sed - 使用命令行追加到下一行

Question

I wish to append " dddd" to the next line whenever I encounter "=" in a textfile.

This command

sed -i '/=/s|$| dddd|' *.krn

is close to what I am looking for as it appends to the current line where "=" is. How can I append to the next line instead?

Answer 1

Use append, see here:

• http://www.grymoire.com/Unix/Sed.html#uh-40

Eg:

$ echo $'abc\ndef\ne=f\nqqq'
abc
def
e=f
qqq
$ echo $'abc\ndef\ne=f\nqqq'|sed '/=/adddd'
abc
def
e=f
dddd
qqq

Edited to clarify as per comment from @je4d- if you want to append to what is present in the next line, you can use this:

$ echo $'abc\ndef\ne=f\nqqq\nyyy'
abc
def
e=f
qqq
yyy
$ echo $'abc\ndef\ne=f\nqqq\nyyy'|sed '/=/{n;s/$/ dddd/}'
abc
def
e=f
qqq dddd
yyy

See here for a great sed cheatsheet for more info if you want:

• http://www.catonmat.net/download/sed.stream.editor.cheat.sheet.txt

Answer 2

So to reiterate the question, when you match on one line, you want to append a string to the next line---a line that already exists, rather than adding a new line after it with the new data.

I think this will work for you:

sed '/=/ { N; s/$/ ddd/ }'

Say you have a file like:

=
hello
world
=
foo
bar
=

Then processing this command on it will yield:

=
hello ddd
world
=
foo ddd
bar
=

The trick here is using the N command first. This reads in the "next" line of input. Commands following it will be applied to the next line.

Answer 3

I'm not a sed guru, but I can do what you want with awk:

'{PREV=MATCH; MATCH="no"}
 /=/{MATCH="yes"} 
 PREV=="yes"{print $0 " dddd"}
 PREV!="yes"{print}'

Demo:

$ echo -e 'foo\nba=r\nfoo\n=bar\nfoo\nfoo\nb=ar\nx' 
foo
ba=r
foo
=bar
foo
foo
b=ar
x

$ echo -e 'foo\nba=r\nfoo\n=bar\nfoo\nfoo\nb=ar\nx' | awk '{APPEND=LAST; LAST="no"} /=/{LAST="yes"} APPEND=="yes"{print $0 " dddd"} APPEND!="yes"{print}'
foo
ba=r
foo dddd
=bar
foo dddd
foo
b=ar
x dddd

Answer 4

This might work for you:

echo -e "=\nx " | sed '/=/{$q;N;s/$/dddd/}'
=
x dddd