Re: Universal B-tree

For research purpose, I think that expression index is a good idea. I just want to do a proof of concept.<br /><br
/>Theother issue is that my algorithm break a z-order interval into several intervals that represents the query box.
Howshould I create it without creating any overhead?<br /><br />Best regards,<br /><br />daniel<br /><br /><div
class="gmail_quote">2010/8/9Greg Stark <span dir="ltr"><<a
href="mailto:gsstark@mit.edu">gsstark@mit.edu</a>></span><br/><blockquote class="gmail_quote" style="margin: 0pt 0pt
0pt0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div class="im">On Mon, Aug 9, 2010 at 5:31
PM,Daniel Oliveira<br /> <<a href="mailto:danielmarquesoliveira@gmail.com">danielmarquesoliveira@gmail.com</a>>
wrote:<br/> > I don't need to change B-tree estructure. I just need integrate my encode<br /> > function that
transformsmultiple keys into one key by bit-interleaving and<br /> > to acess elements given several intervals
(rangesearch).<br /><br /></div>You could build a "expression" index on the function which will build<br /> a regular
btree.Then any range clause on a similar expression in the<br /> query would be able to use the index. Postgres is
capableof detecting<br /> multiple range clauses that can use one or more indexes and combine<br /> the results.  It
mightnot be perfect for your use case, being a<br /> general purpose tool.<br /><br /> I'm not sure how feasible it
wouldbe to implement a special purpose<br /> case that matches your needs. Perhaps a special index type and index<br />
accessmethod similar to gist "tsquery" data type. But that's a lot<br /> more work and a lot more C coding.<br /><br />
--<br/><font color="#888888">greg<br /></font></blockquote></div><br />