kinemotion 0.66.0__tar.gz → 0.66.2__tar.gz

{kinemotion-0.66.0 → kinemotion-0.66.2}/CHANGELOG.md

@@ -7,6 +7,18 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- version list -->
 
+## kinemotion-v0.66.2 (2025-12-19)
+
+### Fix
+
+- **codecs**: remove Fly.io references, document actual Cloud Run architecture
+
+## kinemotion-v0.66.1 (2025-12-19)
+
+### Fix
+
+- **warnings**: remove redundant 'h264' codec, use standard 'avc1' only
+
 ## kinemotion-v0.66.0 (2025-12-19)
 
 ### Feat

{kinemotion-0.66.0 → kinemotion-0.66.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kinemotion
-Version: 0.66.0
+Version: 0.66.2
 Summary: Video-based kinematic analysis for athletic performance
 Project-URL: Homepage, https://github.com/feniix/kinemotion
 Project-URL: Repository, https://github.com/feniix/kinemotion

{kinemotion-0.66.0 → kinemotion-0.66.2}/backend/Dockerfile

@@ -12,12 +12,13 @@ ENV PYTHONUNBUFFERED=1 \
     PYTHONPATH=/app/src
 
 # Install system dependencies required for MediaPipe and OpenCV
-# ffmpeg: Video processing
+# ffmpeg: Video processing (includes libx264 for H.264 encoding)
 # libsm6, libxext6, libxrender-dev: OpenCV/MediaPipe GUI support
 # libgomp1: OpenMP for NumPy optimization
 # hadolint ignore=DL3008
 RUN apt-get update && apt-get install -y --no-install-recommends \
     ffmpeg \
+    libx264-dev \
     libsm6 \
     libxext6 \
     libxrender-dev \

{kinemotion-0.66.0 → kinemotion-0.66.2}/docs/guides/local-testing.md

@@ -1,6 +1,6 @@
 # Issue #12: Local Testing Guide
 
-This guide shows how to test the complete MVP stack locally before deploying to Fly.io and Vercel.
+This guide shows how to test the complete MVP stack locally before deploying to Google Cloud Run and Vercel.
 
 ## Prerequisites
 
@@ -582,4 +582,4 @@ If local testing fails:
 ______________________________________________________________________
 
 **Status:** Ready for local testing
-**Next:** Deploy to Fly.io + Vercel after confirming local testing works
+**Next:** Deploy to Cloud Run + Vercel after confirming local testing works (automatic via GitHub Actions)

{kinemotion-0.66.0 → kinemotion-0.66.2}/docs/guides/setup-issue-12.md

@@ -12,7 +12,7 @@ This document provides a complete setup guide for the Kinemotion Web UI MVP (Iss
           ▼                           ▼
     ┌─────────────┐         ┌──────────────────┐
     │  Frontend   │         │     Backend      │
-    │  (Vercel)   │         │    (Fly.io)      │
+    │  (Vercel)   │         │  (Cloud Run)     │
     │             │         │                  │
     │ React+Vite  │◄────────┤ FastAPI + uv     │
     │ Yarn        │ API     │ Real metrics     │
@@ -83,30 +83,23 @@ yarn dev
 
 ## Deployment
 
-### Backend: Deploy to Fly.io
+### Backend: Deploy to Google Cloud Run
 
-```bash
-cd backend
+Backend deploys automatically via GitHub Actions when you push to main.
 
-# Install Fly CLI if needed
-# https://fly.io/docs/hands-on/install-flyctl/
-
-# Launch app (first time only)
-flyctl launch --image-label kinemotion-backend
+```bash
+# Manual deployment (if needed)
+# Ensure you have gcloud CLI installed and authenticated
 
-# Optional: Set R2 storage credentials
-flyctl secrets set R2_ENDPOINT=https://YOUR_ACCOUNT_ID.r2.cloudflarestorage.com
-flyctl secrets set R2_ACCESS_KEY=YOUR_KEY
-flyctl secrets set R2_SECRET_KEY=YOUR_SECRET
-flyctl secrets set R2_BUCKET_NAME=kinemotion
+cd backend
 
