Re: [Plastic-devs] spectrum load message

[Mark T. <m.b...@br...>]

On Thu, 14 Sep 2006, Isa Barbarisi wrote:

> Hi,
> I also agree with point 4, it seems to be the most flexible solution, 
> with just a couple of remarks.
> It should be clear in the standard that all the information from a SSAP 
> Servers has to be sent and no information should be lost, in order to 
> not lose important metadata.
> The other point is that we would prefer to have the keys of the map as a 
> combination of UCD and utype (both of them) so that if one of them is 
> not found (maybe because the SSAP server doesn't have it) we can make 
> the search on the other one. Eventually all the fields of the row (like 
> ucd, utype, name...) could be sent.

If they're all going in the same map I don't think that column name 
should be included as well.  It's probably reasonably safe to say
that there won't be (confusing) namespace clashes between UCD and
utype, but column names could be anything (and are unlikely to be
machine-intelligible), so it seems like asking for trouble to use
them as keys in the same map.

Mark

-- 
Mark Taylor   Astronomical Programmer   Physics, Bristol University, UK
m.b...@br... +44-117-928-8776 http://www.star.bris.ac.uk/~mbt/