EmergencyEMERGENCY? Get 24/7 Help Now!

GROUP_CONCAT useful GROUP BY extension

 | September 4, 2006 |  Posted In: Insight for Developers

PREVIOUS POST
NEXT POST

(There is an updated version of this post here)

MySQL has useful extention to the GROUP BY operation: function GROUP_CONCAT:

GROUP_CONCAT(expr) – This function returns a string result with the concatenated non-NULL values from a group.

Where it can be useful?

For example to get PHP array without looping inside PHP:

Table:

Handling in PHP:
old way:

with group_concat:

This should work faster, as we remove loop from PHP to MySQL server side.

Also it can be handy to use result concatenated string as part of IN statement:

Sure, last example can be handled with one query with joins, but sometimes we need the temporary ids in clients code, for example to execute query on another server.

One more thing: you may want to add ORDER BY NULL statement after GROUP_BY to avoid
unnecessary sorting with filesort

PREVIOUS POST
NEXT POST
Vadim Tkachenko

Vadim Tkachenko co-founded Percona in 2006 and serves as its Chief Technology Officer. Vadim leads Percona Labs, which focuses on technology research and performance evaluations of Percona’s and third-party products. Percona Labs designs no-gimmick tests of hardware, filesystems, storage engines, and databases that surpass the standard performance and functionality scenario benchmarks. Vadim’s expertise in LAMP performance and multi-threaded programming help optimize MySQL and InnoDB internals to take full advantage of modern hardware. Oracle Corporation and its predecessors have incorporated Vadim’s source code patches into the mainstream MySQL and InnoDB products. He also co-authored the book High Performance MySQL: Optimization, Backups, and Replication 3rd Edition.

51 Comments

  • GROUP_CONCAT in MySQL…

    GROUP_CONCAT(expr) – This function returns a string result with the concatenated non-NULL values from a group.
    Where it can be useful?
    For example to get PHP array without looping inside PHP:
    CREATE TABLE services (
    id INT UNSIGNED NOT NULL,
    client_id …

  • A litte problem I had with the group_concat function was when selecting integers only I was getting a blob instead of a string to solve this I used a cast:

    SELECT GROUP_CONCAT(CAST(myInt as CHAR)) myInts FROM aTable;

  • I have created a table (with only 2 fields) with the following query

    1.CREATE TABLE users (
    2.id INT NOT NULL AUTO_INCREMENT PRIMARY KEY ,
    3.name VARCHAR( 20 ) NOT NULL
    4.) ENGINE = MYISAM ;”

    There are 20,000 users in this table with ids from 1 to 20,000
    On executing the following query
    Code: ( text )

    1. select group_concat(id separator ‘,’) from users

    returns only 283 ids separated with ‘,’

  • Hi, is it possible to use some kind of argument in group_concat that allows it to group 2 by 2? p.ex:
    SELECT id,client_id FROM services WHERE id = 3;
    +—-+———–+
    | id | client_id |
    +—-+———–+
    | 3 | 5 |
    | 3 | 6 |
    | 3 | 7 |
    | 3 | 8 |
    | 3 | 9 |
    | 3 | 10 |
    | 3 | 11 |
    +—-+———–+

    SELECT id,GROUP_CONCAT(client_id,2) FROM services WHERE id = 3 GROUP BY id;
    +—-+————————-+
    | id | GROUP_CONCAT(client_id) |
    +—-+————————-+
    | 3 | 5,6 |
    +—-+————————-+
    | 3 | 6,7 |
    +—-+————————-+
    | 3 | 7,8 |
    +—-+————————-+
    | 3 | 8,9 |
    +—-+————————-+
    | 3 | 9,10 |
    +—-+————————-+
    | 3 | 10,11 |
    +—-+————————-+
    Thanks!

  • A litte problem I had with the group_concat function was when selecting integers only I was getting a blob instead of a string to solve this I used a CONCAT:

    SELECT CONCAT(GROUP_CONCAT(myInt),”) myInts FROM aTable;

  • Don’t use!