-# Deploy
-flyctl deploy --remote-only
+# Build and push to Google Cloud Run
+# This is handled by .github/workflows/deploy-backend.yml automatically
 
 # View logs
-flyctl logs
+gcloud run services logs tail kinemotion-backend --project=kinemotion-backend
 
-# Your backend is now live at https://kinemotion-api.fly.dev
+# Your backend is live at https://kinemotion-backend-1008251132682.us-central1.run.app
 ```
 
 ### Frontend: Deploy to Vercel
@@ -118,12 +111,12 @@ cd frontend
 # 1. Visit https://vercel.com/new
 # 2. Select this repository
 # 3. Set Root Directory: frontend
-# 4. Add environment variable: VITE_API_URL=https://kinemotion-api.fly.dev
+# 4. Add environment variable: VITE_API_URL=https://kinemotion-backend-1008251132682.us-central1.run.app
 # 5. Deploy
 
 # Option 2: Deploy via Vercel CLI
 vercel --cwd=frontend
-# Set: VITE_API_URL=https://kinemotion-api.fly.dev
+# Set: VITE_API_URL=https://kinemotion-backend-1008251132682.us-central1.run.app
 
 # Your frontend is now live at https://kinemotion-mvp.vercel.app
 ```
@@ -137,14 +130,12 @@ kinemotion/
 │   │   ├── __init__.py
 │   │   └── app.py                    ← Main FastAPI application
 │   ├── pyproject.toml                ← Dependencies (uv)
-│   ├── Dockerfile                    ← Container for Fly.io
-│   ├── fly.toml                      ← Fly.io configuration
+│   ├── Dockerfile                    ← Container for Cloud Run
 │   ├── .env.example                  ← Environment template
 │   ├── .dockerignore
 │   ├── README.md                     ← Backend docs
 │   ├── SETUP.md                      ← Backend deployment guide
 │   ├── IMPLEMENTATION_SUMMARY.md     ← Technical details
-│   ├── FLY_DEPLOYMENT.md             ← Fly.io guide
 │   └── uv.lock                       ← Locked dependencies
 │
 ├── frontend/                         ← React frontend (can move to separate repo later)
@@ -178,7 +169,7 @@ kinemotion/
 
 ## API Endpoints
 
-### Backend: `http://localhost:8000` (dev) or `https://kinemotion-api.fly.dev` (prod)
+### Backend: `http://localhost:8000` (dev) or `https://kinemotion-backend-1008251132682.us-central1.run.app` (prod)
 
 #### Health Check
 
@@ -237,7 +228,17 @@ R2_BUCKET_NAME=kinemotion
 
 # API Configuration
 LOG_LEVEL=info
-CORS_ORIGINS=http://localhost:3000,http://localhost:5173,https://kinemotion-mvp.vercel.app
+CORS_ORIGINS=http://localhost:3000,http://localhost:5173,https://kinemotion.vercel.app
+
+# R2 Storage (optional - leave empty for development)
+R2_ENDPOINT=https://your-account-id.r2.cloudflarestorage.com
+R2_ACCESS_KEY=your_access_key
+R2_SECRET_KEY=your_secret_key
+R2_BUCKET_NAME=kinemotion
+
+# Supabase Database (configured in Google Cloud Secret Manager for production)
+SUPABASE_URL=https://smutfsalcbnfveqijttb.supabase.co
+SUPABASE_ANON_KEY=your_key_here
 ```
 
 ### Frontend (`frontend/.env`)
@@ -247,7 +248,7 @@ CORS_ORIGINS=http://localhost:3000,http://localhost:5173,https://kinemotion-mvp.
 VITE_API_URL=http://localhost:8000
 
 # Production
-# VITE_API_URL=https://kinemotion-api.fly.dev
+# VITE_API_URL=https://kinemotion-backend-1008251132682.us-central1.run.app
 ```
 
 ## Cloudflare R2 Setup (Optional)
@@ -280,14 +281,12 @@ R2 is optional for MVP - videos can be stored temporarily without it. To enable:
 # Format: https://YOUR_ACCOUNT_ID.r2.cloudflarestorage.com
 ```
 
-### 4. Set Fly.io secrets
+### 4. Set secrets in Google Cloud Secret Manager
 
