I think as a more general rule, it would be best if the extended components set behaved exactly like the regular component set by default. This would ease entry into the component set for other developers.
On Mon, 21 Mar 2005 07:42:33 -0400, Sylvain Vieujot <[EMAIL PROTECTED]> wrote: > I agree too that preserveDataModel default value should be true. > > I discovered it was set to true in the tag a few days ago, while searching > why an x:dataTable didn't refresh when the back-end bean did refresh. > > I'll set the default to false if nobody has problems with that. > > Sylvain. > > > On Mon, 2005-03-21 at 11:56 +0100, Martin Marinschek wrote: > Well, that would astonish me as well. h:dataTable ought _not_ to serialize > its datamodel by default - only x:dataTable should do that, and x:dataTable > also only if preserveDataModel is set to true! regards, Martin On Sun, > 20 Mar 2005 21:44:55 -0600, Heath Borders <[EMAIL PROTECTED]> wrote: > > using the <x:datatable /> doesn't make you dependent on the MyFaces > > implementation. The components set is implementation independent. > > But, > I thought that the datatable's datamodels were NOT supposed to be > > serializeable. There was a big discussion about this a week or two > ago, > right? > > On Sun, 20 Mar 2005 21:48:29 +0100, Mathias Br�kelmann > > <[EMAIL PROTECTED]> wrote: > > h:datatable now serializes the datamodel > value during requests as a > > default setting. I think that this is quite > problematic since every one > > who is using h:datatable without referring > the special myfaces > > x:datatable must ensure that the values in the > datamodel where serializable. > > > > To minimize dependencies we are using > h:datatable until we need some > > special functions of myfaces. But now we > must refer to x:datatable just > > to disable the serialization of the > datamodel. > > > > -- > -Heath Borders-Wing > [EMAIL PROTECTED] > -- -Heath Borders-Wing [EMAIL PROTECTED]
