，再把处理结构写入新文件。
完成后，关闭新，老文件。删除老文件，最后把新文件改名为老文件。

点赞 2010-6-2 12:36

5楼 yaoyao11

几万行？
文件多大？
可以考虑用内存映射文件操作

点赞 2010-6-2 12:51

6楼麦兜仔仔

把分给我吧，正好写过一个过滤的应用

#include "stdafx.h"
#include
#include
#pragma warning(disable:4244 4267)
////////////////////////////////////////////////////////////////////////////////////////////////////////
template struct Str_LookupA; // ANSI字符集字符串查找模板
template struct Str_LookupW; // UNICODE 字符集字串查找模板
template struct Str_Filter_Word; // 字符串过滤模板框架 -- 只过滤输入单词
template struct Str_Filter_Line; // 字符串过滤模板框架 -- 过滤包含输入单词段落
////////////////////////////////////////////////////////////////////////////////////////////////////////
//
// 模板参数 t_fMatch, 指明查找字串是否区分大小写。
// t_fMatch = true，查找函数将不区分大小写
// t_fMatch = false, 默认方式，按照输入查找字串进行匹配
//
template struct Str_LookupA
{
enum { TYPE_SIZE = sizeof(CHAR)};
typedef CHAR* _Type;
typedef const CHAR* _constType;
typedef _Type _ReturnType;
_ReturnType operator()(_Type pSrc,_constType pSearch)
{
return t_fMatch ? StrStrIA(pSrc,pSearch) : StrStrA(pSrc,pSearch);
}
_ReturnType LineStart(_Type pSrc,_Type pLast)
{
_Type pStart = StrRStrIA(pSrc,pLast,"\r\n");
return pStart ? pStart + 2 : pSrc;
}
_ReturnType LineEnd(_Type pSrc)
{
_Type pEnd = StrStrA(pSrc,"\r\n");
return pEnd ? pEnd + 2 : NULL;
}
DWORD Length(_constType pStr)
{
return lstrlenA(pStr);
}
};
template struct Str_LookupW
{
enum { TYPE_SIZE = sizeof(WCHAR)};
typedef WCHAR* _Type;
typedef _Type _ReturnType;
typedef const WCHAR* _constType;
_ReturnType operator()(_Type pSrc, _constType pSearch)
{
return t_fMatch ? StrStrIW(pSrc,pSearch) : StrStrW(pSrc,pSearch);
}
_ReturnType LineStart(_Type pSrc,_Type pLast)
{
_Type pStart = StrRStrIW(pSrc,pLast,L"\r\n");
return pStart ? pStart + 2 : pSrc;
}
_ReturnType LineEnd(_Type pSrc)
{
_Type pEnd = StrStrW(pSrc,L"\r\n");
return pEnd ? pEnd + 2 : NULL;
}
DWORD Length(_constType pStr)
{
return lstrlenW(pStr);
}
};
template >
struct Str_Filter_Word
{
typedef typename _RT::_Type _Type;
typedef typename _RT::_constType _constType;
typedef typename _RT::_ReturnType _ReturnType;
_ReturnType operator()(_Type pBuffer, DWORD cbSize, _constType *pWord, DWORD cbCount, DWORD* pcbRemoved = NULL)
{
ATLASSERT(pBuffer && pWord && cbSize && cbCount);
_RT Str_Lookup;
DWORD dwRemoved = 0;
DWORD cbStart = 0;
DWORD cbEnd = cbSize;
for (DWORD i = 0; i < cbCount; i++)
{
_Type pStart = pBuffer;
_Type pEnd = pBuffer + cbEnd;
_constType pSearch = pWord[i];
DWORD cbLength = Str_Lookup.Length(pSearch);
if (!pSearch)
break;
for (;;)
{
_Type p = Str_Lookup(pStart,pSearch);
if (!p) break;
dwRemoved++;
_Type pNext = p + cbLength;
if (pNext && pNext <=pEnd)
{
cbStart = p - pBuffer;
memmove(p,pNext,_RT::TYPE_SIZE * (cbEnd - cbStart));
