well, let's resume discussion whether Harmony must contain implementation of the endorsed specs of the same version as RI or may contain a newer version
Tim, could you please comment on that? Thanks, Mikhail 2007/4/26, Stuart Ballard <[EMAIL PROTECTED]>:
It's possible. I'm not sure I agree with doing so, though. From my (brief) reading of that page, just because they allow alternative implementations and they're standardized outside the JCP doesn't mean that the compatibility requirements are less. If you want to provide a japi status page of your own that excludes these packages, I can certainly provide guidance (and scripts) for how to do it. Stuart. On 4/26/07, Mikhail Loenko <[EMAIL PROTECTED]> wrote: > Hi Stuart > > resending in case you missed it on the lst > > thanks, > Mikhail > ---------- Forwarded message ---------- > From: Mikhail Loenko <[EMAIL PROTECTED]> > Date: 25.04.2007 15:15 > Subject: Re: [general] Removing "endorsed" packages from JAPI reports > To: [email protected] > > > 2007/4/25, Alexey Petrenko <[EMAIL PROTECTED]>: > > Yep, it is possible of course. > > I've added this task to my nearest-future-to-do-list but have not > > complete it yet. :) > > Alexey > > I was talking about regular statistics that we refer at > http://harmony.apache.org/subcomponents/classlibrary/status.html > My suggestion was to add one more stats: without "endorsed" packages > > The statistics itself is located at > http://www.kaffe.org/~stuart/... > > Not sure that there is write access there from your > nearest-future-to-do-list ;) > > Thanks, > Mikhail > > > > > I'll finish it this week. > > > > SY, Alexey > > > > 2007/4/25, Mikhail Loenko <[EMAIL PROTECTED]>: > > > Hi Stuart, > > > > > > As it was discussed in the thread I'm copying, we are free > > > to replace classes in the packages listed here [1] by "classes implementing > > > a more recent version of the API as defined by the appropriate > > > standards body" [2] > > > > > > Would it be possible to generate JAPI diffs not taking into account packages > > > listed in [1] so that we get clear picture about amount of API work remained? > > > > > > Thanks, > > > Mikhail > > > > > > [1] http://java.sun.com/j2se/1.5.0/docs/guide/standards/index.html > > > [2] http://java.sun.com/j2se/1.5.0/README.html > > > > > > > > > > > > 2007/1/23, Tim Ellison <[EMAIL PROTECTED]>: > > > > Alexey Petrenko wrote: > > > > > 2007/1/17, Geir Magnusson Jr. <[EMAIL PROTECTED]>: > > > > >> so what do we do? I'm hoping that we can find a way to achieve this > > > > >> w/o us having to have "shadow" or duplicate code here... > > > > > That would be great. Can you suggest something? > > > > > > > > As Alexey wrote, the level of CORBA support in the RI 5.0 [1] is > > > > predominantly 2.3.1 based. However, we are free to implement a later > > > > version of the Corba spec and still be compliant since the CORBA code is > > > > an "endorsed standard" (see [2]). > > > > > > > > In this case I suggest we support Yoko in their goal of implementing > > > > corba 2.4 (and ignore the JAPI diffs). > > > > > > > > [1] http://java.sun.com/j2se/1.5.0/docs/guide/idl/compliance.html > > > > [2] http://java.sun.com/j2se/1.5.0/README.html > > > > > > > > Regards, > > > > Tim > > > > > > > > > > > > > > -- http://sab39.netreach.com/
