Most websites these days are like a menu at a restaurant. They're decorated objects that sit in front of you to help you make sensible requests. Then the requests you made are passed along to the kitchen (the backend web server) which does what you asked for and sends you back the result.

Stealing the source code for the website and hoping to duplicate the web service is like making a copy of a restaurant's menu and hoping to duplicate the restaurant. You won't know anything about how their kitchen runs. You might find it useful to have a list of what requests the web service needs to answer, but most of the hard work hasn't been done for you.