home / skills / thebushidocollective / han / elixir-ecto-patterns

This skill helps you master Elixir Ecto patterns for schemas, changesets, queries, and transactions in robust database apps.

npx playbooks add skill thebushidocollective/han --skill elixir-ecto-patterns

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
8.4 KB
---
name: elixir-ecto-patterns
user-invocable: false
description: Use when Elixir Ecto patterns including schemas, changesets, queries, and transactions. Use when building database-driven Elixir applications.
allowed-tools:
  - Bash
  - Read
---

# Elixir Ecto Patterns

Master Ecto, Elixir's database wrapper and query generator. This skill
covers schemas, changesets, queries, associations, and transactions for
building robust database applications.

## Schema Definition

```elixir
defmodule MyApp.User do
  use Ecto.Schema
  import Ecto.Changeset

  schema "users" do
    field :name, :string
    field :email, :string
    field :age, :integer
    field :is_active, :boolean, default: true
    field :role, Ecto.Enum, values: [:user, :admin, :moderator]

    has_many :posts, MyApp.Post
    belongs_to :organization, MyApp.Organization

    timestamps()
  end

  def changeset(user, attrs) do
    user
    |> cast(attrs, [:name, :email, :age, :is_active, :role])
    |> validate_required([:name, :email])
    |> validate_format(:email, ~r/@/)
    |> validate_number(:age, greater_than: 0, less_than: 150)
    |> unique_constraint(:email)
  end
end
```

## Changeset Validations

```elixir
defmodule MyApp.Post do
  use Ecto.Schema
  import Ecto.Changeset

  schema "posts" do
    field :title, :string
    field :body, :text
    field :published, :boolean, default: false
    field :tags, {:array, :string}, default: []

    belongs_to :user, MyApp.User

    timestamps()
  end

  def changeset(post, attrs) do
    post
    |> cast(attrs, [:title, :body, :published, :tags, :user_id])
    |> validate_required([:title, :body, :user_id])
    |> validate_length(:title, min: 3, max: 100)
    |> validate_length(:body, min: 10)
    |> foreign_key_constraint(:user_id)
  end

  def publish_changeset(post) do
    post
    |> change(published: true)
  end
end
```

## Basic Queries

```elixir
import Ecto.Query

# Get all users
Repo.all(User)

# Get user by ID
Repo.get(User, 1)
Repo.get!(User, 1)  # Raises if not found

# Get by specific field
Repo.get_by(User, email: "[email protected]")

# Filter with where clause
query = from u in User, where: u.age > 18
Repo.all(query)

# Select specific fields
query = from u in User, select: {u.id, u.name}
Repo.all(query)

# Order results
query = from u in User, order_by: [desc: u.inserted_at]
Repo.all(query)

# Limit and offset
query = from u in User, limit: 10, offset: 20
Repo.all(query)
```

## Complex Queries

```elixir
# Combining multiple conditions
query =
  from u in User,
    where: u.is_active == true,
    where: u.age >= 18,
    order_by: [desc: u.inserted_at],
    limit: 10

Repo.all(query)

# Using pipe syntax
User
|> where([u], u.is_active == true)
|> where([u], u.age >= 18)
|> order_by([u], desc: u.inserted_at)
|> limit(10)
|> Repo.all()

# Dynamic queries
def filter_users(params) do
  User
  |> filter_by_name(params["name"])
  |> filter_by_age(params["min_age"])
  |> Repo.all()
end

defp filter_by_name(query, nil), do: query
defp filter_by_name(query, name) do
  where(query, [u], ilike(u.name, ^"%#{name}%"))
end

defp filter_by_age(query, nil), do: query
defp filter_by_age(query, min_age) do
  where(query, [u], u.age >= ^min_age)
end
```

## Associations and Preloading

```elixir
# Preload associations
user = Repo.get(User, 1) |> Repo.preload(:posts)

# Preload nested associations
user = Repo.get(User, 1) |> Repo.preload([posts: :comments])

# Query with preload
query = from u in User, preload: [:posts, :organization]
Repo.all(query)

# Custom preload query
posts_query = from p in Post, where: p.published == true

query = from u in User, preload: [posts: ^posts_query]
Repo.all(query)

# Join and preload
query =
  from u in User,
    join: p in assoc(u, :posts),
    where: p.published == true,
    preload: [posts: p]

Repo.all(query)
```

## Aggregations and Grouping

```elixir
# Count records
Repo.aggregate(User, :count)

# Count with condition
query = from u in User, where: u.is_active == true
Repo.aggregate(query, :count)

# Other aggregations
Repo.aggregate(User, :avg, :age)
Repo.aggregate(User, :sum, :age)
Repo.aggregate(User, :max, :age)

# Group by
query =
  from u in User,
    group_by: u.role,
    select: {u.role, count(u.id)}

Repo.all(query)

# Group with having
query =
  from u in User,
    group_by: u.role,
    having: count(u.id) > 5,
    select: {u.role, count(u.id)}

Repo.all(query)
```

## Inserting and Updating

```elixir
# Insert with changeset
attrs = %{name: "John", email: "[email protected]", age: 30}

%User{}
|> User.changeset(attrs)
|> Repo.insert()

# Insert without changeset
Repo.insert(%User{name: "Jane", email: "[email protected]"})

# Update
user = Repo.get(User, 1)

user
|> User.changeset(%{age: 31})
|> Repo.update()

# Update all
query = from u in User, where: u.is_active == false
Repo.update_all(query, set: [is_active: true])

# Delete
user = Repo.get(User, 1)
Repo.delete(user)

# Delete all
query = from u in User, where: u.is_active == false
Repo.delete_all(query)
```

