On 12/15/2014 03:45 PM, Alexander Korotkov wrote:
> On Mon, Dec 15, 2014 at 4:12 PM, Heikki Linnakangas <hlinnakangas@vmware.com
>> wrote:
>>
>> On 12/15/2014 08:37 AM, Michael Paquier wrote:
>>
>>> On Mon, Dec 15, 2014 at 3:09 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>>
>>>> Michael Paquier <michael.paquier@gmail.com> writes:
>>>>
>>>>> - Point to polygon distance operator
>>>>>
>>>> I looked at that briefly during the last fest, but was unsure whether it
>>>> was too entangled with the GiST patches that Heikki was looking at.
>>>>
>>> Recalling my memories of this morning, things are rather independent.
>>
>> Right. I also looked at it briefly, but I wasn't sure if we really want
>> it. AFAICT, no-one has actually asked for that operator, it was written
>> only to be an example of an operator that would benefit from the knn-gist
>> with recheck patch. If there is some other, real, use for the knn-gist with
>> recheck patch, then I'm OK with that, but otherwise it's dubious to add an
>> operator just so that it can then be made faster by another patch. That
>> said, it seems quite harmless, so might as well commit it.
>
> Lack of recheck is major limitation of KNN-GiST now. People are not asking
> for that because they don't know what is needed to implement exact KNN for
> PostGIS. Now they have to invent kluges like this:
>
> WITH closest_candidates AS (
> SELECT
> streets.gid,
> streets.name,
> streets.geom
> FROM
> nyc_streets streets
> ORDER BY
> streets.geom <->
> 'SRID=26918;POINT(583571.905921312 4506714.34119218)'::geometry
> LIMIT 100
> )
> SELECT gid, name
> FROM closest_candidates
> ORDER BY
> ST_Distance(
> geom,
> 'SRID=26918;POINT(583571.905921312 4506714.34119218)'::geometry
> )
> LIMIT 1;
>
> See blog posts:
> http://blog.light42.com/wordpress/?p=102
> http://workshops.boundlessgeo.com/postgis-intro/knn.html
Ugh. Ok, sold :-). I'll review...
- Heikki
