Substitution on range of lines

Question

I have in 3rd line of a file. I want to replace that with my_dB. How to do this with sed

Answer 1

To replace all instances of <host></host> with <host>my_dB</host> in your file, run:

sed 's|<host></host>|<host>my_dB</host>|g' file

Answer 2

Each can be done with something similar to this:

sed -i "s/<host><\/host>/<host>my_db<\/host>/" foo.txt

The -i means that it'll overwrite the file in-place. Remove that flag to have the changes written to stdout.

If this is for a regular task to fill in details for the file, consider adding anchors to the file to make it easier. For example:

<host>DB_HOST</host>

Then the sed would just need to be:

sed -i 's/DB_HOST/my_db/'

Answer 3

Like this:

$ sed '3,6 s/<host>/<host>my_dB/' file

Demo:

$ cat file
1: <host></host>
2: <host></host>
3: <host></host>
4: <host></host>
5: <host></host>
6: <host></host>
7: <host></host>

$ sed '3,6 s/<host>/<host>my_dB/' file
1: <host></host>
2: <host></host>
3: <host>my_dB</host>
4: <host>my_dB</host>
5: <host>my_dB</host>
6: <host>my_dB</host>

    7:

To store the changes back to the file use the -i option

$ sed -i '3,6 s/<host>/<host>my_bd/' file

However you really should be using a XML parser. Getting into parsing XML with regexp is really a bad idea.

Answer 4

try this:

 sed -r '3,5s#(<host>)(</host>)#\1my_dB\2#' file