Add API test script and remove deployment guide

- Add test_api.sh: Comprehensive test suite for API validation
- Tests health check endpoint
- Validates semantic search with query 'what did jesus say about eternal life'
- Asserts top result is John 17 with similarity >= 0.77
- Tests testament filter functionality
- Checks interactive documentation accessibility
- All 8 tests passing ✓

- Remove DEPLOYMENT_GUIDE.md (no longer needed post-deployment)

Test results: Book jhn, Chapter 17, Similarity 0.789 (> 0.77 threshold)

DEPLOYMENT_GUIDE.md

@@ -1,326 +0,0 @@
-# Deployment Guide: Biblos API to Hugging Face Spaces
-
-This guide walks you through deploying the Biblos Semantic Search API to Hugging Face Spaces.
-
-## Prerequisites
-
-1. **Hugging Face Account** - Sign up at [huggingface.co](https://huggingface.co)
-2. **Git** - Installed on your machine
-3. **Python 3.10+** - For running the data preparation script
-
----
-
-## Step 1: Prepare the Data
-
-First, extract the Bible embeddings from your existing ZIP files:
-
-```bash
-# Navigate to the hf-spaces directory
-cd hf-spaces
-
-# Run the data preparation script (uses desktop embeddings by default)
-python prepare_data.py
-
-# OR, use iOS embeddings (smaller, quantized):
-python prepare_data.py --ios
-
-# OR, specify custom paths:
-python prepare_data.py --source ../public/data --output ./data
-```
-
-This will:
-- Extract JSON files from ZIP archives
-- Validate embeddings
-- Create a `data/` directory with 66 JSON files (one per book)
-- Total size: ~200-250MB
-
-**Verify the extraction:**
-```bash
-ls -lh data/
-# Should show 66 .json files
-```
-
----
-
-## Step 2: Test Locally (Optional but Recommended)
-
-Before deploying, test the API locally:
-
-```bash
-# Install dependencies
-pip install -r requirements.txt
-
-# Run the server
-python -m uvicorn app:app --host 0.0.0.0 --port 7860
-
-# In another terminal, test the API
-curl -X POST http://localhost:7860/search \
-  -H "Content-Type: application/json" \
-  -d '{"query": "love one another", "limit": 3}'
-```
-
-Visit http://localhost:7860/docs for interactive API documentation.
-
----
-
-## Step 3: Create a Hugging Face Space
-
-1. **Go to Hugging Face** - Visit [huggingface.co/spaces](https://huggingface.co/spaces)
-
-2. **Click "Create new Space"**
-
-3. **Configure the Space:**
-   - **Space name**: Choose a name (e.g., `biblos-api`)
-   - **License**: MIT
-   - **SDK**: Select **Docker**
-   - **Space hardware**: CPU basic (free)
-   - **Visibility**: Public or Private (your choice)
-
-4. **Click "Create Space"**
-
----
-
-## Step 4: Upload Files to Your Space
-
-### Option A: Using Git (Recommended)
-
-```bash
-# Clone your Space repository (replace YOUR-USERNAME and YOUR-SPACE-NAME)
-git clone https://huggingface.co/spaces/YOUR-USERNAME/YOUR-SPACE-NAME
-cd YOUR-SPACE-NAME
-
-# Copy all files from hf-spaces directory
-cp -r ../hf-spaces/* .
-
-# Add all files to git
-git add .
-
-# Commit
-git commit -m "Initial deployment of Biblos API"
-
-# Push to Hugging Face (this triggers deployment)
-git push
-```
-
-### Option B: Using Web Interface
-
-1. Click "Files and versions" tab in your Space
-2. Click "Add file" → "Upload files"
-3. Upload these files from the `hf-spaces/` directory:
-   - `app.py`
-   - `requirements.txt`
-   - `Dockerfile`
-   - `README.md`
-   - The entire `data/` directory (66 JSON files)
-4. Commit the changes
-
-**Important:** Make sure to upload the `data/` directory with all 66 JSON files!
-
----
-
-## Step 5: Wait for Deployment
-
-After pushing/uploading:
-
-1. **HF Spaces will automatically build** your Docker container
-2. You'll see logs in the "Logs" tab
-3. Initial build takes **5-10 minutes**
-4. Once complete, you'll see: `Application startup complete`
-
-**Build process:**
-- Install Python dependencies
-- Download the BGE-large model (~639MB)
-- Load all Bible embeddings into memory
-- Start the FastAPI server
-
----
-
-## Step 6: Test Your Deployed API
-
-Once deployment is complete, your API is live!
-
-**Your API URL:**
-```
-https://YOUR-USERNAME-YOUR-SPACE-NAME.hf.space
-```
-
-### Test Endpoints
-
-**Health Check:**
-```bash
