How to insert a text at the beginning of a file?

Question

So far I've been able to find out how to add a line at the beginning of a file but that's not exactly what I want. I'll show it with an example:

File content

some text at the beginning

Result

<added text> some text at the beginning

It's similar but I don't want to create any new line with it...

I would like to do this with sed if possible.

Anything in here work for you? http://stackoverflow.com/questions/54365/prepend-to-a-file-one-liner-shell - you don't need to use `sed`. — wkl, Mar 02 '12 at 12:59

score 476 · Accepted Answer · edited Oct 30 '17 at 18:48

476

sed can operate on an address:

$ sed -i '1s/^/<added text> /' file

What is this magical 1s you see on every answer here? Line addressing!.

Want to add <added text> on the first 10 lines?

$ sed -i '1,10s/^/<added text> /' file

Or you can use Command Grouping:

$ { echo -n '<added text> '; cat file; } >file.new
$ mv file{.new,}

edited Oct 30 '17 at 18:48

Cy Rossignol

16,216
4
57
83

answered Mar 02 '12 at 13:00

kev

155,172
47
273
272

116

To actually insert a new line: sed -i '1s/^/ \n/' file – jslatts Sep 17 '13 at 13:27
2

Why using -i ?? On man it says that it is for suffix. http://linux.die.net/man/1/sed – endrigoantonini May 15 '15 at 02:56
2

-i stands for in-place, you can append a suffix to -i to make a copy rather than overwrite. -i.new would make a new file ending with .new, but just -i would edit the file directly. – Jay Kamat May 22 '15 at 21:31
With command grouping, note the last semicolon before } .. it needs to be there, otherwise shell complaints about unexpected end of file. – pcv Aug 20 '15 at 00:42
6

Note that the sed won't work on an empty file - afaict sed can't be made to do anything at all with 0-length input. – jthill Oct 05 '15 at 20:38
1

@Jirapong It is GNU sed specific. You can achieve the same on e.g. OS X by using gsed. – PJK Feb 14 '16 at 18:59
1

@PJK thank you. You're right but gsed need to install from macport or brew first. – Jirapong Feb 15 '16 at 03:42
First solution adds to first line not to beginning of file. – petroslamb Mar 03 '16 at 10:53
What does the `1` mean in the script `sed -i '1s/^/ /' file` – Alston Mar 09 '16 at 18:39
4

The flaw of this solution is that it doesn't add the text if file is empty. – DBedrenko Apr 16 '16 at 20:45
What if instead of giving text directly, I have to take first line from one file and add as first line to another file ? – Ankit Seth Jan 04 '18 at 08:01
1

Love the last command grouping way. The spaces around the curl brackets are not optional!! – Tianyu Z Aug 24 '20 at 16:38
How to perform this in all files in current directory ? – ajinzrathod Sep 14 '20 at 17:58

score 82 · Answer 2 · edited Feb 05 '14 at 10:42

82

If you want to add a line at the beginning of a file, you need to add \n at the end of the string in the best solution above.

The best solution will add the string, but with the string, it will not add a line at the end of a file.

sed -i '1s/^/your text\n/' file

edited Feb 05 '14 at 10:42

Denim Datta

3,740
3
27
53

answered Feb 05 '14 at 10:18

Ali Reza Ebadat

948
6
5

On Mac OS, was getting error with "undefined label". Found that you need to give an extension for a backup file; see https://mkyong.com/mac/sed-command-hits-undefined-label-error-on-mac-os-x/ – bagsmode Aug 10 '21 at 17:18
7

Under Mac OS's `sed` version you have to supply backup file name with `-i` option. One can pass just an empty string for no backup like `sed -i '' '1s/^/new test\n/' filename` – Grijesh Chauhan Feb 04 '22 at 09:52

score 32 · Answer 3 · answered Mar 02 '12 at 13:00

If the file is only one line, you can use:

sed 's/^/insert this /' oldfile > newfile

If it's more than one line. one of:

sed '1s/^/insert this /' oldfile > newfile
sed '1,1s/^/insert this /' oldfile > newfile

I've included the latter so that you know how to do ranges of lines. Both of these "replace" the start line marker on their affected lines with the text you want to insert. You can also (assuming your sed is modern enough) use:

sed -i 'whatever command you choose' filename

to do in-place editing.

