Fwd: Undirect index update

[Hilaire Fernandes]

Le lundi 18 mai 2009 à 10:35 -0500, Chris Muller a écrit :

> > I thought this message should be used with the object from
> > the Magma collection using the lexical terms as an index.
> 
> yes, it should be used with the whole object that is a member in the
> MagmaCollection.  Unfortunately, I don't know of any easy way other
> than you must find a way to #noteOldKeysFor:
> the-object-which-is-*indexed-by*-the-lexical term.

Ok, I prefer to double check.


> > These lexical terms design a classification, existing
> > independently
> 
> I hope there are at least one thousand different classification
> members for the index, otherwise you may end up with a lot of
> hash-collisions in the Magma index, reducing performance..

Oh, not one thousand members for an index, but from a few unit to
several tenths.
Hum, why should I have hash-collisions if my indexes produce unique
indexing values?

I was believing index value is used as a sort of key to a dictionary
whose values are the indexed objects (classification members) 


Best,

Hilaire
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Ceci est une partie de message
	=?ISO-8859-1?Q?num=E9riquement?= =?ISO-8859-1?Q?_sign=E9e?=
Url : http://lists.squeakfoundation.org/pipermail/magma/attachments/20090519/fcaa1876/attachment.pgp