-curl https://YOUR-USERNAME-YOUR-SPACE-NAME.hf.space/
-```
-
-**Search:**
-```bash
-curl -X POST https://YOUR-USERNAME-YOUR-SPACE-NAME.hf.space/search \
-  -H "Content-Type: application/json" \
-  -d '{
-    "query": "faith without works",
-    "testament": "new",
-    "limit": 5
-  }'
-```
-
-**Interactive Docs:**
-Visit: `https://YOUR-USERNAME-YOUR-SPACE-NAME.hf.space/docs`
-
----
-
-## Step 7: Update Your Frontend
-
-Update your Biblos frontend to use the new API:
-
-**In your frontend code:**
-```javascript
-// Instead of client-side search, call your API
-const API_URL = 'https://YOUR-USERNAME-YOUR-SPACE-NAME.hf.space'
-
-async function searchBible(query, options = {}) {
-  const response = await fetch(`${API_URL}/search`, {
-    method: 'POST',
-    headers: { 'Content-Type': 'application/json' },
-    body: JSON.stringify({ query, ...options })
-  })
-
-  const data = await response.json()
-  return data.results
-}
-```
-
----
-
-## Troubleshooting
-
-### Build Fails
-
-**Check logs** in the "Logs" tab. Common issues:
-
-1. **Missing data files**
-   - Ensure all 66 JSON files are in `data/` directory
-   - Re-run `prepare_data.py` if needed
-
-2. **Out of memory**
-   - Reduce model size (use iOS embeddings)
-   - Upgrade to larger Space hardware (paid)
-
-3. **Import errors**
-   - Check `requirements.txt` versions
-   - Ensure all dependencies are listed
-
-### API Returns 503 Error
-
-This means the model hasn't loaded yet:
-- Wait a few more seconds
-- Check logs for model loading progress
-- First request after cold start takes 2-3 seconds
-
-### Slow Response Times
-
-- First request: Expected (~2-3s for model loading)
-- Subsequent requests: Should be 50-100ms
-- If consistently slow:
-  - Check Space hardware in settings
-  - Consider upgrading to GPU (faster inference)
-
----
-
-## Monitoring & Maintenance
-
-### View Logs
-1. Go to your Space page
-2. Click "Logs" tab
-3. See real-time logs of API requests and responses
-
-### Update the API
-1. Make changes locally
-2. Commit and push:
-   ```bash
-   git add .
-   git commit -m "Update API"
-   git push
-   ```
-3. HF Spaces automatically rebuilds
-
-### Space Sleep
-- Free CPU Spaces may sleep after 48h of inactivity
-- First request after sleep has ~30s cold start
-- To prevent: Upgrade to persistent hardware (paid)
-
----
-
-## Cost & Limits
-
-### Free Tier (CPU Basic)
-- ✅ Perfect for moderate traffic
-- ✅ No time limit
-- ✅ No request limit
-- ⚠️ May sleep after 48h inactivity
-- ⚠️ Slower inference (CPU only)
-
-### Paid Tiers
-If you need more performance:
-- **CPU Upgrade** ($0.04/hour) - Faster CPU, no sleep
-- **GPU T4** ($0.60/hour) - 10x faster inference
-- **GPU A10G** ($3.15/hour) - Best performance
-
-Upgrade in Space Settings → "Change hardware"
-
----
-
-## Security Considerations
-
-1. **CORS** - Currently allows all origins. To restrict:
-   ```python
-   # In app.py, change:
-   allow_origins=["https://yourdomain.com"]
-   ```
-
-2. **Rate Limiting** - Consider adding:
-   ```bash
-   pip install slowapi
-   ```
-   See [slowapi docs](https://slowapi.readthedocs.io/)
-
-3. **API Keys** - For private use, add authentication:
-   ```python
-   from fastapi.security import HTTPBearer
-   ```
-
----
-
-## Next Steps
-
-1. ✅ Deploy your API
-2. ✅ Test all endpoints
-3. ✅ Update frontend to use API
-4. ✅ Share API URL with other developers
-5. ✅ Monitor usage and performance
-6. Consider: Add caching, rate limiting, analytics
-
----
-
-## Support
-
-- **HF Spaces Docs**: [huggingface.co/docs/hub/spaces](https://huggingface.co/docs/hub/spaces)
-- **FastAPI Docs**: [fastapi.tiangolo.com](https://fastapi.tiangolo.com)
-- **Community Forum**: [discuss.huggingface.co](https://discuss.huggingface.co)
-
----
-
-## Summary
-
-```bash
-# Quick deployment checklist:
-☐ Run prepare_data.py to extract embeddings
-☐ Create new Space on Hugging Face (Docker SDK)
-☐ Upload all files (app.py, requirements.txt, Dockerfile, README.md, data/)
-☐ Wait for build to complete (~5-10 min)
-☐ Test API endpoints
-☐ Update frontend to use new API URL
-☐ 🎉 Your API is live!
-```
-
-**Estimated total time:** 15-30 minutes
-
-Good luck! 🚀

