Thanks, Isaac. I'll try and get to it today. On Sat, Aug 8, 2009 at 4:48 PM, Isaac Truett <[email protected]> wrote:
> Joel, > > Same fix, but with a unit test. > > - Isaac > > On Tue, Aug 4, 2009 at 9:25 PM, Isaac Truett<[email protected]> wrote: > > Joel, > > > > The patch is attached. I wanted to write you a JUnit test to go with > > it, but I'm having trouble even getting the existing tests to run. I > > did verify this fix manually. > > > > Thanks, > > Isaac > > > > On Tue, Aug 4, 2009 at 3:22 PM, Isaac Truett<[email protected]> wrote: > >> Thanks, Joel. I'll see if I can put something together tonight. > >> > >> On Tue, Aug 4, 2009 at 3:12 PM, Joel Webber<[email protected]> wrote: > >>> Sounds like a bug to me. It's hard to imagine how this behavior could > be > >>> considered useful. I would assume the appropriate behavior would be to > >>> center, but keep the top-[left right] on the screen, depending upon the > RTL > >>> mode. Can anyone see a problem with this? > >>> @Isaac: If you feel like writing up a patch, I'd be happy to review. > >>> > >>> On Tue, Aug 4, 2009 at 2:49 PM, Isaac Truett <[email protected]> > wrote: > >>>> > >>>> I've just noticed that when the content of a PopupPanel grows larger > >>>> then the browser window, center() can position the panel with a > >>>> negative top/left, making part of the panel unreachable (the window > >>>> won't scroll up or left anymore to see the off-screen portion). Is > >>>> this considered a "feature" of the center() method? If so, would > >>>> people be open to adding an overloaded center(boolean > >>>> dontGoOutOfBounds) that would keep the top and left from going > >>>> negative, assuming a better name for the argument could be found? > >>>> > >>>> - Isaac > >>>> > >>>> > >>> > >>> > >>> > >>> > >>> > >> > > > --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/Google-Web-Toolkit-Contributors -~----------~----~----~----~------~----~------~--~---
