LGTM. I have no objection to loosening up these assertions when a case like
this comes up. Others (e.g., input type='password') may have to be loosened
up at some point as well.

On Thu, Sep 3, 2009 at 3:01 AM, Thomas Broyer <[email protected]> wrote:

>
>
> On 2 sep, 19:56, [email protected] wrote:
> > Revision: 6073
> > Author: [email protected]
> > Date: Wed Sep  2 10:55:41 2009
> > Log: Removing an assertion that introduced a breaking change.
> >
> > Patch by: jlabanca
> > Review by: jgw (TBR)
> >
> > http://code.google.com/p/google-web-toolkit/source/detail?r=6073
> >
> > Modified:
> >   /trunk/user/src/com/google/gwt/user/client/ui/Button.java
> >
> > =======================================
> > --- /trunk/user/src/com/google/gwt/user/client/ui/Button.java   Wed Sep
>  2
> > 07:58:52 2009
> > +++ /trunk/user/src/com/google/gwt/user/client/ui/Button.java   Wed Sep
>  2
> > 10:55:41 2009
> > @@ -54,7 +54,6 @@
> >       assert Document.get().getBody().isOrHasChild(element);
> >
> >       Button button = new Button(element);
> > -    assert
> "button".equalsIgnoreCase(button.getButtonElement().getType());
> >
> >       // Mark it attached and remember it for cleanup.
> >       button.onAttach();
>
> As I said when posting the initial patch, that was something I
> was"unsure about" (and I explicitly pointed out that it was breaking
> backwards compat') http://gwt-code-reviews.appspot.com/61809
>
> ...but on the other hand, TextBox.wrap() has an assertion on type=text
> (actually in the TextBox(Element) ctor) which makes it unable to wrap
> an <input type=password> (use a PasswordTextBox for that) whereas
> PasswordTextBox doesn't add any particular method or behavior change
> to TextBox. This is a very similar situation to Button vs.
> SubmitButton and ResetButton.
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~----------~----~----~----~------~----~------~--~---

Reply via email to