Consistency is one of wicket's strengths. My tiny vote for 1.4.4 -- Tony
On Sun, Dec 13, 2009 at 3:35 PM, Girts Ziemelis <[email protected]> wrote: > I also liked the behaviour - it made the code shorter, as I did not have to > mirror the component tree in both then and else branches. > I guess it is not a big deal, except for the testing headaches - this breaks > the code at runtime :( > I now, i know - I should have test cases covering all branches in the form > :( > > > On 12/13/2009 08:14 AM, Douglas Ferguson wrote: >> >> I did find the behavior handy, but it is easy to work around. >> >> D/ >> >> On Dec 12, 2009, at 11:12 PM, Igor Vaynberg wrote: >> >> >>> >>> i think you guys misunderstand. >>> >>> i believe what we are talking about here is the requirement for >>> presence of components *other* then the component specified by >>> enclosure's child attribute. >>> >>> essentially if i do this: >>> >>> add(new webmarkupcontainer("container").setvisible(false)); >>> and have this in my markup: >>> <div wicket:id="container"><div wicket:id="foo"/></div> >>> >>> wicket will not throw an error even though i never added the "foo" >>> component to my component hierarchy because as soon as it determins >>> that the container div is not visible it will skip over until the >>> closing tag. >>> >>> the enclosures, however, as of 1.4.4 *will* throw an error for *any* >>> missing child declared inside enclosure's markup *even though* the >>> enclosure has been determined as hidden. >>> >>> hope this clears it up somewhat >>> >>> -igor >>> >>> > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
