I am working on an email parsing cloud flow. The email comes in, and I use an HTML-to-text action to remove the HTML tags.
Note, the only HTML tags in the email are <p></p> tags at the start and end of the email. If I copy the output of the HTML to text action into Word and turn on character recognition, I can see carriage returns in the content.

Then, I use a compose action to hold a new line or carriage return.

Next, I use another compose with the following expression to split the lines of the email on the carriage returns.

split(outputs('Html_to_text')?['body'], outputs('New_Line'))
But it doesn't work. The lines all run together in one big paragraph. Any ideas????

Ultimately I want the original line breaks in the output.