## Transactions

```elixir
# Basic transaction
Repo.transaction(fn ->
  user = Repo.insert!(%User{name: "Alice"})
  Repo.insert!(%Post{title: "First post", user_id: user.id})
end)

# Multi for complex transactions
alias Ecto.Multi

Multi.new()
|> Multi.insert(:user, User.changeset(%User{}, user_attrs))
|> Multi.insert(:post, fn %{user: user} ->
  Post.changeset(%Post{}, Map.put(post_attrs, :user_id, user.id))
end)
|> Multi.run(:send_email, fn _repo, %{user: user} ->
  send_welcome_email(user)
end)
|> Repo.transaction()
```

## Embedded Schemas

```elixir
defmodule MyApp.Address do
  use Ecto.Schema
  import Ecto.Changeset

  embedded_schema do
    field :street, :string
    field :city, :string
    field :state, :string
    field :zip, :string
  end

  def changeset(address, attrs) do
    address
    |> cast(attrs, [:street, :city, :state, :zip])
    |> validate_required([:city, :state])
  end
end

defmodule MyApp.User do
  use Ecto.Schema
  import Ecto.Changeset

  schema "users" do
    field :name, :string
    embeds_one :address, MyApp.Address

    timestamps()
  end

  def changeset(user, attrs) do
    user
    |> cast(attrs, [:name])
    |> cast_embed(:address, required: true)
  end
end
```

## Custom Ecto Types

```elixir
defmodule MyApp.Encrypted do
  use Ecto.Type

  def type, do: :binary

  def cast(value) when is_binary(value), do: {:ok, value}
  def cast(_), do: :error

  def dump(value) when is_binary(value) do
    {:ok, encrypt(value)}
  end

  def load(value) when is_binary(value) do
    {:ok, decrypt(value)}
  end

  defp encrypt(value) do
    # Encryption logic
    value
  end

  defp decrypt(value) do
    # Decryption logic
    value
  end
end

# Usage in schema
schema "users" do
  field :secret, MyApp.Encrypted
end
```

## When to Use This Skill

Use elixir-ecto-patterns when you need to:

- Build database-backed Elixir applications
- Define schemas and data models with validations
- Write complex database queries with Ecto's DSL
- Manage database relationships and associations
- Handle data transformations with changesets
- Implement transactions for data consistency
- Work with PostgreSQL, MySQL, or other databases
- Build Phoenix applications with database access
- Create robust data validation layers

## Best Practices

- Always use changesets for data validation
- Preload associations to avoid N+1 queries
- Use transactions for multi-step database operations
- Leverage Ecto.Multi for complex transaction logic
- Keep schemas focused and avoid god objects
- Use virtual fields for computed or temporary data
- Index foreign keys and frequently queried fields
- Use fragments for complex SQL when needed
- Write composable query functions
- Test database constraints and validations

## Common Pitfalls

- Not preloading associations (N+1 query problem)
- Forgetting to validate required fields
- Not using transactions for related operations
- Hardcoding queries instead of composing them
- Ignoring database constraints in schemas
- Not handling changeset errors properly
- Overusing embedded schemas for relational data
- Missing indexes on foreign keys
- Not using Repo.transaction for multi-step operations
- Exposing raw Ecto queries in business logic

## Resources

- [Ecto Documentation](https://hexdocs.pm/ecto/)
- [Ecto Query Guide](https://hexdocs.pm/ecto/Ecto.Query.html)
- [Ecto Changeset](https://hexdocs.pm/ecto/Ecto.Changeset.html)
- [Phoenix with Ecto](https://hexdocs.pm/phoenix/ecto.html)
- [Ecto Best Practices](https://hexdocs.pm/ecto/getting-started.html)

Overview

This skill teaches practical patterns for using Ecto in Elixir applications, covering schemas, changesets, queries, associations, transactions, embedded schemas, and custom types. It focuses on building reliable, testable, and maintainable database layers for Phoenix or standalone Elixir apps. The content emphasizes real-world best practices and common pitfalls to avoid.

How this skill works

I provide actionable examples and idiomatic patterns: schema definitions with fields and associations, changeset construction and validations, basic and complex query composition, preloading strategies, aggregations, inserts/updates/deletes, transactions with Ecto.Multi, embedded schemas, and custom Ecto types. Each pattern includes sample code and guidance on when to use each technique to maintain data integrity and performance.

When to use it

  • When designing database-backed Elixir or Phoenix applications and modeling domain data.
  • When you need robust parameter validation and transformation before database persistence.
  • When composing safe, performant queries and avoiding N+1 problems via preloads.
  • When coordinating multi-step data changes that must succeed or roll back together.
  • When integrating embedded data structures or custom serialization/encryption logic.

Best practices

  • Always validate and sanitize input using changesets before persisting data.
  • Preload associations intentionally to prevent N+1 queries and control eager loads.
  • Use Repo.transaction or Ecto.Multi for operations that must be atomic.
  • Keep schemas focused; use embedded schemas for nested value objects, not relationships.
  • Index frequently queried fields and foreign keys to improve query performance.

Example use cases

  • User and post management: schemas, associations, validations, and publish flows.
  • Search and filter endpoints: build composable, dynamic query builders for APIs.
  • Batched imports: validate rows with changesets and use transactions for consistency.
  • Multi-step onboarding: create user, profile, and welcome post via Ecto.Multi.
  • Encrypted fields: implement a custom Ecto.Type to transparently encrypt/decrypt sensitive data.

FAQ

Should I always use changesets for inserts and updates?

Yes. Changesets centralize validation and casting, reducing bugs and enforcing constraints before DB operations.

When should I use embeds vs associations?

Use embeds for nested value objects stored in the same table (JSON/binary). Use associations for separate relational entities with their own lifecycle and queries.