Erlang MySQL driver
Erlang Lua Perl Other
Clone or download
Pull request Compare This branch is 116 commits behind Eonblast:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

Emysql 0.3.0

Note: Automatic conversion to the encoding of the underlying MySQL server was removed in the 0.3.x branch. If you specify, e.g., utf-8 encoding, then the MySQL server will reject wrongly-encoded strings and data, but the driver will not perform any encoding by itself anymore.

It is now the driver callers responsibility to ensure that data is properly encoded. This change makes it possible to pass binary BLOB data to the MySQL server once again.

This is an Erlang MySQL driver, based on a rewrite at Electronic Arts. Easy to use, strong connection pooling, prepared statements & stored procedures. Optimized for a central node architecture and OLTP.

While you can use mysql via ODBC, you should see better performance when using a driver like Emysql. For samples and docs see below. Read the brief on choosing a package and about the history of the various MySQL drivers.

Emysql is a cleaner rewrite of erlang-mysql-driver, see History. This fork is a direct continuation of the original emysql with fixes, updates, more documentation and samples.

This is the master branch. Should you run into problems, please report them by opening an issue at github and try if they go away by checking out the 'stable' branch. Thank you.

Which fork/package should I use? Likely this one, but see Choosing.
Why are there so many? See History.
Who used this fork? Electronic Arts.
How do I ...? See Samples.
Hello ...? See Samples.





In most cases, especially for high performance and stability, this package, Emysql, will be the best choice. It was rewritten from the ground up to overcome fundamental issues of 'erlang-mysql-driver'. It also has some usable docs meanwhile.


If you are looking for the plain necessities, you should use the ejabberd mysql driver. It is simple, battle tested and stable. There are comprehensive instructions in the source comments.


For mnesia-style transactions, one of the multiple 'erlang-mysql-drivers' may suite you best. There are quite many branches of it out there, and they are based on the same project as the ejabberd driver. To learn more about out the differences between the drivers, see the mysql driver history.

Getting Emysql

$ git clone git:// Emysql


Hello World

This is a hello world program. Follow the three steps below to try it out.


run() ->


	emysql:add_pool(hello_pool, 1,
		"hello_username", "hello_password", "localhost", 3306,
		"hello_database", utf8),

		<<"INSERT INTO hello_table SET hello_text = 'Hello World!'">>),

    Result = emysql:execute(hello_pool,
		<<"select hello_text from hello_table">>),

	io:format("~n~p~n", [Result]).

We'll be coming back to this source to make it run on your machine in a minute. But let's look at the basic building blocks first:

Executing an SQL Statement

emysql:execute(my_pool, <<"SELECT * from mytable">>).

For the exact spec, see below, Usage. Regarding the 'pool', also see below.

Executing a Prepared Statement

emysql:prepare(my_stmt, <<"SELECT * from mytable WHERE id = ?">>).

emysql:execute(my_pool, my_stmt, [1]).

Executing Stored Procedures

emysql:execute(my_pool, <<"create procedure my_sp() begin select * from mytable; end">>).

emysql:execute(my_pool, <<"call my_sp();">>).

Result Record

-record(result_packet, {seq_num, field_list, rows, extra}).

Converting Row Data To Records

-record(foo, {bar, baz}).

Result = emysql:execute(pool1, <<"select bar, baz from foo">>).
Recs = emysql_util:as_record(Result, foo, record_info(fields, foo)).
Bars = [ || Foo <- Recs].

Adding a Connection to the Connection Pool

Emysql uses a sophisticated connection pooling mechanism.

emysql:add_pool(my_pool, 1, "myuser", "mypass", "myhost", 3306,
  "mydatabase", utf8).

Arbitrary post-connection start-up commands can be added as a list of binaries to the last emysql:add_pool/9 argument:

emysql:add_pool(my_pool, 1, "myuser", "mypass", "myhost", 3306,
  "mydatabase", utf8, [
      <<"SET TIME_ZONE='+00:00'">>,

Running Hello World

Let's run the hello world sample from above:

1. Build Emysql

Build, using make:

$ cd Emysql
$ make

Or use rebar:

$ cd Emysql
$ ./rebar compile

Both yield an option to install but this is not needed for the samples.

2. Make a Sample Database

For use in the above sample (and all of those below, too), create a local mysql database. You should have a mysql server installed and running:

$ mysql [-u<user> -p]
mysql> create database hello_database;
mysql> use hello_database;
mysql> create table hello_table (hello_text char(20));
mysql> grant all privileges on hello_database.* to hello_username@localhost identified by 'hello_password';

3. Paste & Run Hello

Be sure to have ./ebin in your Erlang path. The hello-world source as shown above already waits in the Emysql directory, as hello.erl. Just compile and run it:

$ erlc hello.erl
$ erl -pa ./ebin -s hello run -s init stop -noshell

That's it. If you need to blindly repeat that more often some time, you can also use

$ make hello

There are more sample programs:

More Samples

Sample programs are in ./samples.