test_api.sh

@@ -0,0 +1,146 @@
+#!/bin/bash
+
+# Biblos API Test Script
+# Tests the semantic search API with expected results
+
+API_URL="https://dssjon-biblos-api.hf.space"
+QUERY="what did jesus say about eternal life"
+EXPECTED_BOOK="jhn"
+EXPECTED_CHAPTER=17
+MIN_SIMILARITY=0.77
+
+echo "======================================"
+echo "Biblos API Test Suite"
+echo "======================================"
+echo ""
+
+# Colors for output
+GREEN='\033[0;32m'
+RED='\033[0;31m'
+NC='\033[0m'
+
+pass_count=0
+fail_count=0
+
+print_pass() {
+    echo -e "${GREEN}✓ PASS${NC}: $1"
+    ((pass_count++))
+}
+
+print_fail() {
+    echo -e "${RED}✗ FAIL${NC}: $1"
+    ((fail_count++))
+}
+
+# Test 1: Health Check
+echo "Test 1: Health Check (GET /)"
+health=$(curl -s "$API_URL/")
+status=$(echo "$health" | jq -r '.status' 2>/dev/null)
+
+if [ "$status" = "online" ]; then
+    print_pass "API is online and responding"
+else
+    print_fail "API health check failed"
+fi
+
+books=$(echo "$health" | jq -r '.books_loaded' 2>/dev/null)
+if [ "$books" = "66" ]; then
+    print_pass "All 66 books loaded"
+else
+    print_fail "Expected 66 books, got $books"
+fi
+echo ""
+
+# Test 2: Semantic Search
+echo "Test 2: Semantic Search"
+echo "Query: \"$QUERY\""
+search=$(curl -s -X POST "$API_URL/search" \
+    -H "Content-Type: application/json" \
+    -d "{\"query\":\"$QUERY\",\"limit\":3}")
+
+# Check if we got results
+results_count=$(echo "$search" | jq -r '.results | length' 2>/dev/null)
+if [ "$results_count" -gt 0 ]; then
+    print_pass "Search returned $results_count results"
+else
+    print_fail "No search results returned"
+    exit 1
+fi
+echo ""
+
+# Test 3: Validate Top Result
+echo "Test 3: Validate Top Result"
+book=$(echo "$search" | jq -r '.results[0].book')
+chapter=$(echo "$search" | jq -r '.results[0].chapter')
+similarity=$(echo "$search" | jq -r '.results[0].similarity')
+content=$(echo "$search" | jq -r '.results[0].content' | head -c 80)
+
+echo "Top Result:"
+echo "  Book: $book"
+echo "  Chapter: $chapter"
+echo "  Similarity: $similarity"
+echo "  Content: ${content}..."
+echo ""
+
+if [ "$book" = "$EXPECTED_BOOK" ]; then
+    print_pass "Book matches expected: $EXPECTED_BOOK"
+else
+    print_fail "Expected book '$EXPECTED_BOOK', got '$book'"
+fi
+
+if [ "$chapter" = "$EXPECTED_CHAPTER" ]; then
+    print_pass "Chapter matches expected: $EXPECTED_CHAPTER"
+else
+    print_fail "Expected chapter $EXPECTED_CHAPTER, got $chapter"
+fi
+
+# Check similarity threshold
+if (( $(echo "$similarity >= $MIN_SIMILARITY" | bc -l) )); then
+    print_pass "Similarity score $similarity >= $MIN_SIMILARITY"
+else
+    print_fail "Similarity score $similarity < $MIN_SIMILARITY"
+fi
+echo ""
+
+# Test 4: Testament Filter
+echo "Test 4: Testament Filter"
+nt_search=$(curl -s -X POST "$API_URL/search" \
+    -H "Content-Type: application/json" \
+    -d '{"query":"love","testament":"new","limit":1}')
+testament=$(echo "$nt_search" | jq -r '.results[0].testament')
+
+if [ "$testament" = "NT" ]; then
+    print_pass "Testament filter working correctly"
+else
+    print_fail "Testament filter returned: $testament"
+fi
+echo ""
+
+# Test 5: Interactive Docs
+echo "Test 5: Interactive Documentation"
+docs_code=$(curl -s -o /dev/null -w "%{http_code}" "$API_URL/docs")
+
+if [ "$docs_code" = "200" ]; then
+    print_pass "Interactive docs accessible at /docs"
+else
+    print_fail "Docs endpoint returned HTTP $docs_code"
+fi
+echo ""
+
+# Summary
+echo "======================================"
+echo "Test Summary"
+echo "======================================"
+total=$((pass_count + fail_count))
+echo -e "${GREEN}Passed: $pass_count/$total${NC}"
+
+if [ $fail_count -gt 0 ]; then
+    echo -e "${RED}Failed: $fail_count/$total${NC}"
+    echo ""
+    echo -e "${RED}Some tests failed!${NC}"
+    exit 1
+else
+    echo ""
+    echo -e "${GREEN}All tests passed! ✓${NC}"
+    exit 0
+fi