> > Personally I'd prefer the separation over the new access > value. As long as 'anonymous' still works for access (with a > deprecation > warning) it should be ok IMO. >
Agreed. The new access values is 'ugly' -- it really isn't an access value. > > Tom, do you mind updating the xdoclet module to accommodate > this change? > > I can take a look at it if not, but the seems to be a fair > amount of > > ramp up time in making changes in the xdoclet code. I will let you > > know when I am about to commit these changes, so we can try to time > > the commits of this, and the xdoclet module at the same time. > > Yep, I can do that sometime during the next week, I think. Great! I do appreciate it. > > > > Btw, would it make sense to add a writeonly access type > (which isn't > > > queried, only inserted/updated) ? > > > > I can't see an immediate usecase for this, but it seems to > make sense. > > I would hate to limit someone's mapping by leaving this off, as it > > seems valid, and easy to implement. > > Image for instance an object where the date and time of the > last change shall be stored in the database (eg. for > tracing). If the application has no need for this info, there > could be a writeonly attribute backed by a getter method that > simply returns the current date and time. Yeah, this is a perfect example. A quasi-trigger attribute. I like it. It isn't tough at all to add -- I will do first in 1.0 while Armin finishes up the 1.1 merge, and then will merge it there. -Andrew > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