To run the samples, create the database as listed above at localhost, and simply run the compile & run batches:

$ cd samples
$ ./a_hello
$ ./b_raw
$ ./c_rows_as_records
$ ./d_prepared_statement
$ ./e_stored_procedure
$ ./f_load_from_file

or (after building and the database, as explained above), start a_hello etc. manually along these lines:

$ make
$ cd samples
$ erlc a_hello.erl
$ erl -pa ../ebin -s a_hello run -s init stop -noshell


General Notes on using Emysql, including the actual specs:

Starting an Application

The Emysql driver is an Erlang gen-server, and, application.


Adding a Pool

% emysql:add_pool(PoolName, PoolSize, Username, Password, Host, Port, Database, Encoding) ->
%	 ok | {error, pool_already_exists}  
% PoolName = atom()  
% PoolSize = integer()  
% Username = string()  
% Password = string()  
% Host = string()  
% Port = integer()  
% Database = string()  
% Encoding = atom()  

emysql:add_pool(mypoolname, 1, "username", "mypassword", "localhost", 3306, "mydatabase", utf8).

More Record Types

-record(result_packet, {seq_num, field_list, rows, extra}).

-record(ok_packet, {seq_num, affected_rows, insert_id, status, warning_count, msg}).

-record(error_packet, {seq_num, code, msg}).

For other record types, see include/emysql.hrl.

Executing SQL Statements

% emysql:execute(PoolName, Statement) -> result_packet() | ok_packet() | error_packet()  
% PoolName = atom()  
% Statement = string() | binary()  

emysql:execute(mypoolname, <<"SELECT * from mytable">>).
# result_packet{field_list=[...], rows=[...]}

emysql:execute(mypoolname, <<"UPDATE mytable SET bar = 'baz' WHERE id = 1">>).
# ok_packet{affected_rows=1}

Executing Prepared Statements

% emysql:prepare(StmtName, Statement) -> ok  
% StmtName = atom()  
% Statement = binary() | string()  

emysql:prepare(my_stmt, <<"SELECT * from mytable WHERE id = ?">>).
# ok

% emysql:execute(PoolName, StmtName, Args) -> result_packet() | ok_packet() | error_packet()  
% StmtName = atom()  
% Args = [term()]  

emysql:execute(mypoolname, my_stmt, [1]).
#result_packet{field_list=[...], rows=[...]}

Executing Stored Procedures

% emysql:execute(PoolName, StmtName, Args) -> result_packet() | ok_packet() | error_packet()  
% StmtName = atom()  
% Args = [term()]  

	<<"create procedure sp_hello() begin select * from hello_table; end">>).

emysql:execute(hello_pool, <<"call sp_hello();">>).
                [[<<"Hello World!">>],[<<"Hello World!">>]],

Note that you are getting back a list of results here.

Converting Row Data To Records

% emysql_util:as_record(ResultPacket, RecordName, Fields) -> Result  
% ResultPacket = result_packet()  
% RecordName = atom() (the name of the record to generate)  
% Fields = [atom()] (the field names to generate for each record)  
% Result = [record()]  

-record(foo, {bar, baz, bat}).

fetch_foo() ->
   Result = emysql:execute(pool1, <<"select bar, baz, bat from foo">>),
   Recs = emysql_util:as_record(Result, foo, record_info(fields, foo)),
	  io:format("foo: ~p, ~p, ~p~n", [, Foo#foo.baz, Foo#foo.bat])
    end || Foo <- Recs].

Converting Row Data To JSON

% emysql_util:as_json(ResultPacket) -> Result
% Result = [json()]

Result = emysql:execute(pool1, <<"select bar, baz from foo">>),

JSON = emysql_util:as_json(Result).
#[[{<<"bar">>,<<"bar_value">>}, {<<"baz">>,<<"baz_value">>}], ...]

Note that you are getting back a list of erlang terms in accordance with EEP-18. For actual utf8 binary JSON string you will need external library like jsx or jiffy

Loading Data From a File

	<<"LOAD DATA INFILE 'hello.txt' INTO TABLE hello_table (hello_text)">>).

Note, you must grant:

grant file on *.* to hello_username@localhost identified by'hello_password';

You need to give LOCAL or an absolute path, else the file is expected in the database server root. To use the current directory:

{ok, Dir} = file:get_cwd(),
    list_to_binary("LOAD DATA INFILE '" ++ Dir ++
        "/hello.txt' INTO TABLE hello_table (hello_text)")).


