Slow SQL queries can hide expensive scans and missed indexes; inspecting the PostgreSQL execution plan helps pinpoint bottlenecks and optimize query performance with explain and pev2.

Telemetry and analytics in open source CLIs and frameworks often require explicit opt-out settings or environment variables to stop data collection and crash reporting.

Cookie expiry and HTTP cache lifetimes depend on the Date header when present; missing it can make mocked-time tests and browser caching behave inconsistently.

Custom Ruby objects can gain colored, multiline IRB output by defining pretty_print; pretty_inspect is a simpler fallback when pretty printing is not needed.

PostgreSQL enums keep a defined sort order, unlike string-backed columns whose collation can produce surprising results. Ordered enum values make priority fields behave predictably.

Editing Rails credentials for staging or production requires server-only secret keys; a Capistrano task opens the decrypted file without copying keys to a dev machine.

Database query limits prevent runaway SQL from consuming all resources; PostgreSQL statement_timeout and MySQL max_execution_time can cancel slow statements automatically.

ILIKE searches in PostgreSQL can be slow without trigram indexing; a GIN index with pg_trgm can replace sequential scans with bitmap index scans.

Missing secrets can stay unnoticed in some environments and break behavior later. Using ! on Rails.application.credentials and Rails secrets raises immediately when a value is absent.

Project-specific RuboCop rules can flag conventions and stop the pipeline with little code, such as banning TODO and WIP comments in Ruby files.

String comparison can misorder gem and Ruby versions, letting newer releases slip past checks; Gem::Version gives reliable semantic version comparisons.

Rails npm packages need a version mapping from 4-part gem releases to npm’s 3-part semver, so package ranges keep working across patch and prerelease builds.

Ruby classes with many fields can become hard to instantiate or validate. Options range from positional and keyword arguments to ActiveModel::Attributes, Struct, hashes, and OpenStruct.

Custom attribute labels in Rails can be stored in locale files and reused by human_attribute_name and error messages instead of duplicating names in code.

Rails nested attributes become cumbersome with dynamic subrecords; Unpoly enables friendlier add/remove workflows without losing server-side validation and form helpers.

Different ways to build HTML-safe markup in Rails helpers affect readability, buffering, and XSS safety. concat, safe_join, capture, and html_safe each behave differently.

Rails performance debugging often hides slow SQL, view rendering, and N+1 queries; rack-mini-profiler helps surface bottlenecks during development.

Rails 7.1 changes production logging to standard out, which can break file-based logs in opscomplete setups unless the previous logger configuration is restored.

File downloads are slow and fragile to test through Capybara. Request specs make exports fast, stable, and easy to verify, with a small end-to-end check for the link.

Chaining Capybara matchers with and retries the full expectation when one check fails, making multiple assertions more reliable on pages that change asynchronously.

Browser support is strong for the Web Animations API, which lets JavaScript animate DOM elements and await or cancel CSS transitions and animations.

blank? on an ActiveRecord::Relation can load every row into memory and stall workers; use nil checks or exists?/empty? for cheap scope checks.

Numeric form fields on mobile get digit-focused keyboards, locale-aware decimal display, and consistent point-based values in JavaScript and form submission.

Selenium WebDriver dismisses browser dialogs by default and can close alerts before tests react. Setting unhandled_prompt_behavior to ignore leaves prompts open until handled.