Given that the GSOC Rails projects involve some ActiveRecord
refactorings, doesn't this type of improvement make sense?

--dwf

On Mar 8, 12:42 pm, Niels Ganser <[email protected]> wrote:
> I realize us Rubyists don't necessarily think the classic works in the
> field such as Fowler's "Refactoring" apply to our day-to-day coding in
> their entirety but in this case I fully agree with Fowler when he says
> "You don't decide to refactor, you refactor because you want to do
> something else, and refactoring helps you do that other thing".
>
> I myself have had my faire share of frustration when browsing around in
> the AR associations code and as anyone who has ever done more than a few
> hours of programming knows confusion always – and I mean always! – leads
> not only to more bugs but also hinders in adding features to the
> confusing code.
>
> I'd argue those "odd bugs"Adam mentioned have already been there. Many
> times. There is no *one* bug that shouts "I'm here because this code
> needs to be refactored!", it usually is much more subtle than this.
> Years of experience generally show though that the harder code is to
> understand, the more buggy it will be.
>
> Now *if* there is fantastic test coverage and *if* someone is willing
> and able (regarding skills and free time) to give it a shot, I say go
> for it: Fork, refactor, push, and show us!
>
> IMHO there's never a better time to refactor than right *now*. Besides
> yesterday of course..
>
> Best,
> Niels
>
> Pratik wrote:
> >> More importantly, the added complexity created by importing all of the
> >> collection logic and interface into a non-collection association class
> >> just adds to rigidity and potential for odd bugs in the future.
>
> > Let's do this refactoring when that actually happens.
>
> >> What concerns me most about this is that resistance to cleaning up
> >> code likely implies a lack of confidence in the test suite.
> >> Considering how core associations are to Rails, and the not
> >> insignificant amount of cruft in that code, there should be tests on
> >> associations like fat kids on an ice cream truck.  What are you
> >> concerned the changes will break, given that all the tests pass?
>
> > There is no lack of confidence. But I'm not very much in favour of
> > refactoring without any performance gains or any obscure bug fixes.
> > Refactoring like these makes almost all the relevant patches in LH
> > stale and screws up the history. I just don't think the patch in
> > question here is worth that.

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Ruby 
on Rails: Core" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/rubyonrails-core?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to