Nick Roz · Answer 4 · 2021-01-27T15:31:57.990

26

Use subshell:

echo "$(echo -n 'hello'; cat filename)" > filename

Unfortunately, command substitution will remove newlines at the end of file. So as to keep them one can use:

echo -n "hello" | cat - filename > /tmp/filename.tmp
mv /tmp/filename.tmp filename

Neither grouping nor command substitution is needed.

edited Jan 27 '21 at 15:31

answered Feb 15 '19 at 19:23

Nick Roz

3,918
2
36
57

1

this solution delete empty lines at the end of the file – dafnahaktana Sep 09 '20 at 05:34
2

This is the superior and simpler solution to all the other ones using `sed`, which don't work for empty files. – WalksB Nov 12 '20 at 18:32
@dafnahaktana that's because of command substitution – Nick Roz Nov 13 '20 at 13:40

score 18 · Answer 5 · answered Jan 04 '16 at 03:56

18

To insert just a newline:

sed '1i\\'

answered Jan 04 '16 at 03:56

2

I wonder why the top voted answers all use s instead of i. Any downsides to this? – Caesar Jul 03 '20 at 05:04
4

@Caesar Because when all you have is a hammer, every problem looks like a nail. – Tobia Jun 21 '21 at 13:34

score 15 · Answer 6 · edited Feb 19 '15 at 12:37

15

You can use cat -

printf '%s' "some text at the beginning" | cat - filename

edited Feb 19 '15 at 12:37

gniourf_gniourf

44,650
9
93
104

answered Feb 19 '15 at 12:30

Lex

183
1
2

8

This would only output the text followed by the file's content, but it does not modify the file at all. – Adaephon Feb 19 '15 at 14:16
That's a good solution, I wonder why it didn't get any upvotes. Here's mine my good sir. Also, why printf and not a simple echo ? – ychaouche Jun 22 '15 at 13:38
1

@ychaouche - presumably because there's no portable way to prevent `echo` from adding a newline? – Toby Speight Mar 30 '16 at 09:47
I tried directing this into a file by appending `> file` to the command, it jus spammed my terminal with `"some text at the beginning"` – theonlygusti May 10 '17 at 11:29
2

`printf '%s' "some text" | cat - filename > tmpfile && mv tmpfile filename` – luca Oct 18 '19 at 23:49

score 12 · Answer 7 · answered Mar 30 '16 at 07:29

12

To add a line to the top of the file:

sed -i '1iText to add\'

answered Mar 30 '16 at 07:29

solidsnack

1,631
16
26

This includes a newline, which the questioner explicitly doesn't want. – Toby Speight Mar 30 '16 at 09:20

score 12 · Answer 8 · edited Aug 03 '23 at 10:53

12

my two cents:

sed -i '1i /path/of/file.sh' filename

where /path/of/file.sh is the text you want to add.

This will work even if the string contains a forward slash "/"

edited Aug 03 '23 at 10:53

Toby Speight

27,591
48
66
103

answered Dec 17 '18 at 11:35

vardhan

1,377
11
14

This includes a newline, which the questioner explicitly doesn't want. – Toby Speight Aug 03 '23 at 10:53

score 9 · Answer 9 · answered Dec 27 '13 at 10:05

9

Hi with carriage return:

sed -i '1s/^/your text\n/' file

answered Dec 27 '13 at 10:05

m0zgen

569
7
20

score 9 · Answer 10 · answered Jul 09 '15 at 20:56

9

Note that on OS X, sed -i <pattern> file, fails. However, if you provide a backup extension, sed -i old <pattern> file, then file is modified in place while file.old is created. You can then delete file.old in your script.

answered Jul 09 '15 at 20:56

Curt Clifton

91
1
1

4

Or sed -i '' file. Then it'll just edit in-place and not make backups. – Harv Jul 14 '15 at 20:23
1

I've found macOS sed wants a dot to edit in place without creating a backup: sed -i. file – A Bowler Cat Nov 30 '19 at 06:08

score 7 · Answer 11 · answered Nov 30 '17 at 02:04

7

There is a very easy way:

echo "your header" > headerFile.txt
cat yourFile >> headerFile.txt

answered Nov 30 '17 at 02:04

Raphael Villas Boas

229
3
3

Ow... sorry, now that I realize that you wanted to add at the beginning and not a new line. – Raphael Villas Boas Nov 30 '17 at 02:06
But *I* wanted a new line. So thanks for the simplistic approach. – TheSatinKnight Dec 21 '17 at 02:20

Victoria Stuart · Answer 12 · 2017-12-02T01:48:03.073

PROBLEM: tag a file, at the top of the file, with the base name of the parent directory.

I.e., for

/mnt/Vancouver/Programming/file1

tag the top of file1 with Programming.

SOLUTION 1 -- non-empty files:

bn=${PWD##*/}    ## bn: basename

sed -i '1s/^/'"$bn"'\n/' <file>

1s places the text at line 1 of the file.

