lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ryan McKinley <ryan...@gmail.com>
Subject Re: [spatial] Rethinking Cartesian Tiers implementation
Date Wed, 30 Jun 2010 13:40:06 GMT
>
> Here's my proposal:
>
> 1. I'll finish up SOLR-1568, which has a few useful things that are added to contrib/spatial
in the way of distance utils.
> 2. We deprecate contrib/spatial in 3.x.

I think this is reasonable -- if there is not anyone to support it, i
don't think it makes sense to keep around.  As long as it is in the
3.x branch, it can be resurrected where necessary

> 3. You start fresh with your stuff in modules/spatial on trunk and take the geohash stuff
and the utilities I added.
>
> The nice thing about this approach is you can get started now and we won't effect each
other's work and I will no longer be holding you up.
>

As we think about how the tier approach should work, it seems useful
to look at something like MGRS:
http://en.wikipedia.org/wiki/Military_grid_reference_system
http://en.wikipedia.org/wiki/File:MGRSgridSouthPole.png

This is a similar boxid naming convention, but has exceptions for the
poles.  It has the benefit that we could delegate to proj.4 (or
similar) for the calculations and the tokens would have a meaningful
use (in some context).

It might not make sense to use exactly as is since the northing and
easting are encoded sequentially -- for a range query it may make
sense to munge them together as it is currently implemented.

I hope to have more time to look into this, with luck I will -- but I
have been told that "hope" is just delayed disappointment.

ryan

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message