如何实现文件搜索 [英] How to implement a file search

问题描述

我想编写一个程序来读取文件并在其中进行二进制搜索.我不知道此代码是否正确，请检查一下:

I want to write a program to read a file and binary search in it. I don''t know if this code is correct or not, please check it:

//size of record = 100 b
//size of blonck = 4 kb
struct record{
    char name [26] ;
    char family[30] ;
    char stu_no [20] ;
    char mobil [20] ;
    record * pnext ;
};

record search ( record pdata ) 
{
	char item [30];
	char x ;
	fstream file ;
	file.open ( "file2.dat" ) ;
	int med ;
	int low = 0 ;
//counterb is number of blonck
	int hight = counterb - 1 ;
	while ( low <=hight )
	{
		med = ( low + hight )/2 ;
		file.seekp (4096*med , ios::beg ) ;
		for ( int i = 0 ; i < 30 ; i++ )
		{
			file.read ( &x , 1 ) ;
			item[i] = x ;
		}
		for ( int i = 0 ; i < 30 ; i++ )
		{
			int y = strcmp ( item , pdata.family  ) ;
			if ( y == 0 )
			{
				file >> pdata.family >> pdata .name >> pdata .mobil >> pdata .stu_no ;	 
				return pdata ;
			}
			if ( y < 0 )
			{
				file.seekp (-4096 , ios::beg ) ;
				for ( int i = 0 ; i < 30 ; i++ )
				{
					file.read ( &x , 1 ) ;
					item[i] = x ;
				}
				y = strcmp ( item , pdata.family  ) ;
				med = (low + med -1)/2 ;
			}
			if ( y > 0 )
			{
				file.seekp (counterr*100 , ios::beg ) ;
				for ( int i = 0 ; i < 30 ; i++ )
				{
					file.read ( &x , 1 ) ;
					item[i] = x ;
				}
				y = strcmp ( item , pdata.family  ) ;
				med = (hight + med +1)/2 ;
			}
		}
	}
}

推荐答案

请添加一些注释以了解您的逻辑，
我认为二进制搜索逻辑是正确的(假设记录按family的顺序存储).

我认为读取文件时可能会出现一些错误，最好读取写入记录块本身，否则将无法获得预期的结果.

Please add some comments to understand your logic,
I think binary search logic is correct( assuem that records are stored in the order of family ).

I think there may be some error in reading file, it is better to read write the record block itself otherwise it will not get expected result.

这篇关于如何实现文件搜索的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！