N+1 Query Detection

Identify and fix N+1 query anti-patterns in Ecto/Phoenix applications.

Iron Laws - Never Violate These

• Never access associations without preload - Always preload before Enum.map

• No Repo calls inside loops - Restructure to batch queries

• Preload at context boundary - Load associations in context, not controllers/views

• Use joins for filtering - Use join

• preload when filtering by association

Detection Patterns

Pattern 1: Enum.map with Repo

BAD: N+1 queries

users |> Enum.map(fn user -> Repo.get(Order, user.order_id) end)

GOOD: Single query with preload

users |> Repo.preload(:orders)

Pattern 2: Association Access Without Preload

BAD: Lazy loading triggers N queries

for user <- users do user.posts # Triggers query for each user! end

GOOD: Eager load first

users = Repo.all(User) |> Repo.preload(:posts) for user <- users do user.posts # Already loaded end

Pattern 3: Nested Association Access

BAD: N+1 for nested associations

user.posts |> Enum.map(fn post -> post.comments end)

GOOD: Nested preload

Repo.preload(user, posts: :comments)

Quick Detection Commands

Find Enum.map with Repo calls nearby

grep -B5 -A5 "Enum.map" lib/ -r --include="*.ex" | grep -A5 -B5 "Repo."

Find association access patterns

grep -r ".posts|.comments|.orders" lib/ --include="*.ex"

Find Repo calls in loops

grep -B3 "Repo.get|Repo.one" lib/ -r --include="*.ex" | grep -B3 "for|Enum"

Analysis Command

For a context module, run:

grep -n "Repo." lib/my_app/[context].ex

Then verify each query has appropriate preloads.

References

For detailed patterns, see:

• references/preload-patterns.md

• Efficient preloading strategies

• references/query-optimization.md

• Query batching techniques