Without understanding the full issue, if I were you, I'd try the following:

As long as you are sure that it always starts with that same string. I'd throw a \n tag metacharacter after the end square bracket to take out the line break in the files.

BTW, I believe the $ tag indicates that the text must be at the end of the line being checked, so your regex would only remove poseidon (and everything before it on the line)