一个比较复杂的awk比对问题

殇19921011

比对文本
{7106561412,50297001,6075261,571,65000326193972,4740,20140601000000,20140701000000:0,360000}
{7106561412,50297001,6075271,571,65000326193972,4740,20140601000000,20140701000000:0,360000}

{7103210808,50297001,6075271,571,65000242312608,960,20140601000000,20140701000000:0,360000}
{7103210808,50297001,6075271,571,65000242312608,960,20140601000000,20140701000000:1140,360000}

{7106561412,50297001,6075271,571,65000326193972,9000,20140601000000,20140701000000:0,360000}
{7106561412,50297001,6075271,571,65000326193972,9000,20140601000000,20140701000000:4740,360000}

{7103210808,50297001,6075271,571,65000242312608,480,20140601000000,20140701000000:1080,360000}
{7103210808,50297001,6075271,571,65000242312608,480,20140601000000,20140701000000:2220,360000}

{7103210808,50297001,6075271,571,65000242312608,120,20140601000000,20140701000000:960,360000}
{7103210808,50297001,6075271,571,65000242312608,120,20140601000000,20140701000000:2100,360000}

以空格为分界线比对相邻两行数据，每行数据以“，”为分隔符，屏蔽掉第八个域，即除了第八个域。如果两行数据不同，打印出这相邻的两行数据。文本中的数据有多条记录

求大神解答。在线等

yestreenstars

是怎样比对？
1和2比对，然后2和3比对……
1和2比对，然后3和4比对……

ly5066113

回复 1# 殇19921011


try:

1. awk -v RS= 'gensub(/,[^,]*/,"",7,$1)!=gensub(/,[^,]*/,"",7,$2)' file

复制代码

q1208c

好奇怪的需求, 我宁愿写个perl程序来处理, 免得以后自己也不知道要怎么处理了.

Kasiotao

回复 1# 殇19921011

1. awk -vi=0 -F\, '$0==NULL{i=0;if(a[0]!=a[1]) print b[0]"\n"b[1]}$0!=NULL{a[i]=substr($1 $2 $3 $4 $5 $6 $7,2);b[i]=$0;i++}' testfile  

复制代码

我写好长。。。肯定不是最好的
   

殇19921011

1和2对比，然后3和4对比，以空行为分隔回复 2# yestreenstars


   

lifayi2008

1. awk 'BEGIN{FS=","}{a[NR]=$0;b[NR]=$1$2$3$4$5$6$7$9;if(NR%3==2 && b[NR]!=b[NR-1])print a[NR-1]"\n"a[NR]}' c.txt

复制代码

yestreenstars

回复 4# q1208c

你是来宣传perl的~
   

Kasiotao

回复 6# 殇19921011
ly的方法很精辟。。你可以学习下。。以空行做RS，然后用gensub比较，这样也不影响原字符串，碉堡

   

殇19921011

我的机器不能用gensub啊 awk: Function gensub is not defined.回复 9# Kasiotao