Skip to main content
Software Engineering

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

6
  • 5
    It would entirely depend on what the function is for and how it relates (or doesn't relate) to the object in question. Commented May 24, 2016 at 19:19
  • That's the problem. I can't tell when I'd use one or the other. I feel like I could always change the code to accommodate either approach. Commented May 24, 2016 at 19:20
  • 2
    In API terms (and not looking at the implementations at all), the former is abstract and domain oriented (which is good), whereas the latter is not (which is bad). Commented May 24, 2016 at 20:39
  • 1
    The first approach would be more 3-tier OO. But it should be even more so by eliminating the word database from the method. It should be "Store" or "Persist" and do either Account or Thing (not both). As a client of this layer you should not be aware of the storage medium. When retrieving an Account you would need to pass in the id though or a combination of property values (not field values) to identify the desired object. Or/and implenent an enumeration method that passes all accounts. Commented May 24, 2016 at 22:20
  • 1
    Typically, both would be wrong (or, rather, less than optimal). How an object should be serialized into the database should be a property (a member function) of the object, because it typically directly depends on the member variables of the object. In case you change members of the object, you will also need to change the serialization method. That works better if it is part of the object Commented Mar 14, 2017 at 17:24