Planning engine fixes (from adversarial code review):
- Bug #1: sortByLeisure tie-breaking uses totalDrivingHours
- Bug #2: allDates/calculateRestDays guard-let-break prevents infinite loop
- Bug #3: same-day trip no longer rejected (>= in dateRange guard)
- Bug #4: ScenarioD rationale shows game count not stop count
- Bug #5: ScenarioD departureDate advanced to next day after last game
- Bug #6: ScenarioC date range boundary uses <= instead of <
- Bug #7: DrivingConstraints clamps maxHoursPerDriverPerDay via max(1.0,...)
- Bug #8: effectiveTripDuration uses inclusive day counting (+1)
- Bug #9: TripWizardViewModel validates endDate >= startDate
- Bug #10: allDates() uses min/max instead of first/last for robustness
- Bug #12: arrivalBeforeGameStart accounts for game end time at departure
- Bug #15: ScenarioBPlanner replaces force unwraps with safe unwrapping
Tests: 16 regression test suites + updated existing test expectations
Marketing: Remotion canvas set to 886x1920 for App Store preview spec
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Remove Scripts/ directory (scraper no longer needed)
- Add themed background documentation to CLAUDE.md
- Add .gitignore for marketing-videos to prevent node_modules tracking
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This refactor fixes the achievement system by using stable canonical string
IDs (e.g., "stadium_mlb_fenway_park") instead of random UUIDs. This ensures
stadium mappings for achievements are consistent across app launches and
CloudKit sync operations.
Changes:
- Stadium, Team, Game: id property changed from UUID to String
- Trip, TripStop, TripPreferences: updated to use String IDs for games/stadiums
- CKModels: removed UUID parsing, use canonical IDs directly
- AchievementEngine: now matches against canonical stadium IDs
- All test files updated to use String IDs instead of UUID()
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The resolveLocations() function was overwriting valid LocationInput
objects (with coordinates) from LocationSearchSheet. Now it only
resolves locations that don't already have coordinates.
Added debug logging to trace scenario selection.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Adds a new planning mode that lets users follow a team's schedule
(home + away games) and builds multi-city routes accordingly.
Key changes:
- New ScenarioDPlanner with team filtering and route generation
- Team picker UI with sport grouping and search
- Fix TravelEstimator 5-day limit (was 2-day) for cross-country routes
- Fix DateInterval end boundary to include games on last day
- Comprehensive test suite covering edge cases:
- Multi-city routes with adequate/insufficient time
- Optimal game selection per city for feasibility
- 5-day driving segment limits
- Multiple driver scenarios
Enables trips like Houston → Chicago → Anaheim following the Astros.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Three-scenario planning engine (A: date range, B: selected games, C: directional routes)
- GeographicRouteExplorer with anchor game support for route exploration
- Shared ItineraryBuilder for travel segment calculation
- TravelEstimator for driving time/distance estimation
- SwiftUI views for trip creation and detail display
- CloudKit integration for schedule data
- Python scraping scripts for sports schedules
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
