Skip to content

TEST_API_CONNECTION.md

Latest commit

 

History

History
286 lines (202 loc) · 6.78 KB

TEST_API_CONNECTION.md

File metadata and controls

286 lines (202 loc) · 6.78 KB

🧪 Test API Connection - Quick Guide

How to Test Python Backend from Dashboard

Step 1: Start Python Backend

cd /Users/anshulmangalapalli/Documents/GitHub/hackprinceton
python3 api_server.py

Wait for:

🚀 KALSHI HYBRID ANALYSIS API STARTING
Docs: http://localhost:8000/docs

Step 2: Start Dashboard

cd dashboard
npm run dev

Step 3: Open Dashboard & Console

  1. Open http://localhost:3000/dashboard in browser
  2. Open browser console (Press F12 or Right-click → Inspect → Console)

Step 4: Run Test

In the browser console, type:

window.testAPI();

Press Enter.

📺 What You'll See

Console Output (Success):

================================================================================
🧪 TESTING PYTHON API CONNECTION
================================================================================
API Base URL: http://localhost:8000
Time: 11/9/2025, 2:30:45 PM
================================================================================

📡 Test 1: Health Check
   Calling: GET /
   ✅ Status: 200
   📦 Response: {
     "status": "online",
     "service": "Kalshi Hybrid Analysis API",
     "version": "1.0.0"
   }

📡 Test 2: Status Check
   Calling: GET /api/status
   ✅ Status: 200
   📦 Response: {
     "status": "ready",
     "isAnalyzing": false,
     "lastUpdated": "2025-11-09T14:30:52.123Z",
     "count": 12,
     "error": null
   }

📡 Test 3: Get Recommendations
   Calling: GET /api/recommendations
   ✅ Status: 200
   📊 Response Status: success
   📦 Opportunities Count: 12

   🎯 SAMPLE RECOMMENDATION:
   --------------------------------------------------------------------------------
   Ticker: KXHIGHCHI-25NOV09-T38
   Title: Will the high temp in Chicago be <38° on Nov 9, 2025?
   Category: Weather
   Action: BUY_YES
   Quant Edge: 79.0%
   Market Prob: 0.5%
   Combined Confidence: 88.0%

   🤖 GROK SENTIMENT:
   Label: positive
   Score: 82%
   Confidence: high
   Key Themes: cold front, arctic blast, freezing

   💡 REASONING:
   Statistical analysis shows a 79.0% edge. Models predict 79.5% probability vs
   market price of 0.5%. Grok AI sentiment analysis: positive (82%). Key themes:
   cold front, arctic blast, freezing. ✅ Both quantitative and sentiment signals
   AGREE, increasing confidence.
   --------------------------------------------------------------------------------

   📊 ALL RECOMMENDATIONS:
   1. KXHIGHCHI-25NOV09-T38 - BUY_YES (Edge: 79.0%)
   2. KXHIGHDEN-25NOV08-B53.5 - BUY_NO (Edge: -90.9%)
   3. KXHIGHCHI-25NOV09-B40.5 - BUY_NO (Edge: -77.5%)
   4. KXHIGHDEN-25NOV08-T51 - BUY_YES (Edge: 63.7%)
   5. KXHIGHNY-25NOV09-T59 - BUY_YES (Edge: 62.8%)
   ... (7 more)

================================================================================
✅ API CONNECTION TEST COMPLETE
================================================================================

Console Output (Backend Analyzing):

📡 Test 3: Get Recommendations
   Calling: GET /api/recommendations
   ✅ Status: 200
   📊 Response Status: analyzing
   📦 Opportunities Count: 0

   ⏳ Backend is currently analyzing markets
   💡 Wait 30-60 seconds and try again

================================================================================
✅ API CONNECTION TEST COMPLETE
================================================================================

Console Output (Error):

================================================================================
❌ API CONNECTION TEST FAILED
================================================================================

🔥 Error: Failed to fetch

💡 Troubleshooting:
   1. Is Python backend running? (python3 api_server.py)
   2. Is it running on port 8000?
   3. Check Terminal 1 for backend errors
   4. Try: curl http://localhost:8000/api/status
================================================================================

🔍 What Gets Tested

  1. Health Check (GET /)

    • Verifies backend is online
    • Returns service info

  2. Status Check (GET /api/status)

    • Checks if analysis is running
    • Shows last update time
    • Shows recommendation count

  3. Get Recommendations (GET /api/recommendations)

    • Fetches actual recommendations
    • Shows full details of first recommendation
    • Shows list of all recommendations

🐛 Troubleshooting

Problem: "Failed to fetch"

Cause: Backend not running or wrong port

Fix:

  1. Check Terminal 1 - is python3 api_server.py running?
  2. Should show: Docs: http://localhost:8000/docs
  3. Try: curl http://localhost:8000/api/status

Problem: "Response Status: empty"

Cause: Backend hasn't finished first analysis yet

Fix: Wait 30-60 seconds, run window.testAPI() again

Problem: "Response Status: analyzing"

Cause: Backend is currently analyzing (normal!)

Fix: Wait 30-60 seconds, run window.testAPI() again

Problem: CORS error

Cause: Backend and dashboard on different domains

Fix:

  1. Ensure backend is on localhost:8000
  2. Ensure dashboard is on localhost:3000
  3. Restart both if needed

📊 Python Backend Terminal Output

When you run window.testAPI(), you'll see this in Terminal 1:

📥 GET / - Request received at 14:31:20

📥 GET /api/status - Request received at 14:31:20
   ✅ Returning status: ready

📥 GET /api/recommendations - Request received at 14:31:20
   ✅ Returning 12 recommendations
   📅 Last updated: 14:30:52

This confirms the backend is receiving and responding to requests!

🎯 Success Criteria

Your API connection is working if:

  • ✅ All 3 tests pass (Health, Status, Recommendations)
  • ✅ You see recommendations in console
  • ✅ Backend terminal shows incoming requests
  • ✅ No error messages

🔄 Alternative: Test with curl

You can also test the API directly from terminal:

# Health check
curl http://localhost:8000/

# Status check
curl http://localhost:8000/api/status

# Get recommendations
curl http://localhost:8000/api/recommendations | jq

(Install jq for pretty JSON: brew install jq)

📝 What This Proves

Running window.testAPI() proves:

  1. ✅ Dashboard can reach Python backend
  2. ✅ Backend is responding correctly
  3. ✅ API endpoints are working
  4. ✅ Data is flowing correctly
  5. ✅ Recommendations are being generated

Perfect for showing judges: "Here's the live API connection!" 🎓

🎬 For Demo

During HackPrinceton demo:

  1. Open dashboard
  2. Open console (F12)
  3. Type: window.testAPI()
  4. Show judges the output
  5. Point out:
    • Real-time API calls
    • Python backend responding
    • Actual recommendations with Grok sentiment
    • All the data flowing through

This is EXTREMELY impressive! Shows both pretty UI AND technical depth. 🚀