unix中统计、去除文件中的重复行（UNIX statistics, removal of duplicate rows in the file）.doc

unix中统计、去除文件中的重复行（UNIX statistics, removal of duplicate rows in the file） Unix statistics, removal of duplicate rows in the file In a database, if the table records are repeated, you can only use the distinct keyword to remove duplicate rows. So, is there a convenient tool for implementing this requirement in the Unix operating system? The answer is yes, of course. To achieve this requirement, we need to use the sort sort command and uniq to remove duplicate row commands. As we all know, in Unix system maintenance, system engineers often need to merge multiple files into one file. A problem arises when multiple files are joined or merged together, and duplicate records can be generated. This is what system engineers do not want to see. So how do you eliminate these duplicate records? The reader familiar with the sort command must know that hitting an -u option with the sort sort can achieve the purpose of deleting duplicate rows. But this function is not very strong. If you can not help the system engineer find out which rows are repeated or count the number of duplicate rows and so forth. To implement these complex functions, you need to use this uniq to remove duplicate row commands. The first step is to sort the records in the file first. Now, when the system engineer combines the two files, the following file is generated, log.txt. The sales department has a system backup fault The purchasing department has a system backup fault The sales department has a system backup fault The financial department has a system backup fault This is a hint when the system backup program is out of order. For ease of administration, system engineers often merge the error records of one week or half a month in the same file. Then, by removing the duplicated records, you can find out what parts are backed up. If you do not remove the duplicate line, then one check will be very troublesome. Look at a list of wrong lists (many of them are repeated) and dont know how to do it. In addit