Replace index-based discovery with ScraperRegistry (#554)

[danieldotnl]

Summary

• Replace fragile array-index lookups (SCRAPER_IDX/PLATFORM_IDX) with a type-safe ScraperRegistry using deterministic string IDs derived from config names
• Scraper/entity lookup is now reload-safe since order no longer matters
• Duplicate scraper names and entity names are automatically deduplicated with _2, _3 suffixes and a warning log

Changes

• New: registry.py — ScraperInstance dataclass and ScraperRegistry class
• Modified: __init__.py — uses registry instead of dict-of-lists; discovery passes string IDs
• Modified: const.py — removed SCRAPER_IDX/PLATFORM_IDX/SCRAPER_DATA, added SCRAPER_ID/ENTITY_KEY
• New: test_registry.py — unit tests for registry operations
• Modified: test_init.py — updated for registry-based lookups, added duplicate-name tests
• No changes to sensor.py, binary_sensor.py, button.py (they pass discovery_info through without inspecting it)

Test plan

• All 317 tests pass (including 3 new deduplication tests)
• Manual test with HA: configure two scrapers, reload, verify entities map to correct scrapers
• Manual test: configure duplicate scraper names, verify warning logged and both work

Closes #554

🤖 Generated with Claude Code

Summary by CodeRabbit

• Bug Fixes

  • Improved auto-generated naming for scrapers without explicit names, now using slugified resources or numeric fallbacks instead of generic labels.

• Improvements

  • Enhanced deduplication of scraper configurations for improved stability and uniqueness handling.

[coderabbitai]

Caution

Review failed

Pull request was closed or merged during review

📝 Walkthrough

Walkthrough

This pull request replaces index-based scraper discovery with a type-safe ScraperRegistry system. A new ScraperInstance dataclass wraps scrapers with their coordinators, while the registry manages instances by unique IDs. Constants are updated to use SCRAPER_ID and ENTITY_KEY instead of SCRAPER_IDX and PLATFORM_IDX. Discovery now passes IDs rather than indices via load_platform, and config/coordinator retrieval uses registry lookups instead of nested dictionary access.

Changes

Cohort / File(s)	Summary
Registry Core custom_components/multiscrape/registry.py	New module with ScraperInstance dataclass and ScraperRegistry class. Registry provides methods to register, retrieve, and manage scraper instances with deduplication and per-platform configuration tracking.
Constants & Public API custom_components/multiscrape/const.py	Removed legacy constants (SCRAPER_IDX, PLATFORM_IDX, COORDINATOR, SCRAPER, SCRAPER_DATA). Added new constants SCRAPER_ID and ENTITY_KEY for discovery and registry lookups.
Core Integration custom_components/multiscrape/__init__.py	Integrated ScraperRegistry into domain data storage. Added _deduplicate_id() and _deduplicate_entity_key() helpers. Updated setup to register ScraperInstance objects and generate names for unnamed scrapers. Modified async_get_config_and_coordinator() to retrieve from registry using discovery keys instead of indices.
Test Updates tests/test_init.py	Updated test setup and assertions to use ScraperRegistry API (get_all(), get(id)) instead of direct dictionary access. Changed discovery/config flow tests to use SCRAPER_ID and ENTITY_KEY keys. Extended tests to verify deduplication and per-scraper isolation.
Registry Tests tests/test_registry.py	New test module with comprehensive unit tests for ScraperRegistry and ScraperInstance, covering registration, retrieval, deduplication, and error handling.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Poem

🐰 Hoppy times ahead! ✨
No more index confusion, the registry's clear,
Each scraper has a name, its identity near,
With IDs and keys, they won't disappear,
Reload-safe discovery brings us good cheer! 🌟

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'Replace index-based discovery with ScraperRegistry (#554)' accurately summarizes the main change—migrating from array-index lookups to a type-safe registry system.
Linked Issues check	✅ Passed	The PR successfully implements all coding requirements from issue #554: ScraperRegistry class with register/get/get_all methods, ScraperInstance dataclass, deterministic string IDs, deduplication logic, and reload-safe lookups replacing fragile index-based discovery.
Out of Scope Changes check	✅ Passed	All changes are directly aligned with the registry migration objective: registry.py adds the core infrastructure, init.py and const.py refactor to use it, tests validate the new behavior, and platform modules remain unchanged.
Docstring Coverage	✅ Passed	Docstring coverage is 97.14% which is sufficient. The required threshold is 80.00%.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch feature/554-scraper-registry

📝 Coding Plan

• Generate coding plan for human review comments

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}