On Thu, May 30, 2013 at 11:13 PM, Goktug Gokdogan <[email protected]> wrote:

> It is currently inconsistent, it is not enforced,
>

Why was it stopped being enforced?  It used to run :ant_tests (IIRC) that
ran checkstyle, and TAP definitely caught style violations.  We never had
any trouble accepting internal contributions when I was there, as it would
get caught in the submit queue.


> most of GWT contributions are coming from Googlers who are not aware that
> GWT has different rules than all other Google3 code. Moving code from
> Google3 to GWT requires re-sorting.
>

Why is that any different than any other third-party product?  Plus one of
the first rules is to follow the style of the code around your change,
regardless of whether that matches the style guide.


> If you change the name of a function, you need to move the code to a new
> random location and on the review you can't see what really changed in the
> function or even if it has changed. I don't care much about internal stuff
> much but seeing an API in alphabetical order is as good as any other random
> order for the end users of the SDK.
> These are the similar reasons why none of other high profile libs (Guava,
> Guice, Dagger and JDK itself) enforce such rules.
>

I still prefer the way it is, and I think having one order (I don't
particularly care what that is) that is deterministic is better than
anything else, because the others always devolve into random orders as
people don't bother trying to organize it since it is already disorganized,
and it isn't brought up in reviews because there isn't any definitive
answer anyway.  Likewise for formatting rules.

-- 
John A. Tamplin

-- 
http://groups.google.com/group/Google-Web-Toolkit-Contributors
--- 
You received this message because you are subscribed to the Google Groups "GWT 
Contributors" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to