BloodSugar Tracker

中文版

My grandmother has diabetes and used to track her blood sugar in a handwritten notebook. As the pages piled up, finding old records became impossible — every doctor's visit meant flipping through stacks of paper. I built this app to make tracking effortless and sharing with doctors instant.

A lightweight, offline-first Android blood sugar recording app designed for elderly users. Built with Kotlin, Jetpack Compose, and Material 3.

Features

• Quick Recording — 6 meal segments (before/after breakfast, lunch, dinner) with automatic time-based inference
• Medication Tracking — Log medications with name, dosage, and notes; edit/delete with one tap
• Statistics Summary — Collapsible card showing average, max, min glucose per meal segment
• Unit Toggle — Switch between mmol/L and mg/dL with one tap (persists across restarts)
• PDF Export — One-tap report generation with stats + full record table, share via Bluetooth, email, or print
• Copy Summary — One-tap clipboard copy for quick sharing via messaging apps
• Trend Chart — Smooth bezier curves with gradient fill, month/year toggle, and normal-range indicator
• Time Grouping — Records grouped by This Month / This Year / Earlier with collapsible headers
• Bilingual — Chinese and English UI, follows system language automatically
• Accessibility — Large fonts for elderly users, WCAG AA color contrast for glucose status indicators
• Data Safety — Room database with migration framework — APK updates never delete your data
• Privacy — Fully offline. Zero data collection. All records stay on your device.

Screenshots

Main Screen	Record Dialog	Statistics	Trend Chart	Medications

Tech Stack

Component	Version	Notes
AGP	7.4.2	Compatible with JDK 11
Kotlin	1.8.22	Compose Compiler 1.4.8
JDK	11	Eclipse Adoptium
Compose BOM	2023.06.01
Room	2.5.2	Using kapt

Zero third-party dependencies — PDF generation uses Android's native PdfDocument API. Charts are drawn with Compose Canvas. No charting libraries, no PDF libraries.

Architecture

app/src/main/java/com/bloodsugar/
├── MainActivity.kt          # Entry point
├── data/
│   ├── AppDatabase.kt       # Room database with migration framework
│   ├── Record.kt            # Entity: id, value, segment, note, timestamp
│   ├── RecordDao.kt         # DAO: Flow queries + SQL aggregation
│   ├── Medication.kt        # Entity: id, name, dosage, note, timestamp
│   ├── MedicationDao.kt     # DAO: Flow queries for medications
│   └── SegmentStats.kt      # Stats data class
├── ui/
│   ├── MainScreen.kt        # Main UI: header + stats card + grouped list
│   ├── MainViewModel.kt     # State management (MVVM)
│   ├── RecordSheet.kt       # Record entry/edit dialog
│   ├── MedicationSheet.kt   # Medication add/edit dialog
│   ├── ChartOverlay.kt      # Trend chart (bezier curves + gradient)
│   ├── StatsSummaryCard.kt  # Collapsible statistics card
│   ├── PdfExporter.kt       # PDF report generation (zero dependencies)
│   └── theme/               # Colors, typography, Material 3 theme
└── util/
    ├── GlucoseValidator.kt  # Input validation
    ├── GlucoseUnit.kt       # mmol/L ↔ mg/dL conversion
    └── MealSegment.kt       # 6 meal segments + time-based inference

• MVVM — ViewModel talks directly to DAO (no Repository layer, no DI)
• Jetpack Compose — 100% declarative UI
• Material 3 — Design system with light/dark theme
• StateFlow — Reactive state management
• Unit Tests — 22 tests covering validation, unit conversion, and meal segments

Building

# Set JAVA_HOME (adjust path for your system)
export JAVA_HOME="/path/to/jdk-11"

# Build debug APK
./gradlew assembleDebug

# Output: app/build/outputs/apk/debug/app-debug.apk

Installation

1. Download the APK from Releases or build from source
2. Enable "Install from unknown sources" on your Android device
3. Install the APK

Adding a New Language

1. Create app/src/main/res/values-<locale>/strings.xml
2. Copy all entries from app/src/main/res/values/strings.xml (Chinese) or values-en/strings.xml (English)
3. Translate the values

Data Safety

• No fallbackToDestructiveMigration() — updates never destroy your data
• APK overlay installs preserve all existing records
• Schema changes use Room's Migration framework with SQL scripts

Contributing

Contributions are welcome! Feel free to open issues or submit pull requests.

License

MIT License — free to use, modify, and distribute.

Why This Exists

Diabetes management requires consistent tracking, but existing apps are often bloated, ad-supported, or require cloud accounts. This app is:

• Dead simple — designed for people who find smartphones confusing
• Fully offline — no account, no cloud, no tracking
• Zero cost — no ads, no subscriptions, no paywalls
• Shareable — one-tap PDF export for doctor visits

Built with care for those who need it most.