 ```bash
-cd backend
-flyctl secrets set R2_ENDPOINT=https://YOUR_ACCOUNT_ID.r2.cloudflarestorage.com
-flyctl secrets set R2_ACCESS_KEY=YOUR_ACCESS_KEY
-flyctl secrets set R2_SECRET_KEY=YOUR_SECRET_KEY
-flyctl secrets set R2_BUCKET_NAME=kinemotion
+# R2 credentials are stored in Google Cloud Secret Manager
+# and accessed by Cloud Run service account
+# See backend deployment documentation for details
 ```
 
 ## Key Design Decisions
@@ -304,12 +303,14 @@ flyctl secrets set R2_BUCKET_NAME=kinemotion
 - **Benefit:** Works without R2, scales up gracefully
 - **Implementation:** Backend gracefully handles missing R2 credentials
 
-### 3. Fly.io + Vercel
+### 3. Google Cloud Run + Vercel + Supabase
 
-- **Backend:** Fly.io free tier (3GB RAM, adequate for video processing)
+- **Backend:** Google Cloud Run (video processing, auto-scales)
 - **Frontend:** Vercel free tier (industry standard for React)
-- **Storage:** Cloudflare R2 free tier (10GB/month, covers MVP volume)
-- **Total Cost:** $0 for realistic MVP testing
+- **Database:** Supabase (PostgreSQL database)
+- **Authentication:** Supabase Auth (OAuth, email/password)
+- **Storage:** Cloudflare R2 (video and results file storage)
+- **Total Cost:** Minimal for MVP testing
 
 ### 4. Separate Backend/Frontend Directories
 
@@ -360,22 +361,19 @@ ls -lh your_video.mp4
 file your_video.mp4
 
 # Check backend logs
-flyctl logs  # For production
+gcloud run services logs tail kinemotion-backend --project=kinemotion-backend
 # or check terminal for local development
 ```
 
 ### R2 upload fails (if R2 configured)
 
 ```bash
-# Verify credentials are set correctly
-flyctl secrets list
-
-# Test R2 connection
-flyctl ssh console
-# Then in console:
-import boto3
-s3 = boto3.client('s3', endpoint_url=os.getenv('R2_ENDPOINT'), ...)
-s3.head_bucket(Bucket=os.getenv('R2_BUCKET_NAME'))
+# Check R2 credentials in Google Cloud Secret Manager
+gcloud secrets versions access latest --secret=R2_ACCESS_KEY --project=kinemotion-backend
+gcloud secrets versions access latest --secret=R2_SECRET_KEY --project=kinemotion-backend
+
+# Verify endpoint and bucket name are correct
+# Test R2 connection from Cloud Run logs
 ```
 
 ## Testing Checklist
@@ -390,10 +388,11 @@ s3.head_bucket(Bucket=os.getenv('R2_BUCKET_NAME'))
 - [ ] Results display correctly
 - [ ] Error messages are user-friendly
 - [ ] Mobile responsive on phone
-- [ ] Backend deploys to Fly.io
+- [ ] Backend deploys to Cloud Run (automatic via GitHub Actions)
 - [ ] Frontend deploys to Vercel
 - [ ] Production URLs work
 - [ ] R2 download links work (if R2 configured)
+- [ ] Supabase integration works
 
 ## Next Steps
 
@@ -424,7 +423,6 @@ s3.head_bucket(Bucket=os.getenv('R2_BUCKET_NAME'))
   - `backend/README.md` - Backend overview
   - `backend/docs/setup.md` - Setup and deployment
   - `backend/docs/implementation-summary.md` - Technical details
-  - `backend/docs/fly-deployment.md` - Fly.io specific guide
   - `backend/docs/tests.md` - Test documentation
 
 - **Frontend:**
@@ -441,10 +439,10 @@ s3.head_bucket(Bucket=os.getenv('R2_BUCKET_NAME'))
 For issues:
 
 1. Check troubleshooting section above
-1. Review backend logs: `flyctl logs`
+1. Review backend logs: `gcloud run services logs tail kinemotion-backend --project=kinemotion-backend`
 1. Check frontend console (browser DevTools)
 1. Review README files in backend/ and frontend/
-1. Check GitHub issues #10, #11 for status
+1. Check GitHub issues for known problems
 
