MySQL adapter for

The mysql adapter for MySQL wraps the driver written by Julien Schmidt.

This adapter supports basic CRUD queries, transactions, simple join queries and raw SQL.


Use go get to download and install the adapter:

go get

Setting up database access

The mysql.ConnectionURL{} struct is defined like this:

// ConnectionURL implements a MySQL connection struct.
type ConnectionURL struct {
  User     string
  Password string
  Address  db.Address
  Database string
  Options  map[string]string

The db.Address interface can be satisfied by the db.Host(), db.HostPort() or db.Socket() functions.

Alternatively, a mysql.ParseURL() function is provided:

// ParseURL parses s into a ConnectionURL struct.
mysql.ParseURL(s string) (ConnectionURL, error)

You may use mysql.ConnectionURL as argument for db.Open().


To use this adapter, import and the packages.

# main.go
package main

import (

Then, you can use the db.Open() method to connect to a MySQL server:

var settings = mysql.ConnectionURL{
  Address:  db.Host("localhost"), // MySQL server IP or name.
  Database: "peanuts",            // Database name.
  User:     "cbrown",             // Optional user name.
  Password: "snoopy",             // Optional user password.

sess, err = db.Open(mysql.Adapter, settings)


The following SQL statement creates a table with “name” and “born” columns.

--' example.sql

CREATE TABLE birthday (
  `name` VARCHAR(50),
  `born` DATE

Use the mysql command line tool to create the birthday table on the upperio_tests database.

cat example.sql | mysql -uupperio -pupperio upperio_tests

The Go code below will add some rows to the “birthday” table and then will print the same rows that were inserted.

// example.go

package main

import (
  ""         // Imports the main db package.
  ""   // Imports the mysql adapter.

var settings = mysql.ConnectionURL{
  Database: `upperio_tests`,                            // Database name
  Address:   db.Socket(`/var/run/mysqld/mysqld.sock`),  // Using unix sockets.
  User:     `upperio`,                                  // Database username.
  Password: `upperio`,                                  // Database password.

type Birthday struct {
  // Maps the "Name" property to the "name" column of the "birthday" table.
  Name string `db:"name"`
  // Maps the "Born" property to the "born" column of the "birthday" table.
  Born time.Time `db:"born"`

func main() {

  // Attemping to establish a connection to the database.
  sess, err := db.Open(mysql.Adapter, settings)

  if err != nil {
    log.Fatalf("db.Open(): %q\n", err)

  // Remember to close the database session.
  defer sess.Close()

  // Pointing to the "birthday" table.
  birthdayCollection, err := sess.Collection("birthday")

  if err != nil {
    log.Fatalf("sess.Collection(): %q\n", err)

  // Attempt to remove existing rows (if any).
  err = birthdayCollection.Truncate()

  if err != nil {
    log.Fatalf("Truncate(): %q\n", err)

  // Inserting some rows into the "birthday" table.

    Name: "Hayao Miyazaki",
    Born: time.Date(1941, time.January, 5, 0, 0, 0, 0, time.UTC),

    Name: "Nobuo Uematsu",
    Born: time.Date(1959, time.March, 21, 0, 0, 0, 0, time.UTC),

    Name: "Hironobu Sakaguchi",
    Born: time.Date(1962, time.November, 25, 0, 0, 0, 0, time.UTC),

  // Let's query for the results we've just inserted.
  var res db.Result

  res = birthdayCollection.Find()

  var birthday []Birthday

  // Query all results and fill the birthday variable with them.
  err = res.All(&birthday)

  if err != nil {
    log.Fatalf("res.All(): %q\n", err)

  // Printing to stdout.
  for _, birthday := range birthday {
    fmt.Printf("%s was born in %s.\n", birthday.Name, birthday.Born.Format("January 2, 2006"))


Running the example above:

go run main.go

Expected output:

Hayao Miyazaki was born in January 5, 1941.
Nobuo Uematsu was born in March 21, 1959.
Hironobu Sakaguchi was born in November 25, 1962.

Unique adapter features

Simple JOIN queries

Querying from multiple tables is possible using db.Database.Collection(), just pass the name of all the tables separating them by commas. You can also use the AS keyword to define an alias that you could later use in conditions to refer to the original table.

var err error
var artistPublication db.Collection

// Querying from two tables.
artistPublication, err = sess.Collection(`artist AS a`, `publication AS p`)

if err != nil {

res := artistPublication.Find(
  // Use db.Raw{} to enclose statements that you'd like to pass without
  // filtering.
  db.Raw{` = p.author_id`},
  ``, // We defined "p" as an alias for "publication".
  `p.title as publication_title`, // The "AS" is recognized as column alias.
  db.Raw{` AS artist_name`},

type artistPublication_t struct {
  ID               int64  `db:"id"`
  PublicationTitle string `db:"publication_title"`
  ArtistName       string `db:"artist_name"`

all := []artistPublication_t{}

if err = res.All(&all); err != nil {

If you’re working with more than one collection, the first one you pass becomes your primary collection. Calls to db.Collection.Append(), db.Collection.Remove() and db.Collection.Update() will be performed on your primary collection.

Auto-incremental keys

If you want to use auto-increment keys with a MySQL database, you must define the column type as NOT NULL AUTO_INCREMENT, like this:

  title VARCHAR(255)

Also, you must provide the omitempty option in the db tag when defining the struct:

type Foo struct {
  ID    int64   `db:"id,omitempty"`
  Title string  `db:"title"`

In order for the ID to be returned by db.Collection.Append(), the primary key must be named “id”, this is a known limitation that will be fixed on future releases.


Sometimes you’ll need to run complex SQL queries with joins and database specific magic, there is an extra package sqlutil that you could use in this situation:

import ""

This is an example for sqlutil.FetchRows:

  var sess db.Database
  var rows *sql.Rows
  var err error
  var drv *sql.DB

  type publication_t struct {
    ID       int64  `db:"id,omitempty"`
    Title    string `db:"title"`
    AuthorID int64  `db:"author_id"`

  if sess, err = db.Open(Adapter, settings); err != nil {

  defer sess.Close()

  drv = sess.Driver().(*sql.DB)

  rows, err = drv.Query(`
      p.title AS publication_title, AS artist_name
      artist AS a,
      publication AS p
    WHERE = p.author_id

  if err != nil {

  var all []publication_t

  // Mapping to an array.
  if err = sqlutil.FetchRows(rows, &all); err != nil {

  if len(all) != 9 {
    t.Fatalf("Expecting some rows.")

You can also use sqlutil.FetchRow(*sql.Rows, interface{}) for mapping results obtained from sql.DB.Query() calls to a pointer of a single struct instead of a pointer to an array of structs. Please note that there is no support for sql.DB.QueryRow() and that you must provide a *sql.Rows value to both sqlutil.FetchRow() and sqlutil.FetchRows().

Using db.Raw and db.Func

If you need to provide a raw parameter for a method you can use the db.Raw type. Plese note that raw means that the specified value won’t be filtered:

res = sess.Find().Select(db.Raw{`DISTINCT(name)`})

db.Raw also works for condition values.

Another useful type that you could use to create an equivalent statement is db.Func:

res = sess.Find().Select(db.Func{`DISTINCT`, `name`})