輸出匹配行后面的幾行

kitiz

有一個(gè)文件內(nèi)容：
log/TCP_128/RECEIVE
07071422

629.8336
985.7358
1176.7049
1161.5998
874.2844
1852.2079
2153.1002
2125.9526
log/TCP_128/SEND
07071422

742.2564
978.6179
1044.1491
1070.1761
1427.1565
1969.4728
2076.2248
2012.9897
log/TCP_2048/RECEIVE
07071422

938.2984
985.3628
1199.5154
1133.3772
1882.6661
1906.0648
2130.3019
2461.2140
log/TCP_2048/SEND
07071422

942.7927
980.1376
1062.7501
1054.5864
1885.1118
2447.5760
2388.5727
2210.1754

想輸出比如匹配SEND后面隔兩行的8行數(shù)據(jù)，求各位指教
log/TCP_2048/SEND
07071422

942.7927
980.1376
1062.7501
1054.5864
1885.1118
2447.5760
2388.5727
2210.1754




awk '{a[NR%10]=$0}/TCP.*SEND/{for(i=NR+2;i<NR+10;i++) print a[i%10]}' log.txt
這樣輸出的是匹配SEND前面的8行，不知道是什么原因
小白。。

liion631818

你這樣寫就是錯(cuò)的
數(shù)組a中存儲(chǔ)的就是匹配到send那一行的前10行，
匹配到send后就來(lái)打印數(shù)組a中的8行，當(dāng)然是send之前的8行

kitiz

sed -n "/TCP.*SEND/,+10p" log.txt |sed -n "/^$/,+8p"

blackold

回復(fù) 1# kitiz

try:

awk:

1. $ awk '/SEND/{n=8;i=0}/SEND/,n--==0{if(i++>2)print}' urfile

復(fù)制代碼

sed:

1. $ sed '/SEND/,+8! d;/SEND/,+2d' urfile

復(fù)制代碼

yinyuemi

sed -n '/SEND/,+10{/SEND/,+2d;p}'

huang_wunix

樓上各位高手， 關(guān)于sed 中 +8！  +10p  +2d
這些用法論壇里有相關(guān)的資料嗎？ 我想學(xué)習(xí)一下，麻煩給個(gè)鏈接給我
謝謝！

huang_wunix

我試過了，5樓提示
HP-UX:/hwtest/shell>sed -n "/TCP.*SEND/,+10p" log.txt | sed -n "/^$/,+8p" c.txt
sed: Function /TCP.*SEND/,+10p cannot be parsed.
sed: Function /^$/,+8p cannot be parsed.

六樓
1、
HP-UX:/hwtest/shell>awk '/SEND/{n=8;i=0}/SEND/,n--==0{if(i++>2)print}' c.txt
742.2564
978.6179
1044.1491
1070.1761
1427.1565
1969.4728
942.7927
980.1376
1062.7501
1054.5864
1885.1118
2447.5760

2、
HP-UX:/hwtest/shell>sed '/SEND/,+8!d;/SEND/,+2d' c.txt
sed: Function /SEND/,+8!d;/SEND/,+2d cannot be parsed.

7樓提示
HP-UX:/hwtest/shell>sed -n '/SEND/,+10{/SEND/,+2d;p}' c.txt
sed: Function /SEND/,+10{/SEND/,+2d;p} cannot be parsed.


好像都得不到樓主想要的結(jié)果呢？

yinyuemi

回復(fù) 9# huang_wunix


    HP-UX 不支持GNU 擴(kuò)展？？

1.        GNU sed also supports some special 2-address forms:
   
2. 
   
3.        0,addr2
   
4.               Start  out  in  "matched first address" state, until addr2 is found.  This is similar to 1,addr2,
   
5.               except that if addr2 matches the very first line of input the 0,addr2 form will be at the end  of
   
6.               its range, whereas the 1,addr2 form will still be at the beginning of its range.
   
7. 
   
8.        addr1,+N
   
9.               Will match addr1 and the N lines following addr1.
   
10. 
    
11.        addr1,~N
    
12.               Will  match  addr1 and the lines following addr1 until the next line whose input line number is a
    
13.               multiple of N.
    

復(fù)制代碼