This article is part of the on going Unix sed command tutorial series. In our previous articles we learned sed with single commands — printing, deletion, substitute and file write.
Sed provides lot of commands to perform number of operations with the lines in a file.
In this article let us review how to append, insert, replace a line in a file and how to get line numbers of a file.
Let us first create thegeekstuff.txt file that will be used in all the examples mentioned below.
$cat thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc.
Append Lines Using Sed Command
Sed provides the command “a” which appends a line after every line with the address or pattern.
Syntax: #sed 'ADDRESS a\ Line which you want to append' filename #sed '/PATTERN/ a\ Line which you want to append' filename
Sed Append Example 1. Add a line after the 3rd line of the file.
Add the line “Cool gadgets and websites” after the 3rd line. sed “a” command inserts the line after match.
$ sed '3 a\ > Cool gadgets and websites' thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Cool gadgets and websites Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc.
Sed Append Example 2. Append a line after every line matching the pattern
The below sed command will add the line “Linux Scripting” after every line that matches the pattern “Sysadmin”.
$ sed '/Sysadmin/a \ > Linux Scripting' thegeekstuff.txt Linux Sysadmin Linux Scripting Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc. Linux Scripting
Sed Append Example 3. Append a line at the end of the file
The following example, appends the line “Website Design” at the end of the file.
$ sed '$ a\ > Website Design' thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc. Website Design
Insert Lines Using Sed Command
Sed command “i” is used to insert a line before every line with the range or pattern.
Syntax: #sed 'ADDRESS i\ Line which you want to insert' filename #sed '/PATTERN/ i\ Line which you want to insert' filename
Sed Insert Example 1. Add a line before the 4th line of the line.
Add a line “Cool gadgets and websites” before 4th line. “a” command inserts the line after match whereas “i” inserts before match.
$ sed '4 i\ > Cool gadgets and websites' thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Cool gadgets and websites Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc.
Sed Insert Example 2. Insert a line before every line with the pattern
The below sed command will add a line “Linux Scripting” before every line that matches with the pattern called ‘Sysadmin”.
$ sed '/Sysadmin/i \ > Linux Scripting' thegeekstuff.txt Linux Scripting Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Linux Scripting Windows- Sysadmin, reboot etc.
Sed Insert Example 3. Insert a line before the last line of the file.
Append a line “Website Design” before the last line of the file.
$ sed '$ i\ > Website Design' thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Website Design Windows- Sysadmin, reboot etc.
Replace Lines Using Sed Command
“c” command in sed is used to replace every line matches with the pattern or ranges with the new given line.
Syntax: #sed 'ADDRESS c\ new line' filename #sed '/PATTERN/ c\ new line' filename
Sed Replace Example 1. Replace a first line of the file
The below command replaces the first line of the file with the “The Geek Stuff”.
$ sed '1 c\ > The Geek Stuff' thegeekstuff.txt The Geek Stuff Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc.
Sed Replace Example 2. Replace a line which matches the pattern
Replace everyline which has a pattern “Linux Sysadmin” to “Linux Sysadmin – Scripting”.
$ sed '/Linux Sysadmin/c \ > Linux Sysadmin - Scripting' thegeekstuff.txt Linux Sysadmin - Scripting Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Windows- Sysadmin, reboot etc.
Sed Replace Example 3. Replace the last line of the file
Sed command given below replaces the last line of the file with “Last Line of the file”.
$ sed '$ c\ > Last line of the file' thegeekstuff.txt Linux Sysadmin Databases - Oracle, mySQL etc. Security (Firewall, Network, Online Security etc) Storage in Linux Productivity (Too many technologies to explore, not much time available) Last line of the file
Print Line Numbers Using Sed Command
“=” is a command in sed to print the current line number to the standard output.
Syntax: #sed '=' filename
The above send command syntax prints line number in the first line and the original line from the file in the next line .
sed ‘=’ command accepts only one address, so if you want to print line number for a range of lines, you must use the curly braces.
Syntax:
# sed -n '/PATTERN/,/PATTERN/ {
=
p
}' filename
Sed Line Number Example 1. Find the line number which contains the pattern
The below sed command prints the line number for which matches with the pattern “Databases”
$ sed -n '/Databases/=' thegeekstuff.txt 2
Sed Line Number Example 2. Printing Range of line numbers
Print the line numbers for the lines matches from the pattern “Oracle” to “Productivity”.
$ sed -n '/Oracle/,/Productivity/{
> =
> p
> }' thegeekstuff.txt
2
Databases - Oracle, mySQL etc.
3
Security (Firewall, Network, Online Security etc)
4
Storage in Linux
5
Productivity (Too many technologies to explore, not much time available)
Sed Line Number Example 3. Print the total number of lines in a file
Line number of the last line of the file will be the total lines in a file. Pattern $ specifies the last line of the file.
$ sed -n '$=' thegeekstuff.txt 6
My name is Ramesh Natarajan. I will be posting instruction guides, how-to, troubleshooting tips and tricks on Linux, database, hardware, security and web. My focus is to write articles that will either teach you or help you resolve a problem. Read more about
Comments on this entry are closed.
very nice tutorial !
With Debian version of sed, you need to use the option ‘-i’ to really append lines in your file.By default, the modification appears on the standard output.
@linuxindetails,
Thanks for clarifying how the option -i works on Debian.
Hi,
How to insert content of file1 on the top of the file2, file3, file4, etc in same directory ?