SOLUTION 2 -- empty or non-empty files:

The sed command, above, fails on empty files. Here is a solution, based on https://superuser.com/questions/246837/how-do-i-add-text-to-the-beginning-of-a-file-in-bash/246841#246841

printf "${PWD##*/}\n" | cat - <file> > temp && mv -f temp <file>

Note that the - in the cat command is required (reads standard input: see man cat for more information). Here, I believe, it's needed to take the output of the printf statement (to STDIN), and cat that and the file to temp ... See also the explanation at the bottom of http://www.linfo.org/cat.html.

I also added -f to the mv command, to avoid being asked for confirmations when overwriting files.

To recurse over a directory:

for file in *; do printf "${PWD##*/}\n" | cat - $file > temp && mv -f temp $file; done

Note also that this will break over paths with spaces; there are solutions, elsewhere (e.g. file globbing, or find . -type f ... -type solutions) for those.

ADDENDUM: Re: my last comment, this script will allow you to recurse over directories with spaces in the paths:

#!/bin/bash

## https://stackoverflow.com/questions/4638874/how-to-loop-through-a-directory-recursively-to-delete-files-with-certain-extensi

## To allow spaces in filenames,
##   at the top of the script include: IFS=$'\n'; set -f
##   at the end of the script include: unset IFS; set +f

IFS=$'\n'; set -f

# ----------------------------------------------------------------------------
# SET PATHS:

IN="/mnt/Vancouver/Programming/data/claws-test/corpus test/"

# https://superuser.com/questions/716001/how-can-i-get-files-with-numeric-names-using-ls-command

# FILES=$(find $IN -type f -regex ".*/[0-9]*")        ## recursive; numeric filenames only
FILES=$(find $IN -type f -regex ".*/[0-9 ]*")         ## recursive; numeric filenames only (may include spaces)

# echo '$FILES:'                                      ## single-quoted, (literally) prints: $FILES:
# echo "$FILES"                                       ## double-quoted, prints path/, filename (one per line)

# ----------------------------------------------------------------------------
# MAIN LOOP:

for f in $FILES
do

  # Tag top of file with basename of current dir:
  printf "[top] Tag: ${PWD##*/}\n\n" | cat - $f > temp && mv -f temp $f

  # Tag bottom of file with basename of current dir:
  printf "\n[bottom] Tag: ${PWD##*/}\n" >> $f
done

unset IFS; set +f

score 4 · Answer 13 · answered May 21 '18 at 18:48

Just for fun, here is a solution using ed which does not have the problem of not working on an empty file. You can put it into a shell script just like any other answer to this question.

ed Test <<EOF
a

.
0i
<added text>
.
1,+1 j
$ g/^$/d
wq
EOF

The above script adds the text to insert to the first line, and then joins the first and second line. To avoid ed exiting on error with an invalid join, it first creates a blank line at the end of the file and remove it later if it still exists.

Limitations: This script does not work if <added text> is exactly equal to a single period.

Murilo Perrone · Answer 14 · 2022-01-26T07:03:46.640

3

The simplest solution I found is:

echo -n "<text to add>" | cat - myFile.txt | tee myFile.txt

Notes:

Remove | tee myFile.txt if you don't want to change the file contents.
Remove the -n parameter if you want to append a full line.
Add &> /dev/null to the end if you don't want to see the output (the generated file).

This can be used to append a shebang to the file. Example:

# make it executable (use u+x to allow only current user)
chmod +x cropImage.ts
# append the shebang
echo '#''!'/usr/bin/env ts-node | cat - cropImage.ts | tee cropImage.ts &> /dev/null
# execute it
./cropImage.ts myImage.png

edited Jan 26 '22 at 07:03

answered Jan 04 '22 at 17:30

Murilo Perrone

452
4
7

1

This appends the text twice in the same line – David Heisnam Jan 26 '22 at 06:37
1

It also replaces the content of the entire file. – David Heisnam Jan 26 '22 at 06:41
@DavidHeisnam, yes it replaces the file, unless you remove the tee pipe (I have updated the answer). But It won't append the text twice. You may have run the command twice. It's working in bash and zsh. – Murilo Perrone Jan 26 '22 at 07:07
1

In a for loop (in bash on OSX) it replaces the contents with only the text that should be added. Without for it worked for me – flexus Oct 17 '22 at 12:54
I confirm this does not retain the original file contents – Matěj Račinský Feb 03 '23 at 10:33

score 2 · Answer 15 · edited Jun 05 '15 at 21:29

2

echo -n "text to insert " ;tac filename.txt| tac > newfilename.txt

The first tac pipes the file backwards (last line first) so the "text to insert" appears last. The 2nd tac wraps it once again so the inserted line is at the beginning and the original file is in its original order.

