YuHeng Future Roadmap (Draft)

Feature Priority Matrix

Feature	Priority	Effort	Status	Notes
Calendar View	P1	Low	⏳ Planned	Natural extension of existing history
Meal Plan Presets	P1	Low	⏳ Planned	Extend existing settings framework
Health Mode Switch	P1	Low	⏳ Planned	UI emphasis on different nutrition metrics
PWA Support	P2	Medium	📋 Planning	Next.js built-in support
Calendar Image Sharing	P2	Medium	📋 Planning	Frontend generated, no storage needed
Pre/Post Meal Comparison	P2	Medium	📋 Planning	High value user insights
Blood Sugar Risk Assessment	P3	Medium-High	📋 Planning	Requires carb/GI database
Lipid / Vascular Pressure Assessment	P3	Medium-High	📋 Planning	Requires fat/cholesterol breakdown data
Purine Risk Assessment	P3	Medium-High	📋 Planning	For gout-prone users
Liver Risk Assessment	P4	High	📋 Planning	Requires medical model, use caution

---

Architecture Decision Records

ADR 1: Calendar Image Storage - No S3/MinIO

Decision: Frontend generation + Frontend consumption

User clicks "Share" → html2canvas generates image → User saves/shares → Image discarded

Rationale:

1. Share images are “generate-and-consume”, no persistence needed
2. Frontend html2canvas/satori fully meets requirements
3. Zero storage cost, zero network overhead
4. For future short-link sharing, PostgreSQL bytea + TTL is sufficient
5. Consider S3/MinIO only when user base exceeds 100k+

Implementation Plan:

• Install html2canvas
• Create components/calendar/ShareButton.tsx

---

Implementation Roadmap

⏳ Phase 1: High Value, Low Effort (Next 1-2 Weeks)

Calendar View for Nutrition History

• Calendar grid component showing daily calories
• Click date to view detailed entries
• Based on existing getHistory API

Meal Plan Presets

• Preset templates: Bulk, Cut, Maintenance, Low-Carb
• default_meal_plan setting
• Onboarding flow for new users

Health Mode Switch

• [ ] health_focus setting: general

  glycemic

  lipid

  purine

• UI highlights corresponding nutrition metrics

---

📋 Phase 2: Medium Effort (2-3 Weeks, Planning)

PWA Support

• public/manifest.json configuration
• Enable PWA in next.config.ts
• Service Worker for offline data caching strategy

Calendar Image Sharing

• html2canvas integration
• ShareButton component
• WeChat Moments support (1080x1920)
• Watermark and branding

Pre/Post Meal Comparison

• Architecture and data model designed (✅ 2026-05-12)
• Measurement records table (dual DB support: PostgreSQL + SQLite)
• Pluggable metric analyzer framework
• Nutrition comparison analyzer (Phase 1)
• Blood glucose analyzer skeleton (Phase 2)
• Comparison API endpoints
• ComparisonCard UI component
• Comparison analysis page

Architecture Design:

• Pluggable metric system: nutrition → glucose → uric acid → lipid
• Dual database adapter pattern (PostgreSQL JSONB + SQLite TEXT)
• Comparison result caching for performance
• See: features/comparison_analysis.md

---

📋 Phase 3: Health Risk Assessment (Long-term Planning)

Health Risk Models

• Blood sugar risk assessment model (requires GI database)
• Lipid / vascular pressure assessment
• Purine risk assessment (requires purine database)
• Disclaimer system

---

Code Organization Plan

lib/
├── health/               ← Phase 3 (Planned)
│   ├── glycemic.ts       # Blood sugar assessment
│   ├── lipid.ts          # Lipid assessment
│   └── purine.ts         # Purine assessment

components/
├── calendar/             ← Phase 1 (Planned)
│   ├── CalendarGrid.tsx
│   ├── DayCell.tsx
│   └── ShareButton.tsx   ← Phase 2
├── health/               ← Phase 3 (Planned)
│   ├── RiskIndicator.tsx
│   └── ReportCard.tsx
└── meal-plans/           ← Phase 1 (Planned)
    └── PresetSelector.tsx

---

Important Notes

Health Risk Features

• Must add explicit disclaimer: “This assessment is for reference only, does not constitute medical diagnosis”
• Recommend citing authoritative sources (WHO, Chinese Nutrition Society, etc.)
• Avoid medical terms like “diagnosis”, “treatment”; use “risk indication”, “reference advice” instead

Data Migration

• Set default values when adding user_id columns to avoid orphaned existing data
• Need data ownership strategy when migrating to real users in future

---

Open Discussion Items

• Data source selection for health risk assessment (open source vs paid API)
• PWA offline caching strategy (which data needs offline availability)
• Data model design for pre/post meal comparison