Hi
How I can escape Back slashes in sed. I want to change the pattern or string
from
allow ^127\.0\.0\.1$
to
allow ^192\.168\.1\.10$
Thank You,
Manish
To balast:
for f in `ls file[2-4].txt` ; do
cat file1.txt $f > newfile
cp newfile $f
done
To Manish:
sed ‘s/127\\\.0/192\\\.168/g’ filename
works
@ Blast
cat file2>> file1 should work fine
Very nice tutorial…
But how to write these sed commands in a script… do they work only in command-line?
Thanks! I nominate for the best set of sed instructions on the Internet.
Balaji — they all work in scripts; see here for instance
I want to change the first line of a csv file into lower case. How to do that. thanks
I want to add “something else” after “something” , so i use the below code.
sed ‘
/something/ a\
something else
‘ test.txt
This will just list the contents,but how do i write this new line to the file. I mean the same file and not to a new file.
I agree with arvind, how do you write the changes to the original file. In other words, make the input also the output after sed function has been applied. that was a pun..ha
@Anonymous
cat file1>>file2 only works when the text of file1 may be appended (added to the bottom) to file2.
@Balast
Maybe something like
for f in file[2-4];do sed 1r$f file1>$f;done
could work. It is important not to add quotes around file[2-4], else only the contents of file1 are written to the another files.
for f in file[2-4];do cat file1 $f>$f;done
could also work.
Be careful to first test the strings, because on other computers, it can work different from mine.
sed -i -e ‘s|foo|bar|’ file.text
Note the “-i” is for in-place. For safety it will make a backup if you supply an extension to -i, as in “-i.bak”.
sed ‘1 c\
> The Geek Stuff’ thegeekstuff.txt
this command doesn’t affect the original file.
help me
Hi Shridhara,
I am trying to achieve something similar only solution so far for me is
echo “The Geek Stuff” | sed ‘1a\’ >> thegeekstuff.txt
It should be something cleaner tough.Anyhow hope it helps.
Carlos
Argh, the formatting in this article is so confusing. There shouldn’t be any linefeed after the backslashes. So it’s not this:
sed ‘ADDRESS i\
Stuff to add’ filename
It’s this:
sed ‘ADDRESS i\Stuff to add’ filename.
Otherwise you get this error:
sed: -e expression #1, char 32: extra characters after command
For those who are on SunOS which is non-GNU, the following code will help to insert a line at top of file, to delete first line or delete last line:
sed ‘1i\^J
line to add’ test.dat > tmp.dat
sed ‘1d’ test.dat > tmp.dat
sed ‘$d’ tmp.dat > test.dat
^J is inserted with ^V+^J
Add the newline after ‘1i.
\ MUST be the last character of the line.
The second part of the command must be in a second line.
i want to channge a file with contents like
}
some text here
want to remove the new line after }. so the result should be
}some text here
how to do it using sed
Is variable substitution supported with the sed insert command ?
I tried to insert the public IP as the first line in /etc/hosts , but its not working
eg:
publicIP=$(curl -s checkip.dyndns.org|egrep -o ‘[0-9\.]+’ 2>/dev/null)
sed “1 i\${publicIP} heuristics” /etc/hosts
Any idea ?
For inserting lines from the commandline I use the “s” command.
E.g. inserting “the very first line” to the file geek.txt do
sed “1s/^/the very first line\n/”
My problem with the a and i command (working at the commandline) is, that you cannot define the END of the command. So everything following it will be printed – and usually you want to do multiple commands… (though I’m working on a windows-machine, maybe for unix/linux there is a better solution)
How to used the sed append, insert functions in a script to be excuted on remote server. For example I want the command below to be excuted on a remote server using ssh. (solaris)
$ sed ‘$ i\
> Website Design’ thegeekstuff.txt
Very clean and helpful tutorial. Thanks!
Very good tutorial!
But how can i append something at the end of a matched line?
Thanks
When I try the append examples it doesn’t add a newline to the end of the appended line.
If test.txt is:
LINE1
FOO
LINE3
sed ‘/FOO/ a
BAR’ test.txt
gives
LINE1
FOO
BARLINE3
Makes it easy to understand 🙂
@ bob:
sed ‘/FOO/s/$/BAR/’ test.txt
#substitute the end of the line with BAR
Plz explain command
$mv $$ foo.c; head -n 2 foo.c
Where foo.c n $$ r filenames
hi all
i was using this sed command to append a line at the end.
i was using below command dint work
>sed ‘$ a\ my name is kiran’ details.txt
but the changes were not getting saved, when i used below command it was ok
> sed ‘$ i\ my name is kiran’ -i details.txt
amazing……..
HI Please let me know the solution ..if I want to store in a file value such as $$a=20
and 20 is coming from a variable .How can I do that?
thank you! great stuff…
Thanks for the nice tutorial.
But I couldn’t get it to work in shell script and also to get it to update the existing file.
Hey! How to delete or append a line after a pattern match using SED?
I tried using {N; d} and ,1d – both doesn’t work. Can anyone please help!
hi CJ. To delete lines after a pattern match is:
To append to lines matching the pattern is:
Or if appending after the line matching the pattern is:
hi CJ. Here is the correct way to delete the next file following the line matching the pattern:
Explanation: I created an if-else branching condition labeled as “L0”. If the line does not match the pattern then goto L0 which prints the line. If the line matches the pattern, then print the line first, extract the next line and delete it.