+1 to JodaTime, its better use it than develop our own On Fri, Nov 26, 2010 at 3:13 PM, Dan Haywood <[email protected]> wrote:
> > On 26/11/2010 13:43, Sabine Winkler wrote: > >> Hi Dan, >> >> my "TimePeriod" implementation is just for an example application. totally >> agree with you not to reinvent the wheel. I will have a look at the class >> "org.apache.isis.applib.value.Time" (seems to be unfinished) >> > Rob? > > > and work on an TimePeriod implementation. >> > But do look at either JodaTime or Eric Evan's timeandmoney library. I > think either of those give you a TimePeriod; all that is needed is an > implementation of ValueSemanticsProvider. > > > > >> some further questions: >> - how can I contribute new/refactored implementations? >> > Because you haven't yet been voted in as a committer, the thing to do is to > raise a ticket on JIRA, and attach a patch (doing so requires that you check > a checkbox granting relevant rights to ASF). > > > should I contact the owner of the package first? >> > > I'm not sure if it's "good form" to take of owners of packages, because > ultimately Isis is owned by its community. But certainly if you mail here > we can indicate who would likely be reviewing any patches to decide whether > to apply them. For the applib, it's probably gonna be either Rob or myself. > > > > - should implementation details and questions discussed on this list? >> > Oh yes, definitely! "If it didn't happen on the list, then it didn't > happen". > > >> cheers, >> Sabine >> > > Cheers > Dan > > >> Am 25.11.2010 23:50, schrieb Dan Haywood: >> >>> Hi Sabine, >>> I think that Rob archived these off a long time ago, and they've never >>> been reinstated. I'm not sure that there are any concrete plans to do work >>> on them right now (perhaps we should delete this directory.. it still exists >>> in contrib/initial/nakedobjects?) >>> >>> At any rate, my view on this is that, rather than reinvent the wheel, we >>> should provide value support for a nicely designed class such as JodaTime. >>> This can be done by implementing the ValueSemanticsProvider interface ... I >>> think there's coverage of it in my book. It's not that hard to do, and >>> would be a nice enhancement. >>> >>> Dan >>> >>> On 25/11/2010 21:23, Sabine Winkler wrote: >>> >>>> hi, >>>> >>>> may I ask what are the plans for the classes located in >>>> "old-valueholders"? I was looking for TimePeriod implementation that is >>>> only >>>> located in this folder. after all i implemented a simple one for my test >>>> environment using two fields of class "org.apache.isis.applib.value.Time" >>>> - >>>> following the class documentation there are open issues for that. Is >>>> someone >>>> working at this class at the moment? >>>> >>>> thx, >>>> sabine >>>> >>>> >>> >>> >> >> -- Regards, Alexander