    1) explode() can’t be faster than the loop

    2) GROUP BY slows the query

    Did you do benchmark, or you just guessed the results??

  • Hello !

    I’ve a little problem using group_concat and user variable.
    I tried to set a group_concat(id) result into a @var, and then call the same table where id in(@var)
    but mysql return only the first row :\ do u have any idea ?

    Here is the code :
    ——————
    CREATE TABLE IF NOT EXISTS user (
    id tinyint(3) unsigned NOT NULL AUTO_INCREMENT,
    nom varchar(255) NOT NULL,
    PRIMARY KEY (id)
    ) ENGINE=MyISAM DEFAULT CHARSET=utf8 AUTO_INCREMENT=10 ;
    INSERT INTO user (id, nom) VALUES (1, ‘flo’),(2, ‘seb’),(3, ‘julien’),(4, ‘julie’),(5, ‘aymeric’),(6, ‘renaud’),(7, ‘lucy’),(8, ‘charlotte’),(9, ‘aurelien’);

    mysql> SELECT @lid := CAST(GROUP_CONCAT(id) AS CHAR) FROM user WHERE nom in(‘renaud’, ‘julie’, ‘lucy’);
    +—————————————-+
    | @lid := CAST(GROUP_CONCAT(id) AS CHAR) |
    +—————————————-+
    | 4,6,7 |
    +—————————————-+
    1 row in set (0.00 sec)

    mysql> SELECT @lid;
    +——-+
    | @lid |
    +——-+
    | 4,6,7 |
    +——-+
    1 row in set (0.00 sec)

    mysql> SELECT * FROM user WHERE id IN(@lid);
    +—-+——-+
    | id | nom |
    +—-+——-+
    | 4 | julie |
    +—-+——-+
    1 row in set (0.00 sec)

    I know it is totally useless but it’s just a sample 😉

    Regards.

  • Mysql support saved my time 🙂

    mysql> SELECT * FROM user where FIND_IN_SET(id, @lid);
    +—-+———-+
    | id | nom |
    +—-+———-+
    | 4 | julie |
    | 6 | renaud |
    | 9 | aurelien |
    +—-+———-+
    3 rows in set (0.02 sec)

  • I have two tables – member_directory & categories – that have a many-to-many relationship. I am using a bridge table called member_categories.

    SELECT DISTINCT(md.id),GROUP_CONCAT(categories.category SEPARATOR ‘, ‘) AS category FROM member_directory md LEFT JOIN member_categories mc ON md.id = mc.member_id LEFT JOIN categories ON mc.category_id = categories.id GROUP BY md.name ORDER BY md.isfeatured DESC, md.name ASC LIMIT 0, 10
    ====================================================================
    id | category |
    ——————————————————————–
    951 | CONDOMINIUMS, VACATION RENTALS |
    1711 | VACATION RENTALS, REALTORS |
    1413 | VACATION RENTALS |
    1163 | BOATING/FISHING/MARINE |
    2034 | VACATION RENTALS |
    2240 | TRAVEL AGENCIES, VACATION RENTALS |
    2033 | APARTMENT/HOUSE RENTALS |
    2208 | APARTMENT/HOUSE RENTALS |
    2153 | HEARING AIDS, MEDICAL SERVICES/HOSPITALS/CLINICS, AUDIOLOGY |
    1002 | CONDOMINIUMS |
    ====================================================================

    If I add any conditions regarding the category, the GROUP_CONCAT starts returning only the category mentioned in the WHERE clause (I don’t think I explained that very well, but here is an example). Adding a WHERE clause to the query:
    SELECT DISTINCT(md.id), GROUP_CONCAT(categories.category SEPARATOR ‘, ‘) AS category FROM member_directory md LEFT JOIN member_categories mc ON md.id = mc.member_id LEFT JOIN categories ON mc.category_id = categories.id WHERE mc.category_id = 187 GROUP BY md.id ORDER BY md.id ASC, md.name ASC LIMIT 0, 10

