import re, json, sys, io
from openpyxl import load_workbook
from unidecode import unidecode

sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

# Extract all checks from PowerShell script
ps_content = open(r'D:\VNPT-M\Tool\audit_tool\NEW\Update\audit_lhdsin\audit_check_script\audit_cis_windows.ps1', 'r', encoding='utf-8').read()
checks = re.findall(r'Write-CheckResult\s+"(.+?)"\s+"(.+?)"', ps_content)
print(f'PS script checks: {len(checks)}')

# Build criteria lookup: CheckId -> description
ps_criteria = {}
for cid, desc in checks:
    full = f"{cid}. {desc}"
    key = unidecode(full)
    ps_criteria[key] = (cid, full)

# Load Excel checklist
wb = load_workbook(r'D:\VNPT-M\Tool\audit_tool\NEW\Update\audit_lhdsin\config\Windows_Checklist.xlsx')
ws = wb.worksheets[0]

# Build row mapping: scan Excel for criteria IDs (like "1.1.1" in column A)
# Compare unidecode'd criteria text to match PS output
excel_rows = {}
print(f'\nScanning Excel rows 1-{ws.max_row} for criteria IDs...')
for row in range(1, ws.max_row + 1):
    val_a = ws.cell(row=row, column=1).value
    val_b = ws.cell(row=row, column=2).value
    
    # Try to find criteria ID in column A (like "1.1.1")
    text_a = str(val_a).strip() if val_a else ""
    text_b = str(val_b).strip() if val_b else ""
    
    # Check if this row has a criteria ID pattern (like "1.1.1" or "1.1.1.")
    # Also try matching the full text from column B
    combined = f"{text_a} {text_b}".strip()
    norm = unidecode(combined)
    
    if norm in ps_criteria:
        cid, full_text = ps_criteria[norm]
        excel_rows[row] = full_text
        del ps_criteria[norm]  # Remove matched

# Now try matching by CheckId prefix in column A
still_unmatched = dict(ps_criteria)
for cid_key, (cid, full_text) in still_unmatched.items():
    for row in range(1, ws.max_row + 1):
        val_a = str(ws.cell(row=row, column=1).value or "").strip()
        if val_a and cid in val_a:
            excel_rows[row] = full_text
            del ps_criteria[cid_key]
            break

print(f'\nMatched via full text: {len(excel_rows)}')
print(f'Matched via ID search: {194 - len(ps_criteria) - len(excel_rows)}')
print(f'Still unmatched: {len(ps_criteria)}')

# For unmatched, find the closest row by ID pattern
# Parse CheckId like "1.1.1" into section numbers
for cid_key, (cid, full_text) in list(ps_criteria.items()):
    parts = cid.split('.')
    # Try to find a row where column A contains this ID prefix
    found = False
    for row in range(1, ws.max_row + 1):
        val_a = str(ws.cell(row=row, column=1).value or "").strip()
        if val_a == cid or val_a == cid + '.':
            if row not in excel_rows:
                excel_rows[row] = full_text
                del ps_criteria[cid_key]
                found = True
                break
    if not found:
        print(f'  No Excel row for: [{cid}] {full_text[:60]}')

print(f'\nFinal matched rows: {len(excel_rows)}')
print(f'Final unmatched: {len(ps_criteria)}')

# Build new config
new_data = []
for row in sorted(excel_rows.keys()):
    new_data.append({str(row): excel_rows[row]})

new_config = {"data": new_data}

# Write new config
output_path = r'D:\VNPT-M\Tool\audit_tool\NEW\Update\audit_lhdsin\config\windows_config_new.json'
with open(output_path, 'w', encoding='utf-8') as f:
    json.dump(new_config, f, ensure_ascii=False, indent=4)

print(f'\nNew config written to: {output_path}')
print(f'Entries: {len(new_data)}')

# Print summary of row mapping
print('\nSample row mapping:')
for row in sorted(excel_rows.keys())[:10]:
    print(f'  Row {row}: {excel_rows[row][:70]}')