能給出一個(gè)計(jì)算時(shí)間差和匹配某個(gè)字段的shell嗎？

lqiao

一個(gè)文件如test.txt，格式如下：
a,b,c,20051201180903,20051201180912,d,e,21/122,21
a,b,c,20051201180903,20051201180912,d,e,f,21/121,21,22
a,b,c,20051201180903,20051201180912,d,e,f,g,21/121,21,22,23
其中第4列和第5列都是時(shí)間，格式為YYYYMMDDHHMMSS，需要提取第5列減第4列差為大于60s，并且包含21/121的列，請(qǐng)問這個(gè)shell該如何寫？

lqiao

如果第5列減去第4列為大于20s的，那如何判斷呢？謝謝！

大螞蟻

原帖由 lqiao 于 2005-12-20 11:36 發(fā)表
如果第5列減去第4列為大于20s的，那如何判斷呢？謝謝！

基礎(chǔ)12中的awk篇幅中應(yīng)該有講的

mocou

1. awk -F, '$0~/21\/122/{a=substr($4,1,4)" "substr($4,5,2)" "substr($4,7,2)" "substr($4,9,2)" "substr($4,11,2)" "substr($4,13,2);b=mktime(a);aa=substr($5,1,4)" "substr($5,5,2)" "substr($5,7,2)" "substr($5,9,2)" "substr($5,11,2)" "substr($5,13,2);bb=mktime(aa);cc=bb-b;if (cc>20)print}' file

復(fù)制代碼

誰給個(gè)簡(jiǎn)化的

寂寞烈火

思路簡(jiǎn)單,解決起來比較繁瑣

大螞蟻

原帖由 mocou 于 2005-12-20 14:05 發(fā)表

1. awk -F, '$0~/21\/122/{a=substr($4,1,4)" "substr($4,5,2)" "substr($4,7,2)" "substr($4,9,2)" "substr($4,11,2)" "substr($4,13,2);b=mktime(a);aa= ...

復(fù)制代碼

域分割符號(hào)怎么不取大點(diǎn)？
取

1. －F ",200512011"

復(fù)制代碼

如何？或者更長(zhǎng)點(diǎn)？

waker

1. gawk -F , '/21\/121/{a=$4;b=$5;gsub(/../,"& ",a);
   
2. gsub(/../,"& ",b);
   
3. sub(/ /,"",a);
   
4. sub(/ $/,"",a);
   
5. sub(/ /,"",b);
   
6. sub(/ $/,"",b);
   
7. if(mktime(b)-mktime(a)>60)print} ' file
   

復(fù)制代碼

如果你要求的差值都是xxs的話(注意，如果要求的差值大于小時(shí)的話，結(jié)果是什么俺沒想)

1. 
   
2. awk -F , '/21\/121/&&(substr($5,1,12)-substr($4,1,12))*60+(substr($5,13,2)-substr($4,13,2))>60 ' file
   

復(fù)制代碼

寂寞烈火

原帖由 mocou 于 2005-12-20 14:05 發(fā)表
[code]awk -F, '$0~/21\/122/{a=substr($4,1,4)" "substr($4,5,2)" "substr($4,7,2)" "substr($4,9,2)" "substr($4,11,2)" "substr($4,13,2);b=mktime(a);aa= ...

汗,,,,,, :em11:

寂寞烈火

/home/lee#cat txt
a,b,c,20051201180603,20051201180912,d,e,21/122,21
a,b,c,20051201180902,20051201180913,d,e,f,21/121,21,22
a,b,c,20051201180904,20051201180914,d,e,f,g,21/121,21,22,23

1. 
   
2. /home/lee#cat f
   
3. #!/bin/bash
   
4. ifs=$IFS
   
5. IFS=','
   
6. while read line;do
   
7. set -- $line
   
8. a=$(date +%s -d"${4:0:8} ${4:8:2}:${4:10:2}:${4:12:2}")
   
9. b=$(date +%s -d"${5:0:8} ${5:8:2}:${5:10:2}:${5:12:2}")
   
10. c=$((b-a))
    
11. echo $@ $c
    
12. done<txt|awk '$NF>60&&/21\/122/{gsub(/ /,",");sub(/,([0-9])+$/,"");print}'
    

復(fù)制代碼

result

/home/lee#sh f
a,b,c,20051201180603,20051201180912,d,e,21/122,21
/home/lee#