    ====================================================================
    id | category |
    ——————————————————————–
    1417 | SHOPPING |
    1736 | SHOPPING |
    1768 | SHOPPING |
    2219 | SHOPPING |
    2403 | SHOPPING |
    2407 | SHOPPING |
    2426 | SHOPPING |
    2431 | SHOPPING |
    2456 | SHOPPING |
    2458 | SHOPPING |
    ====================================================================

    Is there any way I can get the desired result? I just need a query to get information from the member_directory and any categories it may belong to.

  • I was able to get what I needed by using a sub-query:

    SELECT DISTINCT(md.id), (SELECT GROUP_CONCAT(categories.category SEPARATOR ‘, ‘) AS category FROM member_categories LEFT JOIN categories ON member_categories.category_id = categories.id WHERE member_categories.member_id = md.id) AS category FROM member_directory md LEFT JOIN member_categories mc ON md.id = mc.member_id LEFT JOIN categories ON mc.category_id = categories.id WHERE mc.category_id IN (138) GROUP BY md.name ORDER BY md.isfeatured DESC, RAND(1283945655) LIMIT 0, 10
    ====================================================================
    id | category |
    ——————————————————————–
    2275 | VACATION RENTALS, MASSAGE THERAPY, CONDOMINIUMS |
    1388 | VACATION RENTALS |
    1839 | VACATION RENTALS, CONDOMINIUMS |
    1845 | CONDOMINIUMS, VACATION RENTALS |
    1808 | REALTORS, VACATION RENTALS |
    1899 | CONDOMINIUMS, VACATION RENTALS |
    951 | CONDOMINIUMS, VACATION RENTALS |
    1713 | VACATION RENTALS, CONDOMINIUMS |
    2240 | TRAVEL AGENCIES, VACATION RENTALS |
    2285 | VACATION RENTALS, CONDOMINIUMS |
    ====================================================================

    However, if anyone has any tips to make this better, please post! Thanks!

  • Details

    The result is truncated to the maximum length that is given by the group_concat_max_len system variable, which has a default value of 1024. The value can be set higher, although the effective maximum length of the return value is constrained by the value of max_allowed_packet. The syntax to change the value of group_concat_max_len at runtime is as follows, where val is an unsigned integer:

    SET [GLOBAL | SESSION] group_concat_max_len = val;

    Current values in Our server

    max_allowed_packet = 67107840
    group_concat_max_len = 1024

    Query:

    To how much extent ,I can increase this variable size .and what is the maximumm data limit of GROUP_CONCAT

  • When I am using in GROUP_CONCAT, It will list all the values associated wit the parent ID, But I need only few like limit in ordinary SQL statement. Anybody know how to do this in MySQL qeury?

    tblmovies
    MovieId MovieName
    1 The Take
    2 Fatal Contact
    3 Flashbacks of a Fool
    4 Teddy Bear
    5 Nanking

    tblgenres
    GenreId GenreName
    1 Action
    2 Horror
    3 Sci-Fi
    4 Documentary
    5 Drama
    6 History

    tblmovie_genres
    movieId GenreId
    1 2
    1 4
    1 5
    2 4
    4 5
    3 5
    5 6

    When I was selecting the Movie whith Id 1 GROUP_CONCAT will result Horror,Documentary, Drama ; But I want to get only 2 genres which is Horror,Documentary.

    Please somebody help me .

    Thanks.

  • thank bosss “@Murz, Jasan ”

    it work for me..

    @Murz, Jasan

    Alternatively you can use

    CONVERT(group_concat(myInt) USING utf8)

  • Hi, this is a useful tutorial but i need some more help. In my table the column I use the GROUP_CONCAT is a foreign key from another table. So, i have to display PC1,PC2,PC3 for example and no 2,3,4 which are the values of my table. I tried with LEFT JOIN but I didn’t get a result. I’m trying to make an extra query inorder to take those names but nothing yet. Can you halp me a little bit?