edited Jun 05 '15 at 21:29

qwerty_so

35,448
8
62
86

answered Jun 05 '15 at 20:51

paulp

29
2

1

Can you please explain how this would help Tomas with the issue? – John Odom Jun 05 '15 at 20:56
tac | tac -- that's how non-scalable mindsets form :-( – Michael Shigorin May 22 '18 at 12:55

Dr Bala Soundararaj · Answer 16 · 2023-05-26T09:27:12.477

1

cat concatenates multiple files. <() sends output of a command as a file. echo doesn't add new lines with -n switch. head command with -c switch can remove trailing newline in files. Combining all these, we can insert text at the beginning and end of a file without any new lines by,

cat <(echo -n "line before the file") <(head -c -1 file.txt) <(echo -n "line after the file")

edited May 26 '23 at 09:27

answered Jan 30 '23 at 02:09

Dr Bala Soundararaj

165
6

`head -c -1` doesn't work on bsd (& Macs) so please use `ghead` from `coreutils` – Dr Bala Soundararaj May 26 '23 at 09:26

Farhad Kazemi · Answer 17 · 2023-07-16T06:30:51.777

1

thanks to all the top answers above, just a small hint:

for the sed approach, if you want to pass a variable while going to a newline, like what I've wanted to do with my /etc/hosts file, don't forget to use double quotes "

sed -i "1s/^/127.0.0.1 $HOSTNAME \n/" /etc/hosts

edited Jul 16 '23 at 06:30

answered Jul 16 '23 at 06:29

Farhad Kazemi

11
2

score 0 · Answer 18 · answered Nov 04 '19 at 16:49

Another solution with aliases. Add to your init rc/ env file:

addtail () { find . -type f ! -path "./.git/*" -exec sh -c "echo $@ >> {}" \; }
addhead () { find . -type f ! -path "./.git/*" -exec sh -c  "sed -i '1s/^/$@\n/' {}" \; }

Usage:

addtail "string to add at the beginning of file"
addtail "string to add at the end of file"

score 0 · Answer 19 · answered Sep 09 '20 at 03:44

With the echo approach, if you are on macOS/BSD like me, lose the -n switch that other people suggest. And I like to define a variable for the text.

So it would be like this:

Header="my complex header that may have difficult chars \"like these quotes\" and line breaks \n\n "

{ echo "$Header"; cat "old.txt"; } > "new.txt"
mv new.txt old.txt

Kajukenbo · Answer 20 · 2022-09-05T07:48:46.533

TL;dr -

Consider using ex. Since you want the front of a given line, then the syntax is basically the same as what you might find for sed but the option of "in place editing" is built-in.

I cannot imagine an environment where you have sed but not ex/vi, unless it is a MS Windows box with some special "sed.exe", maybe.

sed & grep sort of evolved from ex / vi, so it might be better to say sed syntax is the same as ex.

You can change the line number to something besides #1 or search for a line and change that one.

source=myFile.txt
Front="This goes IN FRONT "
man true > $source
ex -s ${source} <<EOF
1s/^/$Front/ 
wq
EOF

$ head -n 3 $source
This goes IN FRONT TRUE(1)                                                    User Commands                                                    TRUE(1)

NAME

Long version, I recommend ex (or ed if you are one of the cool kids).

I like ex because it is portable, extremely powerful, allows me to write in-place, and/or make backups all without needing GNU (or even BSD) extensions.

Additionally, if you know the ex way, then you know how to do it in vi - and probably vim if that is your jam. Notice that EOF is not quoted when we use "i"nsert and using echo:

str="+++ TOP +++" && ex -s <<EOF
r!man true
1i
`echo "$str"`
.
"0r!echo "${str}"
wq! true.txt
EOF

0r!echo "${str}" might also be used as shorthand for :0read! or :0r! that you have likely used in vi mode (it is literally the same thing) but the : is optional here and some implementations do not support "r"ead address of zero. "r"eading directly to the special line #0 (or from line 1) would automatically push everything "down", and then you just :wq to save your changes.

$ head -n 3 true.txt | nl -ba   
     1  +++ TOP +++
     2  TRUE(1)                          User Commands                         TRUE(1)
     3

Also, most classic sed implementations do not have extensions (like \U&) that ex should have by default.

How to insert a text at the beginning of a file?

20 Answers20

Linked

Related