Question

My table currently stands like this:

CREATE TABLE IF NOT EXISTS `x_geodata` (
  `post_id` int(11) NOT NULL,
  `post_type` varchar(20) NOT NULL,
  `lat` float(10,6) NOT NULL,
  `lng` float(10,6) NOT NULL,
  PRIMARY KEY  (`post_id`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;

How do I correctly and efficiently INDEX this table?

The only query ran on this table is as follows:

if(!empty($_SESSION['s_property_radius'])) {$dist = $_SESSION['s_property_radius'];}else{$dist = 50;}
$orig_lat = $_SESSION['s_property_address_lat'];
$orig_lon = $_SESSION['s_property_address_lng'];
$lon1 = $orig_lon - $dist / abs( cos( deg2rad( $orig_lat ) ) * 69 );
$lon2 = $orig_lon + $dist / abs( cos( deg2rad( $orig_lat ) ) * 69 );
$lat1 = $orig_lat - ( $dist / 69 );
$lat2 = $orig_lat + ( $dist / 69 );


$sql = "

SELECT `t`.`post_id`, 3956 * 2 * ASIN( SQRT( POWER( SIN( ( ".$orig_lat." - `t`.`lat` ) * pi() / 180 / 2), 2 ) + COS( ".$orig_lat." * pi() / 180) * COS( `t`.`lat` * pi() / 180 ) * POWER( SIN( ( ".$orig_lon." - `t`.`lng` ) * pi() / 180 / 2 ), 2 ) ) ) AS `distance` FROM (
    SELECT `post_id`, `lat`, `lng` FROM `x_geodata` WHERE `post_type` = 'some post type' AND `lng` BETWEEN '".$lon1."' AND '".$lon2."' AND `lat` BETWEEN '".$lat1."' AND '".$lat2."'
) AS `t` HAVING `distance` <= ".$dist."

";

The query checks to make sure we are looking at the correct post type and then does a square radius check on the lat and lng. The returned results are then run through a circular radius check.

What I’m looking for is updated CREATE TABLE SQL or UPDATE TABLE SQL to get this INDEXED correctly.

Any help is greatly appreciated.

EDIT:
Did an EXPLAIN on my query based on arnep answer and I got this:

id    select_type    table    type    possible_keys    key    key_len    ref    rows    Extra
1    PRIMARY    <derived2>    ALL    NULL    NULL    NULL    NULL    3   
2    DERIVED    zch_geodatastore    range    post_type    post_type    70    NULL    3    Using where

No idea what it means though…

Answer 1

As your inner SELECT uses post_type, lat and lon in WHERE clause I would recommend to put an index on those.

Use EXPLAIN [QUERY] to see if the index is used and what benefit you get from it.