prefetch_related

prefetch_related solves the N+1 problem for ManyToMany and reverse FK relationships using two queries instead of a JOIN.

The Problem

# N+1 queries!
authors = await Author.objects.all()
for author in authors:
    posts = await author.posts.all()  # Query per author!

The Solution

from ryx.relations import apply_prefetch_related

# 2 queries total
authors = await Author.objects.all()
authors = await apply_prefetch_related(authors, fields=["posts"])

for author in authors:
    posts = await author.posts.all()  # Cached, no query

How It Works

Instead of a JOIN, prefetch_related does two queries:

-- Query 1: Get all authors
SELECT * FROM "authors"

-- Query 2: Get all posts for those authors
SELECT * FROM "posts" WHERE "author_id" IN (1, 2, 3, 4, 5)

The results are then matched in Python and cached on each instance.

select_related vs prefetch_related

	select_related	prefetch_related
Strategy	JOIN	Two queries + IN
Best for	ForeignKey, OneToOne	ManyToMany, reverse FK
Queries	1	2
Memory	Higher (denormalized rows)	Lower (normalized)

Multiple Relations

posts = await Post.objects.filter(active=True)
posts = await apply_prefetch_related(posts, fields=["author", "tags"])

Next Steps

→ CRUD — Create, read, update, delete