Even though the current form is pretty flexible and you can mix and match
the two, it doesn't mean it makes sense to do so. I'm having hard time
thinking about out a good use case that would make a single complex mix and
match more appealing than having a separate group of class replacements and
a codegen (AFAIK, there is not even single example in the GWT-SDK itself).
Actually separating the two can result in something much more
useful&powerful [*].

GWT may treat them similarly and call it "deferred binding" and that is
what it is from the compiler perspective. However IMHO, it looks like the
compiler detail inside the compiler leaks into user space when you think
them as the same thing.
In user space, conceptually they are different and always used in different
scenarios (think UiBinder vs DOMImpl).

[*] If we don't require GWT.create for class replacement, anyone can
replace any class for any purpose (e.g. browser compatibility, bug
patching), this is way more useful to than mix and match.

On Sat, Aug 10, 2013 at 8:15 AM, Thomas Broyer <[email protected]> wrote:

> There aren't really two concepts: you can mix and match generate-with and
> replace-with for the same type, and/or switch from one to the other over
> time.
>
> Actually, replace-with is only a shorthand for a generator returning a
> constant value (of a class that already exists).
>
> --
> 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.
>
>
>

-- 
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