[Date Prev][Date Next][Thread Prev][Thread Next][Author Index][Date Index][Thread Index]
Re: [zzdev] d.masterdim and slices
- To: Tuukka Hastrup <Tuukka.Hastrup@xxxxxx>
- Subject: Re: [zzdev] d.masterdim and slices
- From: Tuomas Lukka <lukka@xxxxxxxxxxx>
- Date: Tue, 25 Jul 2000 22:10:32 +0300 (EETDST)
- Cc: zzdev@xxxxxxxxxx
- In-reply-to: <20000725190100.D9836990@xxxxxxxxxxxxxxxxxxx>
On Tue, 25 Jul 2000, Tuukka Hastrup wrote:
> On 25 Jul, Tuomas Lukka wrote:
> > On Tue, 25 Jul 2000, Tuukka Hastrup wrote:
> >> On 25 Jul, Tuomas Lukka wrote:
> >> >> If s.0 is non-empty, then one would like to see the difference between
> >> >> getSpaceHome and getSliceHome(s.0) - if there's something to do with
> >> >> slices, why not with s.0 too?
> >> >
> >> > Sorry, now I don't understand... why can't the Space home be designated as
> >> > the home cell of slice 0?
> >> If there's (and it seems there is) some cases we want to distinguish
> >> between a slice and the compound place, then we should make the
> >> difference visible somehow.
> > The point is that s.0 is *always* there. So think of the compound place as
> > s.0 with other things (slices) loaded in. That's why s.0's home is the
> > whole hog's home.
> Still, if we were to do something to s.0, like export new default
> settings, then how do we distinguish between exporting s.0 and
> exporting the compound space as a whole? (Another example: you want
> the slice's d.masterdim rank or the space's d.masterdim rank)
> OK, s.0 is special. It might be it's not a problem if it has special and
> normal slice role at the same time. I guess we'll see as we go on.
Some points to think about:
- the default settings may well come as a slice that we attach to s.0.
Then it would be no problem to just get a new version of that.
- exporting s.0 will not export anything imported from other slices, like
e.g. those cells on dimension lists that come from the other slices