Wow, everyone. This has been a great discussion so far.
Based on what I'm hearing, there's some definite expertise in this
area by members of the list and sufficient interest to work on it to
some degree, so would it make sense then for me to start taking a
crack at this, check something into a little Google code project when
I have something to show, and report back when it's out there? Maybe
then, we could get a few folks involved who are interested to help out
with snags?
To be clear, I have some dedicated time to work on this over the next
1-2 weeks, so I have to do it either way. I just want to try and
collaborate with anyone who is interested and open up the solution for
everyone so it can mature and grow. Even if the end result were just
included as a sophisticated example, it sounds like it would be
helpful to a lot of people.
On Jan 14, 2009, at 10:38 AM, Vjekoslav Nesek wrote:
Tim Moore wrote:
On Jan 14, 2009, at 1:15 AM, Chris Chabot wrote:
Neither of those does drag-and-drop though, but compared to the
nitty-gritty
Easy to say maybe! :-)
It's trickier than you'd think, especially since iframes are
involved.
Right now it's still a work in progress, and we haven't yet been
able to get it working as smoothly as iGoogle. It's based on jQuery
UI's Sortable, and we've fallen a little bit behind the latest
version so it's a little quirky at times. The biggest problem,
however, is that when you move an iframe to a new DOM location, in
most browsers it reloads the contents of the frame.
...and after it reloads that iframe, rpc communication between a
container and gadget is broken. To fix it
there should be some changes to rpc.js to reinitialize rpc after
frame reload. After a lot of experimentation
I think that only reasonable way to implement it is to use custom
absolute layout. jQueryUI sortable
will not help you when a mouse goes over iframe while dragging it
and you louse mousemove events :)
BR,
Vjekoslav Nesek
