Find all lines containing only the same character and replace with the same amount of another character

Question

My goal is to convert a NEWS format (the output of appstream-util appdata-to-news) to a markdown format (to use on GitHub/GitLab).

A good approximation for me would be to just go from this:

Version 0.5.1
~~~~~~~~~~~~~
Released: 2019-01-03

 * This is a test with a ~ tilde ~~~~
 * Second line

to this

Version 0.5.1
=============
Released: 2019-01-03

 * This is a test with a ~ tilde ~~~~
 * Second line

But my current solution with tr '~' '=' gives me this:

Version 0.5.1
=============
Released: 2019-01-03

 * This is a test with a = tilde ====
 * Second line

The regex to find all the lines containing only ~ should be this: ^~*$ But how can, with bash/sed/awk, replace with the same amount of =?

Answer 1

You can pattern match for lines that contain only tilde characters, and then perform a character-by-character replacement

sed '/^~*$/s/~/=/g'

Answer 2

I Tried with below sed command and it worked fine

Command: sed '/Version/{n;s/~/=/g}' filename

e

n_linux_example ~]# sed '/Version/{n;s/~/=/g}' filename
Version 0.5.1
=============
Released: 2019-01-03

 * This is a test with a ~ tilde ~~~~
 * Second line
[root@praveen_linux_example ~]#