 ## Architecture Diagram
 
@@ -454,8 +452,8 @@ Coach (Phone/Desktop)
          ▼
 ┌─────────────────────────┐
 │   Frontend (Vercel)     │
-│  http://kinemotion-     │
-│  mvp.vercel.app         │
+│  https://kinemotion.    │
+│  vercel.app             │
 │                         │
 │ React + Vite + Yarn     │
 │ - Upload form           │
@@ -466,30 +464,32 @@ Coach (Phone/Desktop)
              │ (multipart: video + jump_type)
              ▼
 ┌─────────────────────────┐
-│  Backend (Fly.io)       │
+│  Backend (Cloud Run)    │
 │  https://kinemotion-    │
-│  api.fly.dev            │
+│  backend...run.app      │
 │                         │
 │ FastAPI + Python 3.12   │
 │ - Upload handler        │
 │ - Real metrics (CMJ/DJ) │
 │ - R2 integration        │
+│ - Supabase client       │
 │ - Error handling        │
 └────────────┬────────────┘
-             │ Upload/download
-             │ Results JSON
-             ▼
-    ┌──────────────────┐
-    │ Cloudflare R2    │
-    │ (Optional)       │
-    │                  │
-    │ - Videos (temp)  │
-    │ - Results (JSON) │
-    │ - Debug videos   │
-    └──────────────────┘
              │
-             ▼ Download links
-         (Frontend)
+    ┌────────┴────────┐
+    │                 │
+    ▼                 ▼
+┌─────────────┐   ┌──────────────────┐
+│  Supabase   │   │ Cloudflare R2    │
+│             │   │                  │
+│ - Auth      │   │ - Videos (temp)  │
+│ - Database  │   │ - Results (JSON) │
+│ - Sessions  │   │ - Debug videos   │
+│ - Metadata  │   └──────────────────┘
+└─────────────┘            │
+                           ▼
+                      (Frontend)
+                   Download links
 ```
 
 ______________________________________________________________________

{kinemotion-0.66.0 → kinemotion-0.66.2}/docs/quick-start.md

@@ -119,7 +119,7 @@ This shows all available endpoints with interactive testing.
 
 ## Next Steps
 
-- **Local testing works?** → Deploy to Fly.io + Vercel
+- **Local testing works?** → Deploy to Cloud Run + Vercel (automatic via GitHub Actions)
 - **Issues?** → See [`local-testing.md`](guides/local-testing.md) for detailed troubleshooting
 - **Ready to deploy?** → Follow [`setup-issue-12.md`](guides/setup-issue-12.md)
 

{kinemotion-0.66.0 → kinemotion-0.66.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "kinemotion"
-version = "0.66.0"
+version = "0.66.2"
 description = "Video-based kinematic analysis for athletic performance"
 readme = "README.md"
 requires-python = ">=3.10,<3.13"

{kinemotion-0.66.0 → kinemotion-0.66.2}/src/kinemotion/core/debug_overlay_utils.py

@@ -68,11 +68,11 @@ def create_video_writer(
     needs_resize = (display_width != width) or (display_height != height)
 
     # Try browser-compatible codecs first
-    # avc1/h264: H.264 (Most compatible, including iOS)
+    # avc1: H.264 (Most compatible, including iOS)
     # mp4v: MPEG-4 (Poor browser support, will trigger ffmpeg re-encoding for H.264)
     # ⚠️  CRITICAL: VP9 (vp09) is EXCLUDED - not supported on iOS/iPhone/iPad browsers!
     #     Adding VP9 will break debug video playback on all iOS devices.
-    codecs_to_try = ["avc1", "h264", "mp4v"]
+    codecs_to_try = ["avc1", "mp4v"]
 
     writer = None
     used_codec = "mp4v"  # Default fallback

{kinemotion-0.66.0 → kinemotion-0.66.2}/tests/core/test_debug_overlay_utils.py

@@ -76,7 +76,7 @@ class TestCodecSelection:
                 )
 
                 assert codec != "vp09", "VP9 codec used! This breaks iOS."
