Build SQL queries in Swift. Extensible, protocol-based design that supports DQL, DML, and DDL.

The table below shows a list of SQLKit major releases alongside their compatible NIO and Swift versions.

Use the SPM string to easily include the dependency in your Package.swift file.

.package(url: "https://github.com/vapor/sql-kit.git", from: ...)

PostgresNIO supports the following platforms:

• Ubuntu 16.04+
• macOS 10.15+

SQLKit is an API for building and serializing SQL queries in Swift. SQLKit attempts to abstract away SQL dialect inconsistencies where possible allowing you to write queries that can run on multiple database flavors. Where abstraction is not possible, SQLKit provides powerful APIs for custom or dynamic behavior.

These database packages are built on SQLKit:

• vapor/postgres-kit: PostgreSQL
• vapor/mysql-kit: MySQL and MariaDB
• vapor/sqlite-kit: SQLite

SQLKit does not deal with creating or managing database connections itself. This package is focused entirely around building and serializing SQL queries. To connect to your SQL database, refer to your specific database package's documentation. Once you are connected to your database and have an instance of SQLDatabase, you are ready to continue.

Instances of SQLDatabase are capable of serializing and executing SQLExpression.

let db: SQLDatabase = ...
db.execute(sql: SQLExpression, onRow: (SQLRow) -> ())

SQLExpression is a protocol that represents a SQL query string and optional bind values. It can represent an entire SQL query or just a fragment.

SQLKit provides SQLExpressions for common queries like SELECT, UPDATE, INSERT, DELETE, CREATE TABLE, and more.

var select = SQLSelect()
select.columns = [...]
select.tables = [...]
select.predicate = ...

SQLDatabase can be used to create fluent query builders for most of these query types.

let planets = try db.select()
    .column("*")
    .from("planets")
    .where("name", .equal, "Earth")
    .all().wait()

You can execute a query builder by calling run().

For query builders that support returning results, like select(), there are additional methods for handling the database output.

• all(): Returns an array of rows.
• first(): Returns an optional row.
• run(_:): Accepts a closure that handles rows as they are returned.

Each of these methods returns SQLRow which has methods for access column values.

let row: SQLRow
let name = try row.decode(column: "name", as: String.self)
print(name) // String

SQLRow also supports decoding Codable models directly from a row.

struct Planet: Codable {
    var name: String
}

let planet = try row.decode(model: Planet.self)

Query builders that support returning results have convenience methods for automatically decoding models.

let planets = try db.select()
    ...
    .all(decoding: Planet.self).wait()

The select() method creates a SELECT query builder.

let planets = try db.select()
    .columns("id", "name")
    .from("planets")
    .where("name", .equal, "Earth")
    .all().wait()

This code would generate the following SQL:

SELECT id, name FROM planets WHERE name = ?

Notice that Encodable values are automatically bound as parameters instead of being serialized directly to the query.

The select builder has the following methods.

• columns
• from
• where (orWhere)
• limit
• offset
• groupBy
• having (orHaving)
• distinct
• for (lockingClause)
• join

By default, query components like where will be joined by AND. Methods prefixed with or exist for joining by OR.

builder.where("name", .equal, "Earth").orWhere("name", .equal, "Mars")

This code would generate the following SQL:

name = ? OR name = ?

where also supports creating grouped clauses.

builder.where("name", .notEqual, SQLLiteral.null).where {
    $0.where("name", .equal, SQLBind("Milky Way"))
        .orWhere("name", .equal, SQLBind("Andromeda"))
}

This code generates the following SQL:

name != NULL AND (name == ? OR name == ?)

The insert(into:) method creates an INSERT query builder.

try db.insert(into: "galaxies")
    .columns("id", "name")
    .values(SQLLiteral.default, SQLBind("Milky Way"))
    .values(SQLLiteral.default, SQLBind("Andromeda"))
    .run().wait()

This code would generate the following SQL:

INSERT INTO galaxies (id, name) VALUES (DEFAULT, ?) (DEFAULT, ?)

The insert builder also has a method for encoding a Codable type as a set of values.

struct Galaxy: Codable {
    var name: String
}

try builder.model(Galaxy(name: "Milky Way"))

The update(_:) method creates an UPDATE query builder.

try db.update("planets")
    .where("name", .equal, "Jpuiter")
    .set("name", to: "Jupiter")
    .run().wait()

This code generates the following SQL:

UPDATE planets SET name = ? WHERE name = ?

The update builder supports the same where and orWhere methods as the select builder.

The delete(from:) method creates a DELETE query builder.

try db.delete(from: "planets")
    .where("name", .equal, "Jupiter")
    .run().wait()

This code generates the following SQL:

DELETE FROM planets WHERE name = ?

The delete builder supports the same where and orWhere methods as the select builder.

The raw(_:) method allows for passing custom SQL query strings with support for parameterized binds.

let table = "planets"
let planets = try db.raw("SELECT * FROM \(raw: table) WHERE name = \(bind: planet)")
    .all().wait()

This code generates the following SQL:

SELECT * FROM planets WHERE name = ?

The \(bind:) interpolation should be used for any user input to avoid SQL injection.