sqlite select minimum function value of grouped result

Question

I have a table with gps locations of restaurants chains and would like to return the addresses of the restaurants closest to point (A) within a certain radius

SELECT *
    , MIN(distance($lat, $lon, lat, lon)) as miles 
FROM all_restaurants 
WHERE lat between $lat1 and $lat2 
    AND lon between $lon1 and $lon2 
    AND miles < $miles 
GROUP BY restaurant_id 
ORDER BY miles ASC
    , company_name ASC 
LIMIT 500

returns error

misuse of aggregate: MIN()

Answer 1

Lamak was somewhat right but this is what i ended up doing instead

SELECT *, distance('+lat+','+lon+', lat, lon) as minmiles
FROM all_restaurants 
WHERE restaurant_id || minmiles IN 
( 
    SELECT restaurant_id || MIN(miles) as fewmiles 
    FROM ( 
        SELECT restaurant_id, distance('+lat+','+lon+', lat, lon) as miles 
        FROM restaurant_master_combined 
        WHERE lat BETWEEN $lat AND $lat2 AND lon BETWEEN $lon1 AND $lon2 AND miles < $miles
    ) 
    GROUP BY restaurant_id 
)
ORDER BY ROUND(minmiles) ASC, company_name ASC 

hope this helps somebody else

Answer 2

I think that your problem is that you only are grouping by restaurant_id. When you use an aggregation function as MIN, MAX, SUM, AVG, etc, you need to include every column in the SELECT statement that are not in any aggregation function. In this case you have two options, either in the SELECT you only put restaurant_id as the following:

SELECT restaurant_id 
        , MIN(distance($lat, $lon, lat, lon)) as miles 
    FROM all_restaurants 
    WHERE lat between $lat1 and $lat2 
        AND lon between $lon1 and $lon2 
        AND miles < $miles 
    GROUP BY restaurant_id 
    ORDER BY miles ASC
        , company_name ASC 
    LIMIT 500

Or you put every other column included in the '*' on the grouping (because you can't use GROUP BY *).

Answer 3

I bet that you're calling a function which is distance($lat, $lon, lat, lon), don't you?

Have you tried using a simple subtraction between $lat and lat and have it return the minimum value using the MIN function?

MIN's purpose is for use with a table field or some basic operation such as a subtraction, I guess. This might be the cause of the encountered error.

Up and foremost, avoid using * while using aggregation functions, list the columns you require in your query.

Perhaps providing us with sample data might help us find a way around to work you through.