Refactor NullClassBuilder#generate_class

[padi]

Seems that this can be refactored further.

[coveralls]

Coverage remained the same when pulling 5525994 on padi:initialize_base_class into 0869d4a on avdi:master.

[sferik]

This patch no longer applies cleanly to master. If you’re still interested in having it be applied, can you please rebase from master?

[padi]

@sferik sure. I'll look at it tomorrow and see if it still makes sense to apply this patch.

[padi]

@sferik oops. I still don't know anything about Rubocop, so I guess I'll just continue it tomorrow. The rspec tests pass though.

Edit: I get it now. The NullBuilderClass has 147 out of 144 allowable number of lines (enforced by rubocop.yml). Should I try to shorten the number of non-empty lines or should I increase the class lines to 147? Perhaps applying Replace Method w/ Method Object on the new method initialize_base_class?

[sferik]

@padi If you can find a way to refactor further that doesn’t increase the length of the class, that would be great. Otherwise, I think it’s okay to increase the RuboCop limit to 147 (or an even 150).

[padi]

@sferik it's ok now. Let me know what you think.

[sferik]

@padi I know I suggested refactoring to cut down the length of the NullClassBuilder class but it seems a bit excessive to have a class whose sole purpose is initializing another class.

@avdi Curious to get your thoughts on this.

[sferik]

When there is a class that only does one thing, I like to name that method call. This makes it easy to substitute with a proc, lambda, or Method object. There’s actually a RubyTapas episode about this, which describes why this can be useful. 😄