import glob, re, os, sys

sys.stdout.reconfigure(encoding='utf-8')

target_files = []
target_files.extend(glob.glob('resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/*.md'))
target_files.extend(glob.glob('resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문/*.md'))
target_files.append('resources/knowledge/original/원가계산/건설공사_표준품셈/04_기계설비부문/제13장_플랜트설비공사.md')

print(f"Total target files: {len(target_files)}")

ditto_matches = []
remarks_tables = []

for filepath in target_files:
    fname = os.path.basename(filepath)
    with open(filepath, 'r', encoding='utf-8') as f:
        lines = f.readlines()
    
    in_table = False
    headers = []
    current_sec = 'UNKNOWN'
    table_rows = []
    header_line = 0
    
    for idx, line in enumerate(lines):
        line_s = line.strip()
        sec_m = re.search(r'(\d+-\d+(?:-\d+)?(?:\s+[^\n(]+)?)', line_s)
        if sec_m and not line_s.startswith('|'):
            current_sec = sec_m.group(1).strip()
            
        if line_s.startswith('|') and not line_s.startswith('|---'):
            cols = [c.strip() for c in line_s.split('|')[1:-1]]
            if not in_table:
                # Check if this row is header
                if any('비고' in c or '비 고' in c or '구분' in c or '직종' in c or '규격' in c for c in cols):
                    headers = cols
                    in_table = True
                    header_line = idx + 1
                    table_rows = []
                    continue
            if in_table:
                table_rows.append((idx + 1, cols))
                for col_idx, col in enumerate(cols):
                    if col in ['〃', '"', '″', '”'] or '〃' in col:
                        ditto_matches.append({
                            'file': fname,
                            'sec': current_sec,
                            'line': idx + 1,
                            'col_idx': col_idx,
                            'val': col,
                            'row': cols
                        })
        else:
            if in_table and (line_s == '' or not line_s.startswith('|')):
                # Check table for remarks
                has_remarks = any('비고' in h or '비 고' in h for h in headers)
                if has_remarks or any(m['line'] >= header_line and m['line'] <= idx for m in ditto_matches):
                    remarks_tables.append({
                        'file': fname,
                        'sec': current_sec,
                        'header_line': header_line,
                        'headers': headers,
                        'row_count': len(table_rows),
                        'rows': table_rows
                    })
                in_table = False
                headers = []
                table_rows = []

print(f"Total ditto matches: {len(ditto_matches)}")
print(f"Total tables with remarks/dittos: {len(remarks_tables)}")

# Summary by section
by_sec = {}
for m in ditto_matches:
    k = f"{m['file']} :: {m['sec']}"
    by_sec.setdefault(k, []).append(m)

for k, v in by_sec.items():
    print(f"  {k} -> {len(v)} dittos (lines {v[0]['line']}~{v[-1]['line']})")