import sys
import re
sys.path.append(r"c:\Users\WIN10\Desktop\work\11th-week\法律风险提示-new\市监局-lawRisk-backend")

from lawrisk.services.licensing_repo import _normalize_import_row

def test_normalization():
    print("Testing _normalize_import_row...")
    
    # Mock a raw row that mimics what _parse_import_workbook produces
    raw_row = {
        "row_index": 10,
        "permit_name": None, # Should come from defaults if missing
        "risk_content": "Test Risk",
        "legal_basis": "Law 101",
        "document_no": "No.1",
        "summary": "Summary",
        "permit_status": "Active",
        "filler_name": "John Doe",
        "unit_name": "Test Unit",
        "source_update_date": "2023-12-19",
        "responsible_contact": "Contact Info",
        "jurisdiction_scope": "City Level",
        "subitem_text": "Subitem A、Subitem B",
        "theme_names": "Theme X", # Should be ignored/cleared if not in new template, but if passed in raw_row...
        "scope_text": "Scope Y",
    }
    
    sheet_defaults = {
        "permit_name": "Test Permit",
        "filler_name": "Default Filler", # Should use raw_row if present
        "unit_name": "Default Unit",
        "source_update_date": "2023-01-01",
    }
    
    normalized = _normalize_import_row(raw_row, "Sheet1", sheet_defaults)
    
    if not normalized:
        print("FAILED: Returned None")
        return

    # Check new fields
    print(f"Filler Name: {normalized.get('filler_name')} (Expected: John Doe)")
    print(f"Unit Name: {normalized.get('unit_name')} (Expected: Test Unit)")
    print(f"Update Date: {normalized.get('source_update_date')} (Expected: 2023-12-19)")
    
    # Check Permit Name (fallback)
    print(f"Permit Name: {normalized.get('permit_name')} (Expected: Test Permit)")
    
    # Check removed fields are gone or handled
    if "subitem_summary" in normalized:
        print("WARNING: subitem_summary still present")
    if "theme_names" in normalized:
        # It might be present but empty list if we removed extraction logic?
        # Wait, I removed theme_names extraction from _normalize_import_row.
        # So it should NOT be in the output dict unless I left it.
        # Let's check.
        pass
        
    # Validation
    assert normalized['filler_name'] == "John Doe"
    assert normalized['unit_name'] == "Test Unit"
    assert normalized['source_update_date'] == "2023-12-19"
    assert normalized['permit_name'] == "Test Permit"
    
    print("SUCCESS: Normalization logic verified.")

if __name__ == "__main__":
    test_normalization()