optimize-performance

NEVER optimize without a baseline measurement
ALWAYS write a regression spec before optimizing (query count assertion)
ALWAYS verify with EXPLAIN ANALYZE for database changes

NEVER write the report as "I applied includes(:author), then wrote a spec
to lock it in." The spec MUST be written and shown failing BEFORE the fix
appears in your output. Reordering for narrative flow fails the audit even
when the underlying work was correct.

# Performance Optimization — [Description]

## 1. Baseline
<N> queries for <endpoint/action> — source: <log line / profiler output>

## 2. Bottleneck
<N+1 on association X / missing index on column Y> — tool: <bullet / rack-mini-profiler / EXPLAIN ANALYZE>

## 3. Regression Spec — RED
`make_database_queries(count: <N>)` at <path>:<line> — failure: expected <M>, got <N>

## 4. Fix
<path>:<line> — <includes(:association) / add_index / cache block>

## 5. Regression Spec — GREEN
Spec passes ✓ (<M> queries)

## 6. EXPLAIN ANALYZE
Before: Seq Scan, actual time=<X>ms → After: Index Scan, actual time=<Y>ms

## 7. Quantified Improvement
Queries: <N> → <M> | p95: <X>ms → <Y>ms

# Use counter_cache to avoid COUNT queries in loops
# In migration: add_column :users, :posts_count, :integer, default: 0
# In Post model: belongs_to :user, counter_cache: true
user.posts_count  # no extra query

RSpec.describe "Post index performance" do
  it "loads posts with authors in a fixed number of queries" do
    create_list(:post, 10, :with_author)

    expect do
      get posts_path
    end.to make_database_queries(count: 2) # 1 posts query + 1 authors query
  end
end

EXPLAIN ANALYZE
  SELECT posts.*, users.name
  FROM posts
  INNER JOIN users ON users.id = posts.author_id
  WHERE posts.published = true;