app_wide_search 0.1.1

app_wide_search: ^0.1.1 copied to clipboard

Added #

• Top-level example/ quickstart app with custom suggestion/result builders demonstrating Provider overrides and Hive initialization
• Documentation comments for all Hive adapters to satisfy pub.dev checks

Changed #

• Upgraded to flutter_riverpod ^3.0.1 and riverpod_annotation ^3.0.1 to stay on the supported release train without API changes
• Tightened adapter constructors with explicit docs (no runtime behavior changes)

Fixed #

• Regained pub.dev documentation score (20/20) and resolved "missing example" feedback

Notes #

• No public API changes; consumers can adopt v0.1.1 without migrations

---

Added #

• Production-safe search cancellation to prevent memory leaks
• Prevents race conditions from rapid typing or navigation
• Optional cancellationToken parameter in SearchProvider.search()
• CancelledException for handling cancelled operations
• Backward compatible - existing implementations work unchanged
• See API_CHANGELOG.md for migration details

Changed #

• ⚠️ Upgraded go_router from ^14.0.2 to ^16.0.0 (apps may need to upgrade)
• Upgraded flutter_lints to ^6.0.0 (stricter analysis rules)
• Upgraded intl to ^0.20.0 (latest i18n features)
• Updated to latest stable flutter_riverpod 2.6.1
• Added json_annotation ^4.9.0 and json_serializable ^6.8.0

Fixed #

Documentation #

Notes #

Performance optimization release.

Performance Improvements #

• SearchScreen Optimizations

  • Added configurable debounce for search-as-you-type (300ms default)
  • Reduced rebuilds by 85% using ValueNotifier for clear button
  • Added searchOnChange parameter (default: true)
  • Added debounceDuration parameter (default: 300ms)
  • Impact: 70% fewer search calls, smoother typing experience

• InMemorySearchProvider Optimizations

  • Implemented cached string normalization with _IndexedSearchItem
  • Pre-compute toLowerCase() on initialization instead of every search
  • Impact: 30% faster cold searches (15ms → 10ms for 100 items)
  • Impact: 50% faster warm searches (2ms → 1ms)

• SearchCacheRepository Improvements

  • Added maxEntries parameter (default: 50 queries)
  • Added maxSizeBytes parameter (default: 5MB)
  • Implemented LRU (Least Recently Used) eviction policy
  • Automatic cleanup of oldest 20% when limits exceeded
  • Impact: Memory bounded, prevents unlimited growth

• Code Quality

  • Enhanced analysis_options.yaml with strict linting
  • Added strict-casts, strict-inference, strict-raw-types
  • Added performance rules: prefer_const_constructors, etc.
  • Zero analyzer errors or warnings in library code

Overall Performance Gains #

• ~40% overall performance improvement ✅
• 85% reduction in widget rebuilds
• 70% reduction in search calls
• Memory bounded to prevent leaks
• All 18 tests passing ✅

Documentation #

• Added PERFORMANCE_REPORT.md with detailed metrics
• Added AUDIT_REPORT_FINAL.md with production audit
• Added OPTIMIZATION_SUMMARY.md with implementation details

Breaking Changes #

• None - all changes are backward-compatible ✅

---

Initial release of app_wide_search package.

Features #

• Flexible Search Interfaces

  • AppWideSearchDelegate for modal search using SearchDelegate
  • SearchScreen for full-screen search experience
  • Both can be used in the same application

• Grouped Results

  • GroupedSearchResults widget with ExpansionTile
  • Customizable groups with icons, colors, and priorities
  • Smooth animations and state preservation

• State Management

  • Full Riverpod integration with AutoDispose providers
  • Reactive search results and suggestions
  • Efficient state updates with minimal rebuilds

• Offline Support

  • Automatic result caching with Hive
  • Search history tracking
  • Configurable cache duration and retention limits

• Routing Integration

  • Deep-link support with go_router
  • SearchRouteConfig for easy route setup
  • ShellRoute support for navigation bars
  • Query parameter parsing for pre-filled searches

• Customization

  • SearchProvider abstract class for custom backends
  • InMemorySearchProvider for quick prototyping
  • Custom item builders
  • Custom result builders
  • Customizable empty and error states

• Internationalization

  • Built-in localization with intl package
  • SearchLocalizations for UI strings
  • Easy to add more languages

• Performance

  • Follows Flutter best practices
  • Efficient build methods
  • Const constructors where possible
  • Automatic resource cleanup with AutoDispose

• Cross-Platform

  • Supports Android, iOS, Web, macOS, Windows, and Linux
  • Consistent behavior across platforms

Documentation #

• Comprehensive README with multiple usage examples
• Inline documentation following Effective Dart guidelines
• Example app demonstrating all major features
• Advanced examples for custom implementations

Known Limitations #

• Hive adapters need to be generated with build_runner
• For production use, consider implementing platform-specific search optimizations