Hi Chris,
My response to what you <CO> wrote me on Wednesday, February 5, 2003 at 2:39 PM:
Why not always append and get the next key with:
nextkey = v[-1].key + 1
or have you already rejected this?
--
/\_/\ Jerry McRae \|||/ mail safely: The Bat! 1.60q
(~o o~) (. .)
)'Y'( _______________-o00-(_)-00o-___________________________
( ) / If I had a nickel for every time Windows crashes... Oh yea, I do!
--Bill Gates
========== your original message below ==========
From: Chris Olds
To: [EMAIL PROTECTED]
on: 2:39:11 PM
re: [Metakit] Q: rowids?
wise words:
CO> I'm working on a calendering/PIM application, and I have several independent
CO> datasets I need to work with (all in the same MK storage). Can anyone give me
CO> a recipe for generating surrogate key values? Since I'm only allowing one
CO> opener for the storage, I suppose I could get the last key value used by
CO> sorting the view and giving new rows a key value greater than the last value
CO> used, but that doesn't feel like the best solution - I've spent years avoiding
CO> just this kind of thing in SQL...
CO> To make this concrete, here's how it would look in Python:
CO> import metakit as mk
CO> db = mk.storage()
CO> v = db.getas("contact[key:I,name:S,address:S,city:S,state:S,zip:I]")
CO> nextkey = 1
CO> if v:
CO> nextkey = v.sort(v.key)[-1]+1
CO> and then any inserts into the view would use the current value of nextkey and
CO> increment it (all of this is actually in a class that keeps all of the ugly
CO> stuff hidden, so the keeping track part isn't as bad as it looks here).
CO> I don't want to use the row numbers returned when I add a row, because then
CO> I'd be up a creek if I ever deleted a row (which I don't expect to do a lot
CO> of, but I don't want to preclude).
CO> So - is there a better way?
CO> /cco
CO> --
CO> GPG Key Fingerprint: B375 A4E7 752B DB8C 4359 852E C3CF BF64 379A E9B2
CO> Debian Project (http://www.debian.org)
CO> _______________________________________________
CO> metakit mailing list - [EMAIL PROTECTED]
CO> http://www.equi4.com/mailman/listinfo/metakit
_______________________________________________
metakit mailing list - [EMAIL PROTECTED]
http://www.equi4.com/mailman/listinfo/metakit
