I have 34 different methods that take 1 parameter of my ViewModel class, but each of them only uses two or three properies on the class. I think I’d rather just pass those properties to these methods instead of the entire object. Is that a way to easily refactor that? I tried to just select all the text inside my function and do Refactor > Extract Method hoping it would be smart enough to see that only 2 properties of the object were being used and make the parameter list just those 2 properties, but it just wanted to take the whole object again.
Maybe I need to explain what I’m doing…
I’m trying to write an application that will generate text for a wiki page. Based on the values entered into the UI it will either generate text for a template and text for using that template, or it will just generate the text for the page if a template isn’t necessary. There are 17 different parts to it so there are 17 functions that, when combined, create the text of the template page, and there are 17 functions that combine to make the actual page if a template isn’t needed. I did it this way so I can test smaller pieces of it at a time instead of testing the entire output of one massive function.
So currently all 34 of those methods take 1 parameter, the whole object that’s built from the UI. To me that doesn’t make sense any more. A function that generates 1 section of the entire page will never need to know about the data that’s used in a different section, and if anyone does try to use it there then I know something’s wrong.
Is there a reason not to pass the object? Objects are passed by reference and it’s quite efficient, plus if your implementation needs more properties, it’s easy to modify the function than to maintain the function declaration and go after all the callers. I’d be inclined to leave it alone.
But to answer your specific question, I don’t know of a way to refactor it in Visual Studio. That’s another part of why I’d be inclined to leave it alone if it’s working. 🙂