On Sun, 2009-06-21 at 01:02 -0400, Gene Amtower wrote:
> On Sat, 2009-06-20 at 21:25 -0400, Derrell Lipman wrote:
>
> > On Sat, Jun 20, 2009 at 6:50 PM, Gene Amtower <[email protected]>
> > wrote:
> >
> > Derrell,
> >
> > I just saw the new snippet in "The week in qooxdoo
> > (2009-06-19)" on avoiding excessive server calls when using
> > the Remote Table Model, and I applied it to my Remote Table
> > Model implementation code. Actually, I wondered why Qooxdoo
> > made multiple calls to the backend each time my application
> > updated the remote table, and this solved the issue. It
> > works like a charm and speeds things up considerably by
> > avoiding repetitive table updates.
> >
> > Will you be making a similar update to RpcExample in the
> > contrib section so that users who implement this for the
> > first time by copying RpcExample include this enhancement?
> > I can't imagine any reason to not have this feature coded as
> > standard practice for any Remote Table Model implementation,
> > but maybe you know of a good reason.
> >
> >
> >
> > Hi Gene,
> >
> > So, /me starts thinking (a dangerous thing, I know)... If this
> > snippet shows a good general procedure, then how about if we fix it
> > in the remote table model instead of having to kludge it in each
> > application? Would you please try reverting your changes which were
> > a result of seeing that snippet and applying the attached patch to
> > qx.ui.table.model.Remote. This should fix the problem at the proper
> > level. (Of course, the real proper solution is to redesign the table
> > so that it doesn't need to issue so many loadRowCount() requests...
> > but that's a much bigger job not currently scheduled.)
> >
> > Derrell
>
>
> Derrell,
>
> Uh-oh, thinking again, eh?
>
> In my code, I override this method anyway in my extended
> RemoteTableModel class. Doesn't that negate any updated code in the
> class method? I thought about suggesting a change to the class as
> well, but many users will need to override this method as done in the
> RpcExample code, so it probably should go into BOTH places rather than
> just the framework class, right?
>
> Of course, the overridden method in the example (and in my class)
> should probably use the same variable names and test conditions as the
> parent class. I thought you were the force behind this new snippet,
> but obviously not, as suggested by your response.
>
> (Upon closer inspection, I think you patched the public method, which
> calls the protected method that I override as long as the test
> conditions are met. Am I reading this correctly?)
>
> What do you think?
>
> Gene
Derrell,
I removed my changes from the snippet in my extended classes and added
the proposed patch to the qx.ui.table.model.Remote class. I then
cleared the app cache and rebuilt my app. The repeat calls have
returned, so it doesn't appear to have resolved the issue. Your new
code looked good to me, so I'm not sure why it's not blocking the repeat
calls to the backend.
Any ideas why?
Gene
------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference: $250.
Need another reason to go? 24-hour hacker lounge. Register today!
http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
qooxdoo-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/qooxdoo-devel
