Using MySQL
, I can do something like:
SELECT hobbies FROM peoples_hobbies WHERE person_id = 5;
My Output:
shopping
fishing
coding
but instead I just want 1 row, 1 col:
Expected Output:
shopping, fishing, coding
The reason is that I'm selecting multiple values from multiple tables, and after all the joins I've got a lot more rows than I'd like.
I've looked for a function on MySQL Doc and it doesn't look like the CONCAT
or CONCAT_WS
functions accept result sets.
So does anyone here know how to do this?
You can use GROUP_CONCAT
:
SELECT person_id,
GROUP_CONCAT(hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
As Ludwig stated in his comment, you can add the DISTINCT
operator to avoid duplicates:
SELECT person_id,
GROUP_CONCAT(DISTINCT hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
As Jan stated in their comment, you can also sort the values before imploding it using ORDER BY
:
SELECT person_id,
GROUP_CONCAT(hobbies ORDER BY hobbies ASC SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
As Dag stated in his comment, there is a 1024 byte limit on the result. To solve this, run this query before your query:
SET group_concat_max_len = 2048;
Of course, you can change 2048
according to your needs. To calculate and assign the value:
SET group_concat_max_len = CAST(
(SELECT SUM(LENGTH(hobbies)) + COUNT(*) * LENGTH(', ')
FROM peoples_hobbies
GROUP BY person_id) AS UNSIGNED);
Have a look at GROUP_CONCAT
if your MySQL version (4.1) supports it. See the documentation for more details.
It would look something like:
SELECT GROUP_CONCAT(hobbies SEPARATOR ', ')
FROM peoples_hobbies
WHERE person_id = 5
GROUP BY 'all';
group by 'all'
isn't necessary (moreover unwanted), because this assign to all rows string all
and then compare strings between these rows. Am I right?
Alternate syntax to concatenate multiple, individual rows
WARNING: This post will make you hungry.
Given:
I found myself wanting to select multiple, individual rows—instead of a group—and concatenate on a certain field.
Let's say you have a table of product ids and their names and prices:
+------------+--------------------+-------+
| product_id | name | price |
+------------+--------------------+-------+
| 13 | Double Double | 5 |
| 14 | Neapolitan Shake | 2 |
| 15 | Animal Style Fries | 3 |
| 16 | Root Beer | 2 |
| 17 | Lame T-Shirt | 15 |
+------------+--------------------+-------+
Then you have some fancy-schmancy ajax that lists these puppies off as checkboxes.
Your hungry-hippo user selects 13, 15, 16
. No dessert for her today...
Find:
A way to summarize your user's order in one line, with pure mysql.
Solution:
Use GROUP_CONCAT
with the the IN
clause:
mysql> SELECT GROUP_CONCAT(name SEPARATOR ' + ') AS order_summary FROM product WHERE product_id IN (13, 15, 16);
Which outputs:
+------------------------------------------------+
| order_summary |
+------------------------------------------------+
| Double Double + Animal Style Fries + Root Beer |
+------------------------------------------------+
Bonus Solution:
If you want the total price too, toss in SUM()
:
mysql> SELECT GROUP_CONCAT(name SEPARATOR ' + ') AS order_summary, SUM(price) AS total FROM product WHERE product_id IN (13, 15, 16);
+------------------------------------------------+-------+
| order_summary | total |
+------------------------------------------------+-------+
| Double Double + Animal Style Fries + Root Beer | 10 |
+------------------------------------------------+-------+
You can change the max length of the GROUP_CONCAT
value by setting the group_concat_max_len
parameter.
See details in the MySQL documantation.
In my case I had a row of Ids, and it was neccessary to cast it to char, otherwise, the result was encoded into binary format :
SELECT CAST(GROUP_CONCAT(field SEPARATOR ',') AS CHAR) FROM table
Use MySQL(5.6.13) session variable and assignment operator like the following
SELECT @logmsg := CONCAT_ws(',',@logmsg,items) FROM temp_SplitFields a;
then you can get
test1,test11
SELECT @logmsg := CONCAT_ws(',',@logmsg,description) from test
I had a more complicated query, and found that I had to use GROUP_CONCAT
in an outer query to get it to work:
Original Query:
SELECT DISTINCT userID
FROM event GROUP BY userID
HAVING count(distinct(cohort))=2);
Imploded:
SELECT GROUP_CONCAT(sub.userID SEPARATOR ', ')
FROM (SELECT DISTINCT userID FROM event
GROUP BY userID HAVING count(distinct(cohort))=2) as sub;
Hope this might help someone.
For somebody looking here how to use GROUP_CONCAT
with subquery - posting this example
SELECT i.*,
(SELECT GROUP_CONCAT(userid) FROM favourites f WHERE f.itemid = i.id) AS idlist
FROM items i
WHERE i.id = $someid
So GROUP_CONCAT
must be used inside the subquery, not wrapping it.