pStart = p;
cbEnd -= cbLength;
}
else
{
memset(pStart,0, _RT::TYPE_SIZE * (cbEnd - cbStart));
break;
}
}
}
if (pcbRemoved) *pcbRemoved = dwRemoved;
return pBuffer;
}
_ReturnType operator()(_Type pBuffer, DWORD cbSize, _constType pWord, DWORD* pcbRemoved = NULL)
{
_constType pWords[] = {pWord,0};
return operator()(pBuffer,cbSize,pWords,1,pcbRemoved);
}
};
template >
struct Str_Filter_Line
{
typedef typename _RT::_Type _Type;
typedef typename _RT::_constType _constType;
typedef typename _RT::_ReturnType _ReturnType;
_ReturnType operator()(_Type pBuffer, DWORD cbSize, _constType *pWord, DWORD cbCount, DWORD* pcbRemoved = NULL)
{
ATLASSERT(pBuffer && pWord && cbSize && cbCount);
_RT Str_Lookup;
DWORD dwRemoved = 0;
DWORD cbStart = 0;
DWORD cbEnd = cbSize;
for (DWORD i = 0; i < cbCount; i++)
{
_Type pStart = pBuffer;
_Type pEnd = pBuffer + cbEnd;
_constType pSearch = pWord[i];
if (!pSearch) break;
for (;;)
{
_Type p = Str_Lookup(pStart,pSearch);
if (!p) break;
dwRemoved++;
pStart = Str_Lookup.LineStart(pBuffer,p);
_Type pLineEnd = Str_Lookup.LineEnd(p);
if (pLineEnd)
{
cbStart = pStart - pBuffer;
memmove(pStart,pLineEnd,_RT::TYPE_SIZE * (cbEnd - cbStart));
cbEnd -= (pLineEnd - pStart);
}
else
{
memset(pStart,0, _RT::TYPE_SIZE * (cbEnd - cbStart));
break;
}
}
}
if (pcbRemoved) *pcbRemoved = dwRemoved;
return pBuffer;
}
_ReturnType operator()(_Type pBuffer, DWORD cbSize, _constType pWord, DWORD* pcbRemoved = NULL)
{
_constType pWords[] = {pWord,0};
return operator()(pBuffer,cbSize,pWords,1,pcbRemoved);
}
};
HRESULT test_file_filter(LPCTSTR lpszFile, LPCSTR *pWords, DWORD cbWords, bool fLine = false)
{
ATL::CAtlFileMapping map;
ATL::CAtlFile file;
HRESULT hr;
ULONGLONG uSize = 0;
if (FAILED(hr = file.Create(lpszFile,GENERIC_READ | GENERIC_WRITE,0,OPEN_EXISTING)))
return hr;
if (FAILED(hr = file.GetSize(uSize)))
return hr;
size_t nSize = (size_t)uSize;
if (FAILED(hr = map.MapFile(file.m_h,nSize,0,PAGE_READWRITE,FILE_MAP_READ | FILE_MAP_WRITE)))
return hr;
LPSTR pData = (LPSTR)map.GetData();
DWORD cbRemoved = 0;
if (fLine)
{
// 区分大小写，过滤包含指定单词的段落
Str_Filter_Line<> filter;
filter(pData,nSize,pWords,cbWords,&cbRemoved);
}
else
{
// 以不区分大小写的模式，过滤指定单词
Str_Filter_Word filter;
filter(pData,nSize,pWords,cbWords,&cbRemoved);
}
printf(pData);
return S_OK;
}
int _tmain(int argc, _TCHAR* argv[])
{
// 过滤指定文件中多个单词
LPCSTR pWords[] = {"Sec","Time","北国","sample","for",0};
HRESULT hr = test_file_filter(_T("c:\\100.txt"),pWords,5);
if (FAILED(hr))
printf("failed! hr = 0x08X\n",hr);
// 过滤文件中包含指定单词的段落
hr = test_file_filter(_T("c:\\300.txt"),pWords,5,true);
if (FAILED(hr))
printf("failed! hr = 0x08X\n",hr);
_getch();
return 0;
}