MySQL JOIN blindness - mysql

Question

This query
SELECT station_id, station_name,
COUNT(event_station) as `total_visit_count`
FROM taps AS t
JOIN event_stations AS s
ON t.event_station = s.station_id
WHERE s.event_id=6
GROUP BY s.station_id
ORDER BY s.station_id;
returns
+------------+--------------+-------------------+
| station_id | station_name | total_visit_count |
+------------+--------------+-------------------+
| 5 | Station one | 24 |
| 6 | Station two | 35 |
| 7 | St. Pancras | 34 |
+------------+--------------+-------------------+
which is just fine.
However, there are some stations in taps which have not been visited and I would like them to be shown with a total_visit_count of zer0.
+------------+--------------+-------------------+
| station_id | station_name | total_visit_count |
+------------+--------------+-------------------+
| 5 | Station one | 24 |
| 6 | Station two | 35 |
| 7 | St. Pancras | 34 |
| 8 | Station four | 0 |
+------------+--------------+-------------------+
How do I rewrite my query to to that? I imagine some kind of JOIN is required, but I can't quite see it :-(
[Update]
describe event_Stations;
+--------------+------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+--------------+------------+------+-----+---------+----------------+
| station_id | int(11) | NO | PRI | NULL | auto_increment |
| event_id | int(11) | NO | | NULL | |
| station_name | text | NO | | NULL | |
| allocated | tinyint(1) | NO | | 0 | |
+--------------+------------+------+-----+---------+----------------+
4 rows in set (0.20 sec)
describe taps;
+---------------+-----------+------+-----+-------------------+-------+
| Field | Type | Null | Key | Default | Extra |
+---------------+-----------+------+-----+-------------------+-------+
| tag_id | int(11) | NO | | NULL | |
| time_stamp | timestamp | NO | | CURRENT_TIMESTAMP | |
| event_station | int(11) | NO | | NULL | |
| device_id | text | YES | | NULL | |
| device_type | text | YES | | NULL | |
| event_id | int(11) | NO | | NULL | |
+---------------+-----------+------+-----+-------------------+-------+
6 rows in set (0.00 sec)
select * from event_stations where event_id=6;
+------------+----------+-----------------+-----------+
| station_id | event_id | station_name | allocated |
+------------+----------+-----------------+-----------+
| 5 | 6 | Station one | 0 |
| 6 | 6 | Station two | 0 |
| 7 | 6 | St. Pancras | 0 |
| 8 | 6 | Station three | 0 |
| 9 | 6 | Station four | 0 |
| 10 | 6 | Station five | 0 |
| 11 | 6 | Station six | 0 |
| 12 | 6 | Station seven | 0 |
| 13 | 6 | Station eight | 0 |
| 14 | 6 | Station nine | 0 |
| 15 | 6 | Station ten | 0 |
| 16 | 6 | Station eleven | 0 |
+------------+----------+-----------------+-----------+
12 rows in set (0.00 sec)

Answer 1

First, swap the order of your join, so the primary table is sorted first (this is for organizational purposes only).
Then, use a LEFT JOIN to accomplish what you're looking for. This will ensure you pull all event_stations records (the left portion of the join), even if there is no corresponding record in the taps table (the right portion of the join). In place of the missing taps, you'll get NULL values.
COUNT will ignore nulls in aggregate, so will only return the count of non-null records. Thus, it will return 0 for your missing event_stations records.
SELECT
station_id,
station_name,
COUNT(event_station) as `total_visit_count`
FROM event_stations AS s
LEFT JOIN taps AS t
ON t.event_station = s.station_id
WHERE s.event_id = 6
GROUP BY s.station_id
ORDER BY s.station_id;
Alternatively, you could just use a RIGHT JOIN with your original join order. I personally don't like doing that, though, because I'm a LTR reader (first in order is more important).