Try this:
DECLARE @Hobbies NVARCHAR(200) = ' '
SELECT @Hobbies = @Hobbies + hobbies + ',' FROM peoples_hobbies WHERE person_id = 5;
TL;DR;
set @sql='';
set @result='';
set @separator=' union \r\n';
SELECT
@sql:=concat('select ''',INFORMATION_SCHEMA.COLUMNS.COLUMN_NAME ,''' as col_name,',
INFORMATION_SCHEMA.COLUMNS.CHARACTER_MAXIMUM_LENGTH ,' as def_len ,' ,
'MAX(CHAR_LENGTH(',INFORMATION_SCHEMA.COLUMNS.COLUMN_NAME , '))as max_char_len',
' FROM ',
INFORMATION_SCHEMA.COLUMNS.TABLE_NAME
) as sql_piece, if(@result:=if(@result='',@sql,concat(@result,@separator,@sql)),'','') as dummy
FROM INFORMATION_SCHEMA.COLUMNS
WHERE
INFORMATION_SCHEMA.COLUMNS.DATA_TYPE like '%char%'
and INFORMATION_SCHEMA.COLUMNS.TABLE_SCHEMA='xxx'
and INFORMATION_SCHEMA.COLUMNS.TABLE_NAME='yyy';
select @result;
we have two way to concatenate columns in MySql
select concat(hobbies) as `Hobbies` from people_hobbies where 1
Or
select group_concat(hobbies) as `Hobbies` from people_hobbies where 1
It is late but will helpfull for those who are searching "concatenate multiple MySQL rows into one field using pivot table" :)
Query:
SELECT pm.id, pm.name, GROUP_CONCAT(c.name) as channel_names
FROM payment_methods pm
LEFT JOIN payment_methods_channels_pivot pmcp ON pmcp.payment_method_id = pm.id
LEFT JOIN channels c ON c.id = pmcp.channel_id
GROUP BY pm.id
Tables
payment_methods
id | name
1 | PayPal
channels
id | name
1 | Google
2 | Faceook
payment_methods_channels_pivot
payment_method_id | channel_id
1 | 1
1 | 2
Output:
https://i.stack.imgur.com/TCxjv.png
GROUP_CONCAT(c.name)
works which concatenating multiple rows into one column
Here, my intension was to apply string concatenation without using group_concat() function:
Set @concatHobbies = '';
SELECT TRIM(LEADING ', ' FROM T.hobbies ) FROM
(
select
Id, @concatHobbies := concat_ws(', ',@concatHobbies,hobbies) as hobbies
from peoples_hobbies
)T
Order by Id DESC
LIMIT 1
Here
select
Id, @concatHobbies := concat_ws(', ',@concatHobbies,hobbies) as hobbies
from peoples_hobbies
will return
Id hobbies
1 , shopping
2 , shopping, fishing
3 , shopping, fishing, coding
Now our expected result is at third position. So I am taking the Last row by using
Order by Id DESC
LIMIT 1
Then I am also removing the First ', ' from my string
TRIM(LEADING ', ' FROM T.hobbies )
In sql server use string_agg to pivot a row field values into a column:
select string_agg(field1, ', ') a FROM mytable
or
select string_agg(field1, ', ') within group (order by field1 dsc) a FROM mytable group by field2
Another interesting example in this case -
Following is the structure of the sample table people_hobbies
-
DESCRIBE people_hobbies;
+---------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+---------+--------------+------+-----+---------+----------------+
| id | int unsigned | NO | PRI | NULL | auto_increment |
| ppl_id | int unsigned | YES | MUL | NULL | |
| name | varchar(200) | YES | | NULL | |
| hby_id | int unsigned | YES | MUL | NULL | |
| hobbies | varchar(50) | YES | | NULL | |
+---------+--------------+------+-----+---------+----------------+
The table is populated as follows -
SELECT * FROM people_hobbies;
+----+--------+-----------------+--------+-----------+
| id | ppl_id | name | hby_id | hobbies |
+----+--------+-----------------+--------+-----------+
| 1 | 1 | Shriya Jain | 1 | reading |
| 2 | 4 | Shirley Setia | 4 | coding |
| 3 | 2 | Varsha Tripathi | 7 | gardening |
| 4 | 3 | Diya Ghosh | 2 | fishing |
| 5 | 4 | Shirley Setia | 3 | gaming |
| 6 | 1 | Shriya Jain | 6 | cycling |
| 7 | 2 | Varsha Tripathi | 1 | reading |
| 8 | 3 | Diya Ghosh | 5 | shopping |
| 9 | 3 | Diya Ghosh | 4 | coding |
| 10 | 4 | Shirley Setia | 1 | reading |
| 11 | 1 | Shriya Jain | 4 | coding |
| 12 | 1 | Shriya Jain | 3 | gaming |
| 13 | 4 | Shirley Setia | 2 | fishing |
| 14 | 4 | Shirley Setia | 7 | gardening |
| 15 | 2 | Varsha Tripathi | 3 | gaming |
| 16 | 2 | Varsha Tripathi | 2 | fishing |
| 17 | 1 | Shriya Jain | 5 | shopping |
| 18 | 1 | Shriya Jain | 7 | gardening |
| 19 | 3 | Diya Ghosh | 1 | reading |
| 20 | 4 | Shirley Setia | 5 | shopping |
+----+--------+-----------------+--------+-----------+
Now, a table hobby_list
is generated having the list of all people and a list of each person's hobbies with each hobby in a new line -
CREATE TABLE hobby_list AS
-> SELECT ppl_id, name,
-> GROUP_CONCAT(hobbies ORDER BY hby_id SEPARATOR "\n")
-> AS hobbies
-> FROM people_hobbies
-> GROUP BY ppl_id
-> ORDER BY ppl_id;
SELECT * FROM hobby_list;
https://i.stack.imgur.com/yBqzm.png
Use GROUP_CONCAT:
SELECT GROUP_CONCAT(hobbies) FROM peoples_hobbies WHERE person_id = 5;
Success story sharing
DISTINCT
parameter, you will not get any doubles.... GROUP_CONCAT(DISTINCT hobbies)