I am trying to figure out the proper syntax for getting que开发者_JS百科ry result I need. Here is the scenario:
I am designing a SaaS application and have a table called trips, like so:
table.trips
- trip_id (pk, ai)
- client_id (fk)
- shop_id (fk)
- trip_date (date)
I need to be able to get the number of trips in the table for each day (by shop_id) and then deliver the number of TRIPS per trip_date, like:
DATE | TRIP SUM
--------------------------
01-01-2011 | 3
01-02-2011 | 2
01-03-2011 | 5
Any thoughts on the proper syntax?
SELECT COUNT(*), trip_date , shop_id
FROM trips
GROUP BY trip_date , shop_id
If the date column also stores hour and minute you will need to apply a substring function to the date, to only have the yyyy-mm-dd part of the date column, like this:
SELECT COUNT(*), substring(trip_date,1,10)
FROM trips
GROUP BY substring(trip_date,1,10);
Something like this?
select
DATE(trip_date) as [DATE],
count(1) as [TRIP SUM]
from
trips
group by
DATE(trip_date), shop_id
I'm not a mySQL guy, but you will need to use a T-SQL DatePart equivalent, where I have used the DATE()
function in order to strip the time portion of the date.
select trip_date,count(*)
from trips
where shop_id=[the shop id]
group by trip_date
This will work as long as the date is on the format you showed. If it has also time then you have to remove the time.
Also I do not know if the client_id has to come in in this query, if you have to group by it also then you have to put it along side trip_date on select or group. If you want to filter by it put it in the where.
SELECT shop_id, trip_date, count(trip_id)
FROM trips
GROUP BY shop_id, trip_date
精彩评论