LinuxQuestions.org - [SOLVED] Insert line using sed or awk at line using line number as variable

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - Insert line using sed or awk at line using line number as variable (https://www.linuxquestions.org/questions/programming-9/insert-line-using-sed-or-awk-at-line-using-line-number-as-variable-893652/)

Insert line using sed or awk at line using line number as variable

Hello All,

I want to insert a line at a particular line number using sed or awk.
where line number is not fixed and is in form of variable.

I want to use variable in sed or awk command.

I tried something like below, but no luck.

line=10
sed '$linei\newline' file

Try changing the single quotes to double and see how you go.

Getting error

$ sed "$linei\abcd" 1.conf
sed: 1: "\abcd": unterminated regular expression

I prefer to use awk for this, since then I don't need to worry about escaping:

Code:

awk -v "n=line-number" -v "s=line to insert" '(NR==n) { print s } 1' input-file

First line is line number 1. If the file is too short, it will not insert anything.

This second variant will append the line, if the line number is larger than the number of lines in the file. It will not add empty lines, though; just the line to be inserted:

Code:

awk -v "n=line-number" -v "s=line to insert" '

    (NR==n) { n=-1 ; print s }

            { print $0 }

        END { if (n>0) print s }' input-file

To do the above safely to any file, I'd write a small shell script. I prefer Bash, but since I don't know which shell you prefer, here's something that should work with any /bin/sh. Note, this is untested code:

Code:

#!/bin/sh



LINENUM=55

LINESTR="inserted line"



if [ ":$*" = ":" ] || [*":$*" = ":-h" ] || [ ":$*" = ":--help" ]; then

    exec >&2

    echo ""

    echo "Usage: $0 [ -h | --help ]"

    echo "      $0 filename(s)"

    echo ""

    exit 0

fi



WORK=`mktemp -d` || exit $?

trap "rm -rf '$WORK'" EXIT



while [ $# -gt 0 ]; do

    if [ ! -f "$1" ]; then

        echo "$1: File not found." >&2

        exit 1

    fi



    awk -v "n=$LINENUM" -v "s=$LINESTR" '

            (NR==n) { n=-1; print s }

                    { print $0 }

            END    { if (n>0) print s }'

        ' "$1" > "$WORK/temp" || exit $?



    chown --reference="$1" "$WORK/temp" 2>/dev/null

    chmod --reference="$1" "$WORK/temp" 2>/dev/null



    mv -f "$WORK/temp" "$1" || exit $?



    echo "$1: Modified." >&2

    shift 1

done

Quote:

Originally Posted by sunilsagar (Post 4425053)

Getting error

$ sed "$linei\abcd" 1.conf
sed: 1: "\abcd": unterminated regular expression

Hi,

which sed version are you using? This should work with GNU sed:

Code:

sed "${line} i abcd" 1.conf

Notice the spaces. Make sure that the variable ${line} is not empty.

Great !! thanks so much Guys ..
This worked
sed "${line} i abcd" 1.conf

If you are not sure why it worked? It is because your previous invocation had it looking for a variable called $linei
and then passing \a which sed does not understand as an escape character.

All,

Seeing the code here, what method is used to count all the lines in a current file and put the new line at X+1?

I particular I need a script to add lines to my "sources.list" repositories file, when configuring new servers.

Thanks!

OMR

You want to append lines ? Try:

Code:

echo "New line appended" >> sources.list

Replacing existing lines

All,

OK the "echo "string" >> "path/to/file" works, but got a couple where I need SED or AWK to rewrite the existing lines,

What is correct syntax for that?

Thanks!

OMR