Please add a Common Test suite if you are proposing a pull request!

Basic Tests

Some Common Tests (Unit Tests) have been added in the test folder. They have no significant coverage yet but can help to test the basics. They might also help you find trip ups in your system set up (environment and basics suites).

For the basic tests you only need the test database set up and a mysql server running, the same as described above for the samples:

$ mysql [-u<user> -p]
mysql> create database hello_database;
mysql> use hello_database;
mysql> create table hello_table (hello_text char(20));
mysql> grant all privileges on hello_database.* to hello_username@localhost identified by 'hello_password';

To run the basic tests, at the command line type:

make test

Some tests can take up to half a minute to finish on a slow machine.

These tests currently check access to the database (environment suite) the same functionality as the samples (basics suite) and for race conditions as described in issue #9. Thank you, Ransom!

Encoding Tests

Currently the main focus is on Unicode test cases and encoding conversions, in the suites utf8_SUITE, latin_SUITE, utf8_to_latindb_SUITE, latin_to_utf8db_SUITE. Especially the silent conversions of list strings to the appropriate binary format were a bit of a challenge.

For the encoding tests, please create these databases:

  create database hello_utf8_database character set utf8;
  use hello_utf8_database;
  create table hello_table (hello_text char(20));
  grant all privileges on hello_utf8_database.* to hello_username@localhost identified by 'hello_password';
  create database hello_latin1_database character set latin1;
  use hello_latin1_database;
  create table hello_table (hello_text char(20));
  grant all privileges on hello_latin1_database.* to hello_username@localhost identified by 'hello_password';

To run the encoding tests do:

make encoding-test

To run all tests (this includes issue tests, see below):

make all-test

You see the test results when opening test/index.html with a browser. It should look like this:

Test Results

All test runs in "test"

Test Name Label Test Run Started _Ok_ Failed Skipped
me.Emysql. basics_SUITE - Tue Dec 13 2011 04:17:29 7 0 0 (0/0) 0 ct@machine
me.Emysql. environment_SUITE - Tue Dec 13 2011 04:17:29 6 0 0 (0/0) 0 ct@machine
Total 13 0 0 (0/0) 0

Copyright (C) 2011 Open Telecom Platform
Updated: Tue Dec 13 2011 04:17:36

Issue Tests

Issue 20

There is a test to check on issue #20. For this test you need two databases like this:

$ mysql [-u<user> -p]
mysql> create database test1;
mysql> create database test2;
mysql> grant all privileges on test1.* to test@localhost identified by 'test';
mysql> grant all privileges on test2.* to test@localhost identified by 'test';
mysql> use test1;
mysql> CREATE TABLE `test` ( `a` int(11) NOT NULL );
mysql> use test2;
mysql> CREATE TABLE `test` ( `b` int(11) NOT NULL );

The test suite is test/pool_SUITE.erl. To run the test, use make:

make test20

Check the test results by opening test/index.html with a browser.

Issue 9

The probably rare but still bad races of issue #9 have been solved at long last, courtesy of Ransom Richardson. They are tested for as part of the standard test. You can also use:

make test9

Check the test results by opening test/index.html with a browser.

Util Tests

For this test you need database like this:

$ mysql [-u<user> -p]
create database test_database;
grant all privileges on test_database.* to test_username@localhost identified by 'test_password';
flush privileges;

The test suite is test/as_record_SUITE.erl. To run the test, use make:

make testutil

Check the test results by opening test/index.html with a browser.


Open Source Erlang MySQL driver efforts are a fractured matter. You may find yourself digging in the sources to find out about their relationships with each other - and which one to pick. Here is a brief history.

Yxa: The first Erlang MySQL driver, in ~270 lines of code, seems to have been written between 2001 and 2004 by Magnus Ahltorp at the Swedish Royal Institute of Technology. It exposes low level, blocking functions to talk 4.0 protocol with a MySQL server. In 2005 Fredrik Thulin brought the driver into its current modular form to use it for the the SIP proxy Yxa while working at the Stockholm University. It has three process layers: a high level, round-robin connection pooling module; then, middle-man, single-connection, blocking processes that do the bit-level wrangling with the MySQL protocol. This module, mysql_conn, can also be used as a single-connection, stand-alone driver. And below this, there are small, protocol-agnostic receiver processes that handle the socket communication with the server, no matter the contents. Fredrik implemented gen-server behavior, added logging, error messages, upgraded authentication, and thoroughly commented the source. This mysql driver is working, complete and stable since at least 2007, it is available as part of Yxa 1.0 (hosted on github). It has no support for transactions or stored procedures. It is the basis for the following two packages. Its basic modular division and general functionality were not changed but only enhanced and it had originally been agreed upon that the Yxa branch should receive and merge the contributions of the later forks upstream. Unfortunately, that did not come to pass.

