Hi Pedro!
I've started examining the draft in some detail, but as it is taking time to do, thought a quick airing of my opinion might be a good idea.
The draft looks like a good start to me!
I was glad to see the presence of the <row value constructor>, which makes the standard more capacious and flexible. Also, the expansion of <join condition> to include <search condition>, which relieves a severe restriction on joins.
I notice Arnold Rots has also given some feedback. I'm really glad, because one of the areas that has worried me is the one pertaining to temporal (and interval) data types. I think this needs careful going over. I was uncertain (not being an astronomer) that the standard SQL/92 approach could be justified on grounds of usefulness and the amount of work it would take to include the standard set.
I know Region is still to be discussed. Can I make a plea that we have a proper syntax for it, and not the magic string that was included as the main Region parameter? This will be easier to implement and will force reasonable documentation, even if only as a result of having a syntax diagram.
More on the way later this week or next.
Regards
Jeff
-- Jeff Lusted tel: +44 (0)116 252 5358 Astrogrid Project mob: +44 (0)7973 492290 Dept Physics & Astronomy email: jl99-at-star.le.ac.uk University of Leicester web: http://www.astrogrid.orgReceived on 2007-02-14Z10:03:40