On Tuesday 04 November 2008 09:01:53 am Paul McNett wrote: > johnf wrote: > > Larry (who is not a member of dev) has presented code to eliminate > > recursive calls in dPermixin.update() on the user list. All the code > > does is to insure that only one self.Form.update() is active at one time. > > IOW before a new update() can be started the current one has to finish. > > > > Does anyone consider this a good proposal? Is it required? > > I think it sounds good, but haven't had a chance to check it out yet. > Not sure it's required, but if it doesn't hurt, and if it improves > performance then I think it is a slam-dunk. > > > I for one can recall when I first started that I called update() all over > > the place in my code. I was under the mistaken impression that it was > > required. It is still required with code that makes changes to the > > underlying datasets. > > I hardly ever call update(). If I do, it is only on the outermost > container that needs it, which is usually a single control or a handful > of controls. Why force all that work on the form when it isn't needed? > > Paul
Issue "update" came about from attempting to change the screen controls. In Larry's case he is using paged controls to display different views. In this case Larry's is displaying text controls that first show LBS and then Kilos depending on the selection. The problem arose because when he switched tabs (either kilos or pounds) on the form - the GUI was not getting updated. So of course the simple way was to use self.update(). And of course Larry's was using dynamic properties to trigger the changes. I have noticed that self.object.Fit() seems to work. Does Fit() call update()? Thanks for the OK. -- John Fabiani _______________________________________________ Post Messages to: [email protected] Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-dev Searchable Archives: http://leafe.com/archives/search/dabo-dev This message: http://leafe.com/archives/byMID/[EMAIL PROTECTED]
