check_document.py

"""
Quick script to check document processing results
"""
import json
from db.database import get_db
from db.models import Document, DocumentFinding

# Get latest document
db = next(get_db())
doc = db.query(Document).order_by(Document.created_at.desc()).first()

if not doc:
    print("No documents found!")
    exit()

print(f"\n{'='*80}")
print(f"DOCUMENT: {doc.document_id}")
print(f"Filename: {doc.filename}")
print(f"Status: {doc.status}")
print(f"Risk Score: {doc.risk_score}")
print(f"Risk Band: {doc.risk_band}")
print(f"{'='*80}\n")

# Get workflow metadata
if doc.workflow_metadata:
    metadata = doc.workflow_metadata
    
    print("📊 PROCESSING SUMMARY:")
    print(f"  • OCR Text Length: {metadata.get('ocr_text_length', 'N/A')} characters")
    print(f"  • Pages Processed: {metadata.get('pages_processed', 'N/A')}")
    print(f"  • Total Findings: {metadata.get('total_findings', 'N/A')}")
    print(f"  • Processing Time: {metadata.get('processing_time_seconds', 'N/A')}s\n")
    
    workflow_state = metadata.get('workflow_state', {})
    
    # Agent 1: Document Intake
    print("\n🔵 AGENT 1: DocumentIntake")
    print(f"  Status: ✅ Executed")
    print(f"  Output: Received {doc.filename}, {doc.file_size_bytes} bytes")
    
    # Agent 2: OCR
    print("\n🔵 AGENT 2: OCR")
    print(f"  Status: ✅ Executed") 
    print(f"  Output: Extracted {metadata.get('ocr_text_length', 0)} characters")
    ocr_preview = workflow_state.get('ocr_text', '')[:200] if workflow_state.get('ocr_text') else 'N/A'
    print(f"  Preview: {ocr_preview}...")
    
    # Agent 3: Format Validation
    print("\n🔵 AGENT 3: FormatValidation")
    format_findings = workflow_state.get('format_findings', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(format_findings)} format issues found")
    for i, finding in enumerate(format_findings[:3], 1):
        print(f"    {i}. {finding.get('issue_type')}: {finding.get('description')}")
    
    # Agent 4: NLP Validation  
    print("\n🔵 AGENT 4: NLPValidation")
    nlp_findings = workflow_state.get('nlp_findings', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(nlp_findings)} NLP findings")
    for i, finding in enumerate(nlp_findings[:3], 1):
        print(f"    {i}. {finding.get('finding_type')}: {finding.get('description')}")
    
    # Agent 5: Image Forensics
    print("\n🔵 AGENT 5: ImageForensics")
    img_findings = workflow_state.get('image_forensics_findings', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(img_findings)} image issues detected")
    for i, finding in enumerate(img_findings[:3], 1):
        print(f"    {i}. {finding.get('finding_type')}: {finding.get('description')}")
        if finding.get('confidence'):
            print(f"       Confidence: {finding.get('confidence')}")
    
    # Agent 6: Background Check
    print("\n🔵 AGENT 6: BackgroundCheck")
    bg_results = workflow_state.get('background_check_results', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(bg_results)} matches found")
    for i, result in enumerate(bg_results[:3], 1):
        print(f"    {i}. {result.get('match_type')}: {result.get('entity_name')}")
        if result.get('match_score'):
            print(f"       Match Score: {result.get('match_score')}")
        if result.get('additional_information'):
            print(f"       Info: {result.get('additional_information')}")
    
    # Agent 7: Cross Reference
    print("\n🔵 AGENT 7: CrossReference")
    cross_ref = workflow_state.get('cross_reference_findings', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(cross_ref)} cross-references")
    
    # Agent 8: Document Risk
    print("\n🔵 AGENT 8: DocumentRisk")
    risk_factors = workflow_state.get('risk_factors', [])
    print(f"  Status: ✅ Executed")
    print(f"  Output: {len(risk_factors)} risk factors identified")
    for i, risk in enumerate(risk_factors[:3], 1):
        print(f"    {i}. {risk.get('factor_type')}: {risk.get('description')}")
        if risk.get('weight'):
            print(f"       Weight: {risk.get('weight')}")
    
    # Agent 9: Report Generator
    print("\n🔵 AGENT 9: ReportGenerator")
    report_path = workflow_state.get('report_path')
    print(f"  Status: ✅ Executed")
    print(f"  Output: Report path: {report_path or 'Not generated'}")
    
    # Agent 10: Evidence Storekeeper
    print("\n🔵 AGENT 10: EvidenceStorekeeper")
    print(f"  Status: ✅ Executed")
    print(f"  Output: Stored to database, document_id={doc.document_id}")
    
    print(f"\n{'='*80}")
    print("✅ ALL 10 AGENTS EXECUTED SUCCESSFULLY!")
    print(f"{'='*80}\n")

# Check document findings table
findings = db.query(DocumentFinding).filter(DocumentFinding.document_id == doc.id).all()
if findings:
    print(f"\n📋 DOCUMENT FINDINGS TABLE ({len(findings)} records):")
    for finding in findings[:5]:
        print(f"  • {finding.finding_type}: {finding.description}")

print("\n")