  • Thank you , Vadim Tkachenko
    i used group_concat in my query but i got i problem
    problem is

    mysql group_concat limit length is 1024 and i need to increase it ( i did it )

    (
    but is there is problem when i increase group_concat length to more than 1024
    )

    it work fine in localhost , but on the website , i got error -> need more premission to do this

  • I have a table like
    userid testid attempts grade
    2 1 6,4,9,2,7,5,3,8,1 20,55,56,67,80,66,33,34,52

    I want to get the attempts in order like
    userid testid attempts grade
    2 1 1,2,3,4,5,6,7,8,9 52,67,33,55,66,20,80,34,56
    And my query looks like
    SELECT li.userid, li.testid, group_concat(li.attempts AS att, group_concat(li.grade) FROM testattempts GROUP BY testid

    Could anyone help me with this?

  • Hi,
    Can we change the comma( , ) symbol to | symbol in group contact present in the mysql …

    appreciates a reply..

    Thank U

  • You can chanage separator as follows:

    mysql> SELECT *,
    -> GROUP_CONCAT(DISTINCT test_score
    -> ORDER BY test_score DESC SEPARATOR ‘ |’)
    -> FROM table_name

  • i have below loop
    for($x=$c;$x!=1;$x–){
    $sql1=”select l2tpCDR.iccid as ic, sum(l2tpCDR.data) as sdata from l2tpCDR inner join sim on sim.ICCID=l2tpCDR.iccid where StatementID=0 and sim.simID='” . $b[$x] . “‘ group by ic”;

    $result1=mysql_query($sql1);
    mysql_error();
    $row1=mysql_fetch_array($result1);
    }

    and i want to tot up all the sum(l2tpCDR.data) as sdata at the end .could you please help me with this?

  • @Andrey:
    GROUP_CONCAT is rather slower than a regular while( mysql_fetch_array() ) PHP loop. The example here is just for simplicity. But GROUP_CONCAT is really powerful, when it comes to get several records with several “subrecords” without getting too much data (duplicated) from related tables. For example:
    SELECT p.name, p.description, group_concat( pf.filename SEPARATOR ‘:’ ) FROM pages p LEFT JOIN pages_files pf USING( pages_id )
    will get all pages and to every page – all files assigned to it (separated by ‘:’). You can even try something like this:
    SELECT p.name, p.description, group_concat( concat( pf.description, ‘: ‘, pf.filename ) SEPARATOR ‘;’ ) FROM pages p LEFT JOIN pages_files pf USING( pages_id )
    to get same as last time, but with files and their descriptions (file and description separated by :, and different files assigned to the same page separated by 😉 in one query. With a little more work you can get a ready-to-use xml code for this kind of tree-structure, and if not for the group_concat limit – you would be able to construct a single SQL query to extract a ready-to-use XML document containing all pages and their assigned files with all the required attributes – it is up to you to use this grouping function for much more then replacing php while loop, as it’s possibilities are really huge.

  • I just wrote up a tiny example to demonstrate the group_concat function. I´m just posting this here since it might be useful for you : http://www.giombetti.com/2013/06/06/mysql-group_concat/

Leave a Reply

 
 

Percona’s widely read Percona Data Performance blog highlights our expertise in enterprise-class software, support, consulting and managed services solutions for both MySQL® and MongoDB® across traditional and cloud-based platforms. The decades of experience represented by our consultants is found daily in numerous and relevant blog posts.

Besides specific database help, the blog also provides notices on upcoming events and webinars.
Want to get weekly updates listing the latest blog posts? Subscribe to our blog now! Submit your email address below and we’ll send you an update every Friday at 1pm ET.

No, thank you. Please do not ask me again.