[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
[nebula-dev] Re: [compositetable] different suggestions
|
> > I think that undo/redo can be handled at the data binding tier. But
> > that's a *long* discussion. Will you be at EclipseCon? We could discuss
> > this more there.
>
> perfect! That heads me to the right direction when I'll be up to this
> feature.
> Unfortunately I cannot come there. I am living/working in europe and my
> employer is too small to be able to pay me that stay. :-(
No problem. Feel free to ask when you get there.
> > Yes. Everything should be just a simple SWT control.
>
> thanks! I dig further into your code and that's exactly my impression.
> Especially the fact that row- and header-object are just subclasses of
> composite and do not have any specialized interface. The fact that you
> chose to clone the rows by reflection (special Stark Trek Replicator tec.
> :-)) and did not choose a factory pattern pointed me to that direction,
> too (due to your usage for the visual editor, I guess).
Yep. :-)
> > The first build implementing this feature landed last night (v0.9.1). I
> > have heard that there are issues on Linux. My day job is on Win32 and I
> > decided to value early feedback from the community over having
> > everything "perfect" on all platforms before doing a release. :-) Also,
> > I don't have access to MacOS, so if someone on the list does and can
> > test there, I'd really appreciate it!
>
> I started working on my tableViewer replacement these days. I hope the
> project priorities allow me to stay on this track, so I'll be able to give
> you feedbacks! I use Windows in my project, too but I use ubuntu at home.
My main laptop is Windows. My desktop at home is Fedora 6. I *love* the new 3d desktop effects with Beryl and AIGLX. I no longer have Mac envy by any stretch. :-)
> I started on the column header yesterday and I found that it is sufficient
> to make GridRowLayout aware of childControl visibility. I added if-clauses
> and bound-calculation alternatives. I thought of implementing subclassed
> grid and header objects that delegate to columns. Should I submit a patch
> for the basic GridRowLayout or a GridRowLayout subclass (if my stuff is
> welcome)?
Yes, please. Please open a bug report and attach your patch there.
Best regards,
Dave Orme
