Implementing Mysql Joins. How to get results in same row? - mysql

Question

Link for Er Diagram:
I want to get data from the child tables.
Eg. I need location, equipments, language, service for all user_id. There are multiple values for location, equipment, language and service for a given user_id.
I tried the following code:
SELECT user_detail.user_id , user_detail.name , user_equipment.equipment , user_location.location , user_service.service , user_language.language
from user_detail
INNER JOIN user_equipment on user_detail.user_id = user_equipment.user_id
INNER JOIN user_location on user_detail.user_id = user_location.user_id
INNER JOIN user_service on user_detail.user_id = user_service.user_id
INNER JOIN user_language on user_detail.user_id = user_language.user_id;
Now the problem is, if there are three location for a user_id I get three row, is there any way I get all location in same row?

Answer 1

You are looking for this "GROUP_CONCAT", your query would look like this:
SELECT user_detail.user_id ,
user_detail.name,
GROUP_CONCAT(user_equipment.equipment SEPARATOR ',') as equipments,
GROUP_CONCAT(user_location.location SEPARATOR ',') as locations,
GROUP_CONCAT(user_service.service SEPARATOR ',') as services,
GROUP_CONCAT(user_language.language SEPARATOR ',') as languages
FROM user_detail
INNER JOIN user_equipment on user_detail.user_id = user_equipment.user_id
INNER JOIN user_location on user_detail.user_id = user_location.user_id
INNER JOIN user_service on user_detail.user_id = user_service.user_id
INNER JOIN user_language on user_detail.user_id = user_language.user_id;
GROUP BY user_detail.user_id

Answer 2

In situations where I can't use a GROUP_CONCAT call, I would do the select, then join them in some resulting code. I am unsure what application language you are using, but for something like php:
$select = $query(); //Returns the SELECT results
for($i = 0; $i < count($select); $i++){
for($a = 0; $a < count($select); $a++){ //Double loop to check the entire list
if($a !== !i){ //No point in selecting the same row and comparing it
$location_count = 0;
if($select[$a]['user_id'] === $select[$i]['user_id']){
$locationref = 'location_'.$location_count++;
$select[$i][$locationref] = $select[$a]['location'];
}
}
}
}
Just my thoughts - untested but the general principle is there.

Answer 3

You can reduce multiple lines to single lines for each user, by grouping by user_id. The GROUP_CONCAT function will enable you to see all values of location in the group - like so:
SELECT user_detail.user_id ,
user_detail.name ,
user_equipment.equipment ,
GROUP_CONCAT(user_location.location SEPARATOR ',') AS user_locations ,
user_service.service ,
user_language.language
from user_detail
INNER JOIN user_equipment on user_detail.user_id = user_equipment.user_id
INNER JOIN user_location on user_detail.user_id = user_location.user_id
INNER JOIN user_service on user_detail.user_id = user_service.user_id
INNER JOIN user_language on user_detail.user_id = user_language.user_id
GROUP BY user_detail.user_id;
Note that this assumes there will always be one (and only one) equipment, service and language record for each user - if there can be more than one of each then you should use GROUP_CONCAT for their values (with a DISTINCT qualifier), while if the user might not have corresponding records the join to the relevant table should be changed to a LEFT OUTER JOIN.