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.

12
  • 7
    Don't forget maintaining consistent levels of abstraction and good names that ensure peeking inside the methods won't surprise you. Don't be surprised if an entire object is hiding in there. Commented Jun 5, 2017 at 15:43
  • 15
    Sometimes splitting methods can be good, but keeping things in-line can also have advantages as well. If a boundary check could sensibly done either inside or outside a block of code, for example, keeping that block of code in-line will make it easy to see whether the boundary check is being done once, more than once, or not at all. Pulling the block of code into its own subroutine would make such things more difficult to ascertain. Commented Jun 5, 2017 at 16:29
  • 7
    The "readability" bullet could be labelled "Abstraction". It's about abstraction. The "bite-sized chunks" do one thing, one low-level thing that you don't really care about the nitty-gritty details for when you're reading or stepping through the public method. By abstracting it to a function you get to step over it and keep focus on the big scheme of things / what the public method is doing at a higher level. Commented Jun 5, 2017 at 16:32
  • 7
    "Testing" bullet is questionable IMO. If your private members want to be tested, then they probably belong in their own class, as public members. Granted, first step to extracting a class/interface is to extract a method. Commented Jun 5, 2017 at 16:35
  • 6
    Splitting up a function purely by line numbers, code blocks and variable scope with no concern for the actual functionality is a terrible idea, and I'd argue the better alternative to that is just leaving it in one function. You should be splitting functions according to functionality. See DaveGauer's answer. You're somewhat hinting at this in your answer (with mentions of names and problems), but I can't help but feel like this aspect needs a whole lot more focus, given its importance and relevance in relation to the question. Commented Jun 6, 2017 at 12:52