It is not a good solution to save the complete HTML tag into a database. Why was this done?

I am currently getting the picture url with regex but unfortunately, i can not access the picture because I have to log in. Is there a way to get the picture with an auth token?