import requests
import os
import json

# Test uploading the new test_ocr_content.pdf file
base_url = 'http://localhost:3015'
pdf_file = 'test_ocr_content.pdf'

# Check if the file exists
if not os.path.exists(pdf_file):
    print(f"❌ {pdf_file} not found in current directory")
    exit(1)

# Login first
print("🔐 Logging in...")
login_data = {'username': 'jleu3482', 'password': 'jleu1212'}
login_response = requests.post(f'{base_url}/login', data=login_data)

if login_response.status_code == 200:
    token = login_response.json().get('access_token')
    headers = {'Authorization': f'Bearer {token}'}
    print('✅ Login successful')
    
    # Upload the new PDF
    print(f"📤 Uploading {pdf_file}...")
    with open(pdf_file, 'rb') as file:
        files = {'file': (pdf_file, file, 'application/pdf')}
        upload_response = requests.post(f'{base_url}/documents/upload', files=files, headers=headers)
    
    print(f"  Upload Status: {upload_response.status_code}")
    if upload_response.status_code == 200:
        result = upload_response.json()
        print(f"  Response: {json.dumps(result, indent=2)}")
        
        if result.get('status') == 'duplicated':
            print("⚠️  File already exists, but let's test search anyway")
        else:
            print('✅ Upload successful')
            
    else:
        print(f'❌ Upload failed: {upload_response.text}')
        
    # Test search with specific content from the new PDF
    print("\n🔍 Testing search functionality...")
    
    # Test queries based on the content we know is in the PDF
    test_queries = [
        "test PDF document for OCR processing",
        "GPU acceleration for faster processing", 
        "Important Heading",
        "Numbers 1234567890",
        "special characters"
    ]
    
    for query in test_queries:
        print(f"\n🔎 Querying: \"{query}\"")
        query_data = {'query': query, 'top_k': 3}
        search_response = requests.post(f'{base_url}/query', json=query_data, headers=headers)
        
        if search_response.status_code == 200:
            results = search_response.json()
            if isinstance(results, dict):
                response_text = results.get('response', '')
                if query.lower() in response_text.lower():
                    print(f"  ✅ Found matching content!")
                else:
                    print(f"  ⚠️  No direct match, but got response")
                print(f"  Response: {response_text[:200]}...")
            else:
                print(f"  Unexpected result format: {results}")
        else:
            print(f'❌ Search failed: {search_response.text}')
            
else:
    print('❌ Login failed')