TAP document implementation issues. Section 2.6

[Doug Tody]

Hi Pat -

Boolean would be the natural choice and it is unfortunate that ADQL
does not support this (but maybe there is some reason).  Given these
limitations and the use of 0/1 for the region functions my preference
would be to use 0/1 for the TAP_SCHEMA booleans as you suggest.

	- Doug


On Sat, 3 Oct 2009, Patrick Dowler wrote:

> On Thursday 01 October 2009 17:28:20 Patrick Dowler wrote:
> > On Thursday 01 October 2009 13:39:41 Douglas Tody wrote:
> > > In our (partial so far) implementation of the TAP schema here we
> > > created DBMS tables much like TAP_SCHEMA.tables and TAP_SCHEMA.columns.
> > > We did use boolean for std,primary,indexed in the native DBMS tables.
> > > I agree we should consider just using boolean in the TAP schema itself.
> > > This was an oversight I think.
> >
> > In clarifying this section I have made the datatypes of columns in the
> > TAP_SCHEMA tables explicit. They are basically VARCHAR (but no size
> >  specified) for everything except some in Columns: size is an INTEGER and
> >  principal, indexed, and std are BOOLEAN.
>
> Correcting myself:
>
> We can't actually do this since ADQL does not have a BOOLEAN type.
>
> We could follow ADQL region function style and use integer, with 1 meaning
> true and 0 meaning false (CONTAINS and INTERSECTS return 0 or 1).
>
> Alternatively, we could use CHAR(1) with values T and F.
>
> In either case, VOTable does have boolean so this is likely to cause
> confusion. The mapping table in sec 2.5 is bidirectional, but we cannot really
> specify a bidirectional mapping of boolean.
>
> Maybe TAP just cannot support VOTable boolean type?
>
> Arbitrary decision: I will go with INTEGER and 0 or 1 (for the flags in
> tap_schema.columns) and I will change the table in section 2.5 to say that
> VOTable boolean is not supported by TAP unless there is strong input to do
> something else (and very soon - see PS)
>
> thanks in advance for help/support/ideas,
>
> PS-My understanding is that TCG review has to start in the next couple of days
> in order to be completed before the interop... so I will need some input or
> will have to make an arbitrary decision in the next 24-36 hours. I also know
> some people are travelling to ADASS in Japan and are likely off-line.
>
> --
>
> Patrick Dowler
> Tel/Tél: (250) 363-0044
> Canadian Astronomy Data Centre
> National Research Council Canada
> 5071 West Saanich Road
> Victoria, BC V9E 2M7
>
> Centre canadien de donnees astronomiques
> Conseil national de recherches Canada
> 5071, chemin West Saanich
> Victoria (C.-B.) V9E 2M7
>