import re, os, glob, sys

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

base_dirs = [
    'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문',
    'resources/knowledge/original/원가계산/건설공사_표준품셈/03_건축부문',
    'resources/knowledge/original/원가계산/건설공사_표준품셈/04_기계설비부문',
    'resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문',
]

results = []

for bdir in base_dirs:
    for fpath in glob.glob(os.path.join(bdir, '*.md')):
        fname = os.path.basename(fpath)
        with open(fpath, encoding='utf-8') as f:
            lines = f.readlines()
        
        in_table = False
        table_start = 0
        table_lines = []
        
        for idx, line in enumerate(lines):
            stripped = line.strip()
            if stripped.startswith('|') and stripped.endswith('|'):
                if not in_table:
                    in_table = True
                    table_start = idx + 1
                    table_lines = []
                table_lines.append((idx + 1, stripped))
            else:
                if in_table:
                    in_table = False
                    # Analyze table
                    # Check if any row has cells with multiple numbers separated by spaces (squashed rows)
                    for lno, row_str in table_lines[2:]: # skip header and divider
                        cells = [c.strip() for c in row_str.split('|')[1:-1]]
                        for c_idx, cell in enumerate(cells):
                            # check if cell has 4+ space-separated numbers or words
                            parts = cell.split()
                            if len(parts) >= 4 and any(re.match(r'^[0-9~.()%]+$', p) for p in parts):
                                # Could be squashed
                                results.append({
                                    'file': fname,
                                    'line': lno,
                                    'cell_idx': c_idx,
                                    'cell': cell[:80]
                                })
                                break

print(f"Found {len(results)} potential squashed rows:")
for r in results:
    print(f"  {r['file']} L{r['line']}: [Col {r['cell_idx']}] {r['cell']}")