-                assert codec in ["avc1", "h264", "mp4v"], f"Unexpected codec: {codec}"
+                assert codec in ["avc1", "mp4v"], f"Unexpected codec: {codec}"
 
 
 class TestVideoWriterCreation:
@@ -199,6 +199,6 @@ class TestRegressionIOS:
                     fps=30.0,
                 )
 
-                # Should use H.264 (avc1/h264) not VP9
-                assert codec in ["avc1", "h264"]
+                # Should use H.264 (avc1) not VP9
+                assert codec == "avc1", f"Expected avc1, got {codec}"
                 assert codec != "vp09", "VP9 breaks iPhone playback!"

{kinemotion-0.66.0 → kinemotion-0.66.2}/uv.lock

@@ -913,7 +913,7 @@ wheels = [
 
 [[package]]
 name = "kinemotion"
-version = "0.65.0"
+version = "0.66.1"
 source = { editable = "." }
 dependencies = [
     { name = "click" },

kinemotion-0.66.0/backend/docs/fly-deployment.md

@@ -1,226 +0,0 @@
-# Fly.io Deployment Guide
-
-Complete deployment configuration for the Kinemotion backend on Fly.io.
-
-## Files Created
-
-1. **`Dockerfile`** - Multi-stage Python 3.12 container optimized for video processing
-1. **`fly.toml`** - Fly.io configuration with free tier settings
-1. **`.dockerignore`** - Excludes unnecessary files from Docker build
-
-## Quick Start
-
-### 1. Install Fly CLI
-
-```bash
-curl -L https://fly.io/install.sh | sh
-flyctl auth login
-```
-
-### 2. Initialize and Deploy
-
-```bash
-cd backend
-flyctl launch --image-label kinemotion-backend
-# OR deploy existing app:
-flyctl deploy --remote-only
-```
-
-### 3. Set Required Secrets
-
-For Cloudflare R2 storage (optional but recommended for production):
-
-```bash
-# Get credentials from Cloudflare R2 dashboard
-flyctl secrets set R2_ENDPOINT=https://YOUR_ACCOUNT_ID.r2.cloudflarestorage.com
-flyctl secrets set R2_ACCESS_KEY=YOUR_ACCESS_KEY
-flyctl secrets set R2_SECRET_KEY=YOUR_SECRET_KEY
-flyctl secrets set R2_BUCKET_NAME=kinemotion
-```
-
-### 4. Update CORS Origins (Optional)
-
-To add your frontend domain:
-
-```bash
-flyctl config set env.CORS_ORIGINS="https://your-domain.com,https://another-domain.com"
-```
-
-## Configuration Details
-
-### Docker Image
-
-- **Base**: Python 3.12-slim (latest, most efficient)
-- **Dependencies**:
-  - ffmpeg (video processing)
-  - libsm6, libxext6, libxrender-dev (OpenCV/MediaPipe support)
-  - libgomp1 (NumPy optimization)
-- **Package Manager**: uv (fast, deterministic dependency installation)
-- **Health Check**: `/health` endpoint monitored every 30 seconds
-
-### Fly.io Settings
-
-- **App Name**: `kinemotion-api`
-- **Region**: `sjc` (San Jose, US - adjust in `fly.toml` if needed)
-- **Machine**: Free tier (shared CPU, 256MB memory, 1 worker)
-- **Port**: 8000 (internal and external)
-- **Workers**: 1 (single worker for free tier stability)
-
-### Graceful Shutdown
-
-- **Timeout**: 60 seconds for video processing completion
-- **Health Check Grace Period**: 5 seconds on startup
-- **Restart**: Automatic on crash or deployment
-
-## Environment Variables
-
-### Default Configuration (in fly.toml)
-
-- `LOG_LEVEL`: "info" (set to "debug" for troubleshooting)
-- `WORKERS`: "1" (shared CPU safe setting)
-- `CORS_ORIGINS`: "https://kinemotion-mvp.vercel.app"
-- `PYTHONUNBUFFERED`: "1" (streaming output)
-- `PYTHONDONTWRITEBYTECODE`: "1" (reduce disk I/O)
-
-### Secrets (set via `flyctl secrets set`)
-
-- `R2_ENDPOINT`: Cloudflare R2 endpoint
-- `R2_ACCESS_KEY`: R2 access key ID
-- `R2_SECRET_KEY`: R2 secret key
