awk中RS问题求助

风一样的男子1

有文件：
cat time.txt

2008-01-08 17:33:26
2008-01-08 17:33:32
00:00:06
2008-01-08 17:34:12
2008-01-08 17:34:19
00:00:07
2008-01-08 16:34:07
2008-01-08 16:34:14
00:00:07

我想输出的结果：
start time: 2008-01-08 17:33:26  End time: 2008-01-08 17:33:32   elapsed time: 00:00:06
start time: 2008-01-08 17:34:12  End time: 2008-01-08 17:34:19   elapsed time: 00:00:07
start time: 2008-01-08 16:34:07  End time: 2008-01-08 16:34:14   elapsed time: 00:00:07

我用语句：
awk 'BEGIN{FS="\n";RS=""}{print "start time: " $1 "\t End time: " $2 "\t elapsed time: " $3}' time.txt

却只能输出想要的第一行记录：start time: 2008-01-08 17:33:26  End time: 2008-01-08 17:33:32   elapsed time: 00:00:06
后面两行未输出，请大家帮忙指点，谢谢了！

ly5066113

xargs -n5 < time.txt | awk '{print "start time:"$1,$2,"End time:"$3,$4,"elapsed time:"$5}'

davistar

通用性不强，给你个思路。gawk

1. 
   
2. awk 'BEGIN{RS="0+:[0-9]+:[0-9]+"}{if($1)printf"start time:%s endtime:%selapsed time:%s\n", $1,$2,RT}'  urfile
   

复制代码

jinl

RS="" 匹配一个或多个空行作为记录分割，按照你现在的代码，一共只有一条记录，你输出了前三个字段而已。建议你先在00:00那行后面插入回车，再送awk 处理
sed -e 's/^[0-9]{2}:[0-9]{2}:[0-9]$/&\n\n/' time.txt|awk 'BEGIN{FS="\n";RS=""}{print "start time: " $1 "\t End time: " $2 "\t elapsed time: " $3}'

mcolinc

原帖由 ly5066113 于 2008-3-10 11:14 发表
xargs -n5 < time.txt | awk '{print "start time:"$1,$2,"End time:"$3,$4,"elapsed time:"$5}'

这个很巧妙

风一样的男子1

原帖由 ly5066113 于 2008-3-10 11:14 发表
xargs -n5 < time.txt | awk '{print "start time:"$1,$2,"End time:"$3,$4,"elapsed time:"$5}'

这个的确很巧妙。
给xargs传参数，换行或者空格都可以作为参数分隔符的吗？

chenwenming

原帖由 ly5066113 于 2008-3-10 11:14 发表
xargs -n5 < time.txt | awk '{print "start time:"$1,$2,"End time:"$3,$4,"elapsed time:"$5}'

想不到xargs 这个命令这么有用，以前只看到find 与xargs配合用的．

但是xargs有个缺点．如果没有特别指定，xargs会把"_"当作结束标志。"_"以后参数就传不过去了．

[ 本帖最后由 chenwenming 于 2008-3-10 13:53 编辑 ]

风一样的男子1

1. awk 'BEGIN{FS="\n";RS=""}{
   
2. x=1
   
3.         while(x<=NF){
   
4.         print "start time: " $x "\t End time: " $(x+1) "\t elapsed time: " $(x+2)
   
5.         x+=3
   
6.         }
   
7. }' time.txt

复制代码

最后用这个笨一点的办法搞定了，
谢谢大家指点！