Migrate cache service from localStorage to IndexedDB

[IvanildoBarauna]

Summary

Migrated the BrowserCache service from localStorage-based implementation to IndexedDB, providing significantly larger storage capacity (GB vs ~5MB) while maintaining the same API surface and type safety.

Key Changes

Core Implementation

• Storage Backend: Replaced localStorage with IndexedDB for the cache service

  • Implemented openDB() helper to manage IndexedDB connections with proper initialization
  • Created object store schema with key as keyPath for efficient lookups
  • All cache operations now return Promises to handle async IndexedDB API

• API Changes: Made all BrowserCache methods async

  • get<T>(key) → async get<T>(key): Promise<T | null>
  • set<T>(key, data, ttl) → async set<T>(key, data, ttl): Promise<void>
  • remove(key) → async remove(key): Promise<void>
  • clearAll() → async clearAll(): Promise<void>
  • clearExpired() → async clearExpired(): Promise<void>
  • getStats() → async getStats(): Promise<CacheStats>
  • clearPortfolioCache() → async clearPortfolioCache(resourceType?): Promise<void>

• Data Structure: Updated cache entry format for IndexedDB

  • Entries now include key field (required for keyPath)
  • Maintained data, timestamp, and expiresAt fields
  • Removed JSON serialization overhead (IndexedDB handles structured cloning)

Hook Updates

Updated all data-fetching hooks to await cache operations:

• useExperience.ts
• useEducation.ts
• useProjects.ts
• useSocialLinks.ts

Testing Infrastructure

• Test Setup: Added fake-indexeddb polyfill for Node.js test environment
• Test Helpers: Created resetDB() and insertRawEntry() utilities for test isolation
• Test Updates:
  • Converted all test cases to async/await pattern
  • Updated assertions to work with IndexedDB cursor-based operations
  • Removed localStorage-specific test assertions
  • Enhanced error handling tests for IndexedDB-specific scenarios

Removed Features

• Removed QuotaExceededError handling logic (IndexedDB has much larger quota)
• Removed JSON parse error handling (IndexedDB uses structured cloning)

Notable Implementation Details

• Proper database connection lifecycle management with db.close() in all code paths
• Cursor-based iteration for clearAll(), clearExpired(), and getStats() operations
• Fire-and-forget async cleanup on expired cache reads (doesn't block the return)
• Graceful error handling maintains backward compatibility (methods never throw)
• SSR compatibility preserved through isBrowser() check for IndexedDB availability

https://claude.ai/code/session_01CWajAL77hbhWSCYir3UKMS

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.