macOS Code files question

I would aim for smaller files, to the extent that this makes sense. It's not very helpful, but files generally map to a class or grouping of functionality. As such, you should have exactly what you need to do one thing well. If you're shoving too much code into one file it MAY indicate an over-complicated object design that could be broken down into simpler, single-purpose objects that compose the larger piece.

Don't consolidate for no reason.

-Lee

First off, nobody talks about the number of words in a source file. They talk about lines, columns, or sometimes statements or functions or classes.

My rules of thumb are:
- Aim for each line to be under 80 columns wide - situations may arise where that isn't possible, though. Don't ever let a line be over 120 columns wide though - find a way to make the line shorter. IE, shorter variable names.
- Aim for each file being between 200 and 2000 lines long. Situations may arise where that isn't possible - make sure to keep related functionality together. Generally a well organized project will have the file line counts within that range.

Although really, the entire concept of files is just silly when you're talking about programming. Light Table looks like an interesting IDE... it looks like it's actually ready for Python development so maybe I'll give that a spin right now...

I started to read a book a while back called "Clean Code", I need to get back to it again. The book covers how to write good code and teaches good habits.

http://www.amazon.com/Clean-Code-Handbook-Software-Craftsmanship/dp/0132350882

Click to expand...

I found myself over the years understanding code pretty well but writing it looked really bad. I would come back to projects that I had not worked on in 6 months and be confused as to what I was looking at even though I wrote it.

I used the same code over and over in methods to do the same thing. My methods and functions did more then one thing. For instance I would have a method generate a random number and the same method would process that data even further. Now I write multiple methods or functions to perform the task that 1 method did before. So now I would have a method just generate a random number and return it. It makes my code cleaner and easier to read.

For code that is used by many different classes like reading and writing to the drive, I would create a BoilerPlate class where I would have those methods be accessed.

I don't know what is the right way to write code and break things down. I just strive to make the code clean, short, understandable and as abstract as possible.