-- `R2_BUCKET_NAME`: S3 bucket name (default: "kinemotion")
-
-## Performance Notes
-
-### Free Tier Considerations
-
-1. **Single Worker**: Video processing is CPU-bound, single worker prevents memory thrashing
-1. **256MB Memory**: Sufficient for sequential video processing with temp file cleanup
-1. **Graceful Timeout**: 60 seconds allows complete video analysis before shutdown
-1. **Auto Restart**: Failed requests trigger immediate restart
-
-### Expected Performance
-
-- **Typical analysis time**: 30-120 seconds (depends on video length/quality)
-- **Concurrent requests**: Limited to 1 (single worker)
-- **Queue timeout**: Requests will fail if processing exceeds 60s
-
-### Upgrade Path
-
-For production with multiple concurrent requests:
-
-1. Increase `WORKERS` to 2-4 in `fly.toml`
-1. Upgrade machine to `shared-cpu-2x` or `shared-cpu-4x`
-1. Increase memory_mb to 512-1024
-1. Consider load balancing across multiple machines
-
-## Monitoring
-
-### View Logs
-
-```bash
-flyctl logs
-flyctl logs -n 100  # Last 100 lines
-flyctl logs --follow  # Stream logs
-```
-
-### Check Status
-
-```bash
-flyctl status
-flyctl machine list
-flyctl machine status <machine_id>
-```
-
-### View Metrics
-
-```bash
-flyctl metrics
-```
-
-## Common Issues
-
-### "Out of memory" errors
-
-- Increase machine memory in `fly.toml`: `memory_mb = 512`
-- Or upgrade machine type: `cpu_kind = "shared-cpu-2x"`
-- Reduce concurrent workers or process videos sequentially
-
-### Health check fails
-
-- Check logs: `flyctl logs | grep health`
-- Ensure `/health` endpoint is working: `curl https://kinemotion-api.fly.dev/health`
-- Increase `grace_period` in `fly.toml` if startup takes time
-
-### Video analysis timeouts
-
-- Increase `timeout-graceful-shutdown` in `[processes]` section
-- Consider processing videos offline with async jobs
-- Implement video streaming/chunking for large files
-
-### R2 storage errors
-
-- Verify secrets are set: `flyctl secrets list`
-- Check R2 credentials in Cloudflare dashboard
-- Ensure bucket name matches R2 settings
-
-## Debugging
-
-### Build locally
-
-```bash
-docker build -t kinemotion-backend .
-docker run -p 8000:8000 kinemotion-backend
-```
-
-### Test health endpoint
-
-```bash
-curl http://localhost:8000/health
-```
-
-### Test analysis endpoint
-
-```bash
-curl -X POST -F "file=@test.mp4" -F "jump_type=cmj" \
-  http://localhost:8000/api/analyze
-```
-
-## Useful Commands
-
-```bash
-# Deploy changes
-flyctl deploy --remote-only
-
-# Check app info
-flyctl info
-
-# View environment variables
-flyctl config view
-
-# Scale machines (paid tier)
-flyctl scale count 2
-
-# Restart machines
-flyctl machines restart <machine_id>
-
-# SSH into machine (debugging)
-flyctl ssh console
-
-# View recent deployments
-flyctl releases
-```
-
-## Next Steps
-
-1. **Customize Region**: Update `primary_region` in `fly.toml` (e.g., "lax", "iad", "lon")
-1. **Add Domain**: Use `flyctl certs add your-domain.com`
-1. **Configure Monitoring**: Set up alerts via Fly dashboard
-1. **Optimize**: Monitor logs and metrics, adjust WORKERS if needed
-1. **Scale**: Upgrade to paid machine when needed for production load
-
-## References
-
-- [Fly.io Documentation](https://fly.io/docs/)
-- [Dockerfile Best Practices](https://docs.docker.com/develop/develop-images/dockerfile_best-practices/)
-- [Python on Fly.io](https://fly.io/docs/languages-and-frameworks/python/)
-- [Fly Machines API](https://fly.io/docs/reference/machines/)