PostgreSQL

The postgresql adapter for PostgreSQL wraps the github.com/lib/pq driver written by Blake Mizerany.

Before starting to read this detailed information, it is advisable that you take a look at the getting started page so you become acquainted with the basics of upper/db, and you can grasp concepts better.

Installation

Use go get to download and install the adapter:

go get github.com/upper/db/v4/adapter/postgresql

Setup

Import the postgresql package into your application:

// main.go
package main

import (
  "github.com/upper/db/v4/adapter/postgresql"
)

Define the postgresql.ConnectionURL struct:

// ConnectionURL defines the DSN attributes.
type ConnectionURL struct {
  User     string
  Password string
  Host     string
  Database string
  Options  map[string]string
}

Pass the postgresql.ConnectionURL value as argument to postgresql.Open() to establish a connection to the server, and create a database session:

settings = postgresql.ConnectionURL{
  ...
}

sess, err = postgresql.Open(settings)
...

Use the postgresql.ParseURL() function to convert a DSN into a postgresql.ConnectionURL:

// ParseURL parses a DSN into a ConnectionURL struct.
postgresql.ParseURL(dsn string) (ConnectionURL, error)

Once the connection is established, you can start performing operations on the database.

Example

In the following example, a table named 'birthday' consisting of two columns ('name' and 'born') will be created. Before starting, the table will be searched in the database and, in the event it already exists, it will be removed. Then, three rows will be inserted into the table and checked for accuracy. To this end, the database will be queried and the matches (insertions) will be printed to standard output.

The birthday table with the name and born columns is created with these SQL statements:

--' example.sql
DROP TABLE IF EXISTS "birthday";

CREATE TABLE "birthday" (
  "name" CHARACTER VARYING(50),
  "born" TIMESTAMP
);

The psql command line tool is used to run the statements in the upperio_tests database:

cat example.sql | PGPASSWORD=upperio psql -Uupperio upperio_tests

The rows are inserted into the birthday table. The database is queried for the insertions and is set to print them to standard output.

package main

import (
	"fmt"
	"log"
	"time"

	"github.com/upper/db/v4/adapter/postgresql"
)

var settings = postgresql.ConnectionURL{
	Database: `upperio_tests`, // Database name
	Host:     `localhost`,     // Server IP or name
	User:     `upperio`,       // Username
	Password: `upperio`,       // Password
}

type Birthday struct {
	// The 'name' column of the 'birthday' table
	// is mapped to the 'name' property.
	Name string `db:"name"`

	// The 'born' column of the 'birthday' table
	// is mapped to the 'born' property.
	Born time.Time `db:"born"`
}

func main() {

	// The database connection is attempted.
	sess, err := postgresql.Open(settings)
	if err != nil {
		log.Fatalf("db.Open(): %q\n", err)
	}
	defer sess.Close() // Closing the session is a good practice.

	// The 'birthday' table is referenced.
	birthdayCollection := sess.Collection("birthday")

	// Any rows that might have been added between the creation of
	// the table and the execution of this function are removed.
	err = birthdayCollection.Truncate()
	if err != nil {
		log.Fatalf("Truncate(): %q\n", err)
	}

	// Three rows are inserted into the 'Birthday' table.
	birthdayCollection.Insert(Birthday{
		Name: "Hayao Miyazaki",
		Born: time.Date(1941, time.January, 5, 0, 0, 0, 0, time.Local),
	})

	birthdayCollection.Insert(Birthday{
		Name: "Nobuo Uematsu",
		Born: time.Date(1959, time.March, 21, 0, 0, 0, 0, time.Local),
	})

	birthdayCollection.Insert(Birthday{
		Name: "Hironobu Sakaguchi",
		Born: time.Date(1962, time.November, 25, 0, 0, 0, 0, time.Local),
	})

	// The database is queried for the rows inserted.
	res := birthdayCollection.Find()

	// The 'birthdays' variable is filled with the results found.
	var birthdays []Birthday

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

	// The 'birthdays' variable is printed to stdout.
	for _, birthday := range birthdays {
		fmt.Printf("%s was born in %s.\n",
			birthday.Name,
			birthday.Born.Format("January 2, 2006"),
		)
	}
}

Compile the example and run it:

go run example.go

The output will be:

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

Adapter particularities

JSON Types

You can save and retrieve data when using JSON types. If you want to try this out, make sure the column type is JSONB and the field type is postgresql.JSONB:

package main

import (
  // ...
  "github.com/upper/db/v4/adapter/postgresql"
  // ...
)

type Person struct {
  ...
  Properties  postgresql.JSONB  `db:"properties"`
  Meta        postgresql.JSONB  `db:"meta"`
}

JSON types are supported on PostgreSQL 9.4+. In addition to these, the adapter features other custom types like postgresql.StringArray and postgresql.Int64Array.

SQL builder

Use the SQL builder for any complex SQL query:

q := sess.SQL().Select(
    "p.id",
    "p.title AD publication_title",
    "a.name AS artist_name",
  ).From("artists AS a", "publication AS p").
  Where("a.id = p.author_id")

var publications []Publication
if err = q.All(&publications); err != nil {
  log.Fatal(err)
}

Auto-incremental Keys (Serial)

If you want tables to generate a unique number automatically whenever a new record is inserted, you can use auto-incremental keys. In this case, the column must be defined as SERIAL.

For the ID to be returned by db.Collection.Insert(), the SERIAL column must be set as PRIMARY KEY too.

CREATE TABLE foo(
  id SERIAL PRIMARY KEY,
  title VARCHAR
);

Remember to use omitempty to specify that the ID field should be ignored if it has a zero value:

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

otherwise, an error will be returned.

Helper functions

Use db.Func to escape function names and arguments:

res = sess.Find().Select(db.Func("DISTINCT", "name"))

Use the db.Raw() function for strings that have to be interpreted literally:

res = sess.Find().Select(db.Raw("DISTINCT(name)"))

db.Raw can also be used as a condition argument, similarly to db.Cond.

Take the tour

Get the real upper/db experience, take the tour.