这样的数据如何筛选？

jiaoli

源文件如下：
37123456,123456789,601234020200051640,"孔霞","03",123456789,"2008/01/06",1,4000,5060.41
37123456,123456789,601234020200062521,"栗汝礼","03",123456010,"2008/01/06",1,100,110.91
37123456,123456789,601234020200069800,"柯纯龙","03",370786017,"2008/01/06",1,20000,19500
37123456,123456789,601234020200069800,"柯纯龙","03",123456030,"2008/01/06",1,31000,500
37123456,123456789,601234020200068018,"严凤书","03",123456789,"2008/01/06",2,50000,100163.39
37123456,123456789,601234020200070039,"刘庆","03",123456789,"2008/01/06",2,4000,8000
37123456,123456789,601234020200060554,"王兰英","03",123456789,"2008/01/06",1,1600,91.26
37123456,123456789,601234020200070039,"刘庆","03",123456789,"2008/01/06",2,4000,4000
37123456,123456789,601234020200067710,"罗有艳","03",123456789,"2008/01/06",2,3000,12012.01
37123456,123456789,601234020200064742,"孙祥婷","03",123456789,"2008/01/06",1,50,12.61
37123456,123456789,601234020200069800,"柯纯龙","03",123456030,"2008/01/06",1,200,300
37123456,123456789,601234020200060554,"王兰英","03",123456789,"2008/01/06",1,1000,1091.26
处理要求如下：
如果第三列中的数据是唯一的，就保留这一行，如果有重复的，就保留最后一个记录行。

源文件格式如图：

[ 本帖最后由 jiaoli 于 2008-1-8 10:57 编辑 ]

jiaoli

格式乱了，我发一个源文件上来。

davistar

1. 
   
2. awk -F, '{a[$3]=$0}END{for( i in a)print a[i]}' urfile
   

复制代码

jiaoli

awk真是无所不能，成了！谢谢楼上朋友！

jinl

原帖由 davistar 于 2008-1-8 10:57 发表

awk -F, '{a[$3]=$0}END{for( i in a)print a}' urfile

awk 数组的经典用法，很喜欢awk的数组下标可以任意组合

guocb

8错