Parse/Extract bad Email address from Undeliverable/Returned emails

[Mojgan]

I am trying to parse the body of an undeliverable/returned email and extract the bad email address.  
Undeliverable/returned emails come in different formats. The email address may be in different positions in the email body and preceded by different words or syntax. Therefore,  "Filter Array' or "Split by Colon" didn't help so far. They did help if I know the position or the preceding word.

In addition, when I use 'HTML to TEXT' function, in certain instances, it duplicates the email according to below example:

    'ABCD@ucu.edu'   to   'ABCD@ucu.eduABCD@ucu.edu'

Which makes it even harder to grab the email. 

Is there a way to accomplish this task? 

Here is my flow so far:

[VictorIvanidze]

Hi @Mojgan,

- do not use HTMLToText. Parse HTML.

- find all email addresses by @(at), then check if there is something like "#550 5.1.0" before given address.