ejabberd: In 2006, a fork of the Yxa driver was created by Mickael Remond at Process One to become part of the successful instant messaging server ejabberd (also hosted on github). It can be assumed to be as stable as the Yxa branch, didn't change a byte in the lowest level, but only slightly enhanced the higher level. The difference from the original Yxa branch consists mainly of added inspection functions that help using the query results, and of an independent adoption of the MySQL 4.1 client-server protocol. The original Yxa branch has meanwhile adopted EDoc comment format, which makes the sources look more different than they actually are. You can find a Jan 2011 diff between Yxa and the ejabberd version here, and one ignoring comments here. These two branches could be merged quite easily, probably without any change in functionality at all.

erlang-mysql-driver: The storied life of this branch began in 2006 when Yariv Sadan created a fork from the ejabberd branch, made it a standalone project, gave it the maximally generic name that stuck, and hosted it on Google Code. Before he moved on to work at Facebook, he added high-level handling of prepared statements and transactions, and at long last completed some loose ends with the connection pooling that had been known to be lagging since the Yxa version. There were changes both in the original Yxa and the ejabberd branch after the forking off that never made their way into this fork, but in all likelihood they must be minor. It is not always obvious if the changes in erlang-mysql-driver had reached their intended final form. The separation of the three process layers seems to have suffered and complicated enhancements as the highest layer module, mysql.erl, started to become entangled with the second, mysql_conn.erl. Maybe that had a part in why the upstream merge never happened. The main repository of this fork lay dormant since Oct '07 when in Feb '10, Dave Smith, the rebar guy, started making some updates and put them on github. The driver is now enjoying a couple of active forks that make a convincing case for the github Network graphs.

A parallel fork from Yariv's branch, not entangled with Dave's tree, is the one by Nick Gerakines. I suspect it could technically be the strongest of the erlang-mysql-driver forks, with a lot of clean up work by smart guys put in, although it is generally less well known. And much less forked. In the end, the branch was abandoned for Emysql. In all branches, documentation beyond the source comments remains lacking.

Emysql was created from scratch in 2009, specifically to achieve better stability and throughput. It was proposed and written by Jacob Vorreuter at Electronic Arts and deployed at Shawn Fanning's, a social network site for gamers. Initially, Nick Gerakines, Jacob's boss at EA, rewrote large parts of erlang-mysql-server to clean it up. But some fundamental problems remained and when half a year in, they still could not overcome performance and stability issues, Nick gave Jacob the green light to rewrite it from the ground up because they felt that, in Jacob's words, the Yxa branch had been touched by so many people that it had become more complicated than necessary. According to Jacob, Bill Warnecke helped in the early design and testing. They abandoned the separation into three process layers and pulled the socket handling and bit-parsing into one module, coupling the functionality into direct function calls. It looks like they borrowed some chore lines from Magnus but generally created a new, straightforward architecture focused on providing a high performance node. Not only can Emysql open multiple connections, but multiple pools of multiple connections each to multiple database servers, which makes for a strong central OLTP node. Jacob says that Emysql is pretty stable and ran without issues in production at EA. Nick remembers: "The primary way we used it was to have a single Erlang node be the MySQL communication point and allow a number of connected nodes to connect through it to MySQL. We wanted very high throughput with many pids across a grid connecting to it and needed the ability to have a number of MySQL connections open for connection pooling." Rupture was killed in the consolidations of 2009. But Shawn could probably keep the money and we the fond memory of Napster and now, the glistening Emysql.

Eonblast Emysql is a continuation fork of Jacob's work, including all his commits and adding docs, samples, fixes and extensions. Henning Diedrich, Vitaliy Batichko, Chris Rempel, Patrick Atambo, Joel Meyer, Erik Seres, Alexey Lebedeff, Logan Owen, Seven Du, Brendon Hogger, Bart van Deenen, Ransom Richardson and Qing Liang have contributed to this branch. Support for stored procedures has been added, remaining issues are being addressed and there is work going on to bring Mnesia-style transactions. The fork is still close to the original, which currently lies dormant, but has started to add features and possibly increased stability.

Fredrik, Nick and Jacob helped shedding light on the matter. Thank you very much! Errors and omissions are mine. Please let me know about any errors you may spot. Thanks. - Henning

Links and References


(+)maintained at the time of writing, Feb 2011.


  • decrementing pool size could close sockets that are in use
  • spawn individual conn_mgr gen_server processes for each pool
  • allow row results to be returned as binary


Copyright (c) 2009-2011 Bill Warnecke, Jacob Vorreuter, Henning Diedrich, Eonblast Corporation

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.