Storing and Querying GPS Coordinates Effectively

Question

I want to create a large database of GPS coordinates that can be queried by saying \"Return all coordinates that are within \'n\' metres of [this coordinate]\".

I need i

Answer 1

If you want to avoid a GIS extension, I adapted the functions from this post to postgres sql:

create or replace function change_in_lat(miles numeric)
returns double precision as $$
with v as (select
    3960.0 as earth_radius,
    180 / pi() as radians_to_degrees
) select ( miles / earth_radius ) * radians_to_degrees from v;
$$ language sql
returns null on null input;

create or replace function change_in_long(lat numeric, miles numeric)
returns double precision as $$
with v as (select
    3960.0 as earth_radius,
    pi() / 180 as degrees_to_radians,
    180 / pi() as radians_to_degrees
) select (
    miles / (earth_radius * cos(lat * degrees_to_radians))
    ) * radians_to_degrees from v;
$$ language sql
returns null on null input;

using those you can do some surrounding-square queries:

--find all "a"s within 25 miles of any "b"
select * from a join b on (
a.gpslat between
    b.gpslat - change_in_lat(25) and b.gpslat + change_in_lat(25)
and a.gpslong between
    b.gpslong - change_in_long(b.gpslat::numeric, 25)
    and b.gpslong + change_in_long(b.gpslat::numeric, 25)
);

if you used it frequently enough I'm sure turning the between statements into a single function would be easy. I never did any actual "within radius" queries with this though.

For anything more complicated, you'll probably want a GIS extension like other answers have said. PostGIS is good, but I found a lot of the gis-specific functions can be hard to get right, and unless you utilize bounding-box indices your spacial queries might take a day if your data set is large enough. But the tradeoff in complexity is definitely worth it for all the fancy things, like outputting your data in geojson format, etc.