Re: is GiST still alive?

> *nod* I have tried this several times - it just doesn't work well,
> because the maps are too different.
>
> You could do something like this:
> . a table for each element type, fields being the attributes, plus
> the node id.
> . a table to tie everything together (parent_id, child_id,
> child_order, child_type).
> In theory you could even generate the DB schema from an XML schema and
> evaluate it with XPath-like expressions.

Although bits separated from a topic, I am developing XML database enviroment
based on PostgreSQL, XpSQL.
http://gborg.postgresql.org/project/xpsql/

The enviroment decompose XML documents into fragments and uses PostgreSQL
to store them in a set of relations.
Users do not have to know that they are stored as relations.

It provides a variety of means to access XML Documents;
(a) DOM functions to build and traverse XML Documents
(b) XPath engine to extract infomation from documents
(c) update functions to modify documents

Although the development is on going,
the performance is better than Apache Xindice.
#if you require, I'll send you the result after.

- points -
* node labeling is based on dewey decimal classification,
so updating performance is well.
* The index method is general (postgres) B-tree.

this is a programing image.
http://gborg.postgresql.org/project/xpsql/cvs/co.php/XpSQL/doc/USAGE?r=1.4

+-------------------------------------------------------------------+
Makoto Yui <yuin(at)bb(dot)din(dot)or(dot)jp>
Key fingerprint = 6462 E285 97D8 1323 40C4 F9E5 EB0F 9DE6 1713 219E
+-------------------------------------------------------------------+