How to insert lines after specific string with `:` and `=`?

Question

a.txt has only one long string as example:

echo 'stack=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU' >a.txt

I would like insert 2 lines after string stack=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU in a.txt as below :

~$ cat a.txt
    stack=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU
#insert after above string
    stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU
    stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU

I tried

sed '/stack\=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU/a stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU' a.txt

but failed.

What's the problem?

Could you please do mention on where to insert the line, if you could elaborate on condition part more it will make question more clear, thank you. — RavinderSingh13, Jun 18 '21 at 17:53
Are you trying to append 2 lines to `a.txt` no matter what it contains (and you're telling us it only contains 1 line, the target line) or are you trying to append 2 lines after the target line wherever that target line appears in `a.txt`? Please [edit] your question provide more truly representative sample input/output that unambiguously demonstrates your requirements. — Ed Morton, Jun 18 '21 at 18:25

score 2 · Answer 1 · answered Jun 18 '21 at 17:56

You can use it like this:

sed -i.bak '/^stack=log1:/r b.txt' a.txt

cat a.txt

stack=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU
stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU
stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU

Where you have:

cat b.txt

stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU
stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU

Details:

/^stack=log1:/: Search for string stack=log1: at the start a line
r b.txt: Insert content from file b.txt after the match

It's a good solution,just need to create another file. – kittygirl Jun 18 '21 at 18:14 — kittygirl, Jun 18 '21 at 18:14

Ed Morton · Accepted Answer · 2021-06-18T18:21:28.033

2

Your question isn't clear. If a.txt only has 1 line and you want to add those 2 other lines to that file then all you need is:

cat <<! >> a.txt
stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU
stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU
!

Otherwise if a.txt is a multi-line file and you want to add 2 lines after your target line no matter where it appears in that file then:

awk '
{ print }
$0 == "stack=log1:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu1:LOGEMU" {
    print "stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU"
    print "stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU"
}
' a.txt

To update a.txt if you have GNU awk use awk -i inplace 'script' a.txt, or with any awk tmp=$(mktemp) && awk 'script' a.txt > "$tmp" && mv "$tmp" a.txt.

edited Jun 18 '21 at 18:21

answered Jun 18 '21 at 18:14

Ed Morton

188,023
17
78
185

I tried method with multi lines,it seemed not workable when `sudo awk ' { print }...`. – kittygirl Jun 18 '21 at 18:28
It's workable just fine. If you have a specific problem please tell me some symptoms so I can help. – Ed Morton Jun 18 '21 at 19:09
When `sudo awk ' { print }...'a.txt`,just print the output,`a.txt` has no any change. – kittygirl Jun 19 '21 at 07:21
How about without the `sudo`? Maybe you have DOS line endings (see https://stackoverflow.com/questions/45772525/why-does-my-tool-output-overwrite-itself-and-how-do-i-fix-it)? Maybe you have white space at the end of lines? As you can see in my answer when you run the script I posted against the input file you posted it does produce the expected output you posted so if it's not doing the same for you then either a) you copy/pasted the script incorrectly or b) the input file you're running it against is different in some way from the input file you posted? – Ed Morton Jun 19 '21 at 13:00

GAD3R · Answer 3 · 2021-06-18T18:50:45.530

2

Insert two lines after a match should work:

sed -i '/^stack=log1/a \
stack=log2:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu2:LOGEMU \
stack=log3:NFLOG,base1:BASE,ifi1:IFINDEX,ip2str1:IP2STR,print1:PRINTPKT,emu3:LOGEMU' a.txt

edited Jun 18 '21 at 18:50

answered Jun 18 '21 at 18:43

GAD3R

4,317
1
23
34

When `sudo`,just print output,file has no change. – kittygirl Jun 18 '21 at 18:49
@kittygirl use `-i` option see my edit please. – GAD3R Jun 18 '21 at 18:50
That usage of `-i` without a backup file is specific to GNU sed. If the OP is using another sed then it won't behave as desired even if it has a `-i` option (e.g. MacOS sed). – Ed Morton Jun 18 '21 at 20:42

How to insert lines after specific string with `:` and `=`?

3 Answers3