EXAMPLE DOCUMENTATION

Subqueries (also known as nested queries) are queries nested within another query. Kysely provides several ways to work with subqueries.

In SELECT Clause

You can use subqueries in SELECT statements to fetch related data:

const result = await db
  .selectFrom("users")
  .select(({ eb, ref }) => [
    "id",
    "firstName",
    eb
      .selectFrom("orders")
      .select("count(*)")
      .where("userId", "=", ref("users.id"))
      .as("orderCount"),
  ])
  .execute();

In WHERE Clause

Subqueries can be used in WHERE clauses for filtering:

// Find users who have made at least one order
const usersWithOrders = await db
  .selectFrom("users")
  .selectAll()
  .where("id", "in", (eb) =>
    eb.selectFrom("orders").select("userId").distinct()
  )
  .execute();

// Find products with price higher than average
const expensiveProducts = await db
  .selectFrom("products")
  .selectAll()
  .where("price", ">", (eb) =>
    eb.selectFrom("products").select(({ fn }) => fn.avg("price").as("avgPrice"))
  )
  .execute();

In FROM Clause

You can use subqueries in the FROM clause to query derived tables:

const result = await db
  .selectFrom((eb) =>
    eb
      .selectFrom("orders")
      .select([
        "userId",
        eb.fn.count("id").as("orderCount"),
        eb.fn.sum("amount").as("totalAmount"),
      ])
      .groupBy("userId")
      .as("orderStats")
  )
  .innerJoin("users", "users.id", "orderStats.userId")
  .select([
    "users.firstName",
    "orderStats.orderCount",
    "orderStats.totalAmount",
  ])
  .execute();

EXISTS and NOT EXISTS

Check for the existence of related records:

const usersWithBigOrders = await db
  .selectFrom("users")
  .selectAll()
  .where(({ exists, eb }) =>
    exists(
      eb
        .selectFrom("orders")
        .select("id")
        .where("userId", "=", eb.ref("users.id"))
        .where("amount", ">", 1000)
    )
  )
  .execute();

Correlated Subqueries

Subqueries that reference the outer query:

const result = await db
  .selectFrom("employees")
  .select(({ eb, ref }) => [
    "id",
    "firstName",
    "salary",
    eb
      .selectFrom("employees as e2")
      .select("salary")
      .where("e2.departmentId", "=", ref("employees.departmentId"))
      .where("e2.salary", ">", ref("employees.salary"))
      .count()
      .as("higherSalaryCount"),
  ])
  .execute();

Best Practices

Use meaningful aliases for subqueries to improve readability
Consider performance implications - some subqueries might be better expressed as JOINs
Use type parameters with .as() to ensure type safety:

.as<{ orderCount: number }>("orderStats");

Break down complex subqueries into smaller, more manageable pieces using CTEs when possible

Common Gotchas

Remember to use ref() when referencing outer query columns
Subqueries in SELECT must return a single value unless used with IN/EXISTS
Correlated subqueries can impact performance on large datasets

Subqueries

On this page