awk取字段 向大侠请教

ys547007

数据如下：
17:40:10:416315  [end task]:dispId:9,taskID:201,rowLen:100,shouldRowNum:6,actualRowNum:6,todayRowNum:1950,lastBeginTime:10-23_17:40:09:8482,lastEndTime:10-23_17:40:09:9722
17:40:11:428225  [end task]:dispId:6,taskID:241,rowLen:111,shouldRowNum:6,actualRowNum:6,todayRowNum:11700,lastBeginTime:10-23_17:40:09:6973,lastEndTime:10-23_17:40:09:8265
17:40:11:428361  [end task]:dispId:6,taskID:261,rowLen:111,shouldRowNum:14,actualRowNum:14,todayRowNum:27300,lastBeginTime:10-23_17:40:10:4717,lastEndTime:10-23_17:40:10:4978
17:40:16:483620  [end task]:dispId:6,taskID:241,rowLen:111,shouldRowNum:6,actualRowNum:6,todayRowNum:11706,lastBeginTime:10-23_17:40:14:5411,lastEndTime:10-23_17:40:14:5794
17:40:16:483740  [end task]:dispId:6,taskID:261,rowLen:111,shouldRowNum:14,actualRowNum:14,todayRowNum:27314,lastBeginTime:10-23_17:40:15:5334,lastEndTime:10-23_17:40:15:5813
17:40:16:483797  [end task]:dispId:7,taskID:181,rowLen:54,shouldRowNum:5,actualRowNum:5,todayRowNum:8130,lastBeginTime:10-23_17:40:15:6233,lastEndTime:10-23_17:40:15:7306

要取每行rowLen:后面的数字和todayRowNum：后的数字 请赐教！

li0924

1. awk -F"[,:]" 'NR%3==2{print $7,$13}' infile

复制代码

WilliBhamlll

1. awk -F, '{for(i=0;i++<NF;)if($i~"rowLen|todayRowNum"){split($i,a,":");printf a[2]" "}print ""}' file

复制代码

wang390750

awk -F "[ \t,:]+" '{if($11 ~/rowLen/ && $17~/todayRowNum/)print $12,$18}' urfile

linuxbrook

awk -F"[: ,]" '{print $12,$13,$18,$19}'  a.txt

reyleon

1. sed 's/.*rowLen:\([0-9]\+\).*todayRowNum:\([0-9]\+\).*/\1 \2/g' file

复制代码

trdcaz

1. $ cat file | awk -F, '{print $3,$6}'
   
2. rowLen:100 todayRowNum:1950
   
3. rowLen:111 todayRowNum:11700
   
4. rowLen:111 todayRowNum:27300
   
5. rowLen:111 todayRowNum:11706
   
6. rowLen:111 todayRowNum:27314
   
7. rowLen:54 todayRowNum:8130

复制代码

如果只要数字

1. $ cat file | awk -F "[,:]" '{print $10,$16}'
   
2. 100 1950
   
3. 111 11700
   
4. 111 27300
   
5. 111 11706
   
6. 111 27314
   
7. 54 8130

复制代码

yestreenstars

1. sed -r 's/.*rowLen:([0-9]+).*todayRowNum:([0-9]+).*/\1 \2/'

复制代码

LikeLx

print $几 的方法不行，源数据稍微一修改就打印错了，还是匹配的精准。

jason680

回复 1# ys547007

another way to get value

# awk 'function get_val(k){if(match($0,k":([^,]+)",a))return a[1]}{r=get_val("rowLen");t=get_val("todayRowNum");print "row="r", tod="t}' FILE
row=100, tod=1950
row=111, tod=11700
row=111, tod=27300