Spaces:
Sleeping
Sleeping
File size: 6,851 Bytes
48cec82 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 | #!/usr/bin/env python3
"""
Test script to verify the Firebase loading and content processing improvements.
"""
import os
import sys
# Add the parent directory to Python path
current_dir = os.path.dirname(os.path.abspath(__file__))
sys.path.insert(0, current_dir)
def test_firebase_loading():
"""Test Firebase loading with language filtering."""
print("π§ͺ Testing Firebase Loading and Content Processing")
print("=" * 60)
try:
from cve_factchecker.firebase_loader import FirebaseNewsLoader
# Test Firebase loader
loader = FirebaseNewsLoader()
print(f"β
Firebase loader initialized - Project: {loader.project_id}")
# Test with very small limit first
print("\nπ Testing with small sample (5 articles)...")
articles = loader.fetch_articles(limit=5)
if articles:
print(f"β
Fetched {len(articles)} sample articles")
# Analyze the first article
first_article = articles[0]
print(f"\nπ Sample Article Analysis:")
print(f" Title: {first_article.title[:100]}...")
print(f" Content length: {len(first_article.content)} characters")
print(f" URL: {first_article.url}")
print(f" Source: {first_article.source}")
print(f" Article ID: {first_article.article_id}")
# Show content preview
if first_article.content:
content_preview = first_article.content[:300].replace('\n', ' ')
print(f" Content preview: {content_preview}...")
else:
print(" β οΈ No content found!")
# Test language-specific fetching
print("\nπ Testing English language filtering...")
english_articles = loader.fetch_articles_by_language("English", limit=10)
if english_articles:
print(f"β
Fetched {len(english_articles)} English articles")
# Check content quality
valid_content = 0
for article in english_articles[:3]: # Check first 3
if article.content and len(article.content) > 100:
valid_content += 1
print(f" π '{article.title[:50]}...' - {len(article.content)} chars")
else:
print(f" β οΈ '{article.title[:50]}...' - insufficient content")
print(f" Content quality: {valid_content}/{min(3, len(english_articles))} articles have substantial content")
else:
print("β No English articles found")
return english_articles
except Exception as e:
print(f"β Firebase test failed: {e}")
import traceback
traceback.print_exc()
return []
def test_vector_processing(articles):
"""Test vector store processing."""
print("\nπ Testing Vector Store Processing")
print("=" * 60)
try:
from cve_factchecker.retriever import VectorRetriever
# Test vector retriever
retriever = VectorRetriever(persist_directory="/tmp/test_vector_db")
print("β
Vector retriever initialized")
# Test article storage
print(f"\nπ¦ Testing storage of {len(articles)} articles...")
retriever.store_articles_in_vector_db(articles, clear_first=True)
# Test retrieval
print("\nπ Testing document retrieval...")
test_query = "security vulnerability"
results = retriever.search(test_query, k=3)
if results:
print(f"β
Found {len(results)} relevant documents for '{test_query}'")
for i, doc in enumerate(results):
content_preview = doc.page_content[:100].replace('\n', ' ')
print(f" {i+1}. {content_preview}...")
else:
print("β οΈ No documents found for test query")
return True
except Exception as e:
print(f"β Vector processing test failed: {e}")
import traceback
traceback.print_exc()
return False
def test_full_system():
"""Test the complete system integration."""
print("\nπ Testing Full System Integration")
print("=" * 60)
try:
from cve_factchecker.orchestrator import FactCheckSystem
# Test system initialization
system = FactCheckSystem(vector_dir="/tmp/test_system_vector")
print("β
Fact check system initialized")
# Test Firebase ingestion
print("\nπ Testing Firebase ingestion...")
result = system.ingest_firebase(limit=10)
if result.get("success"):
print(f"β
Ingestion successful: {result.get('synced')} articles")
else:
print(f"β οΈ Ingestion issues: {result.get('error', 'Unknown error')}")
# Test fact checking
print("\nπ§ Testing fact checking...")
test_claim = "A new security vulnerability was discovered in popular software"
fact_result = system.fact_check(test_claim)
print(f" Claim: {test_claim}")
print(f" Verdict: {fact_result.get('verdict', 'Unknown')}")
print(f" Confidence: {fact_result.get('confidence', 0)}")
reasoning = fact_result.get('reasoning', 'No reasoning provided')
print(f" Reasoning: {reasoning[:200]}...")
return True
except Exception as e:
print(f"β Full system test failed: {e}")
import traceback
traceback.print_exc()
return False
def main():
"""Run all tests."""
print("π§ͺ CVE Fact Checker - Comprehensive Test Suite")
print("=" * 80)
# Test 1: Firebase loading
articles = test_firebase_loading()
# Test 2: Vector processing (if we have articles)
if articles:
vector_success = test_vector_processing(articles)
else:
print("\nβ οΈ Skipping vector processing test - no articles loaded")
vector_success = False
# Test 3: Full system integration
system_success = test_full_system()
# Summary
print("\nπ Test Results Summary")
print("=" * 80)
print(f"Firebase Loading: {'β
PASS' if articles else 'β FAIL'}")
print(f"Vector Processing: {'β
PASS' if vector_success else 'β FAIL'}")
print(f"System Integration: {'β
PASS' if system_success else 'β FAIL'}")
overall_success = bool(articles) and vector_success and system_success
print(f"\nOverall Result: {'β
ALL TESTS PASSED' if overall_success else 'β SOME TESTS FAILED'}")
return overall_success
if __name__ == "__main__":
success = main()
sys.exit(0 if success else 1) |