import os, sys
sys.path.append('scratch')
from find_strict_grouped_tables import true_grouped_tables

sys.stdout.reconfigure(encoding='utf-8')
plant_tables = [t for t in true_grouped_tables if t['file'] == '제13장_플랜트설비공사.md']

fpath = 'resources/knowledge/original/원가계산/건설공사_표준품셈/04_기계설비부문/제13장_플랜트설비공사.md'
with open(fpath, 'r', encoding='utf-8') as f:
    lines = f.readlines()

print(f"Total plant grouped tables: {len(plant_tables)}")

# Filter for tables with process / work categories (like 공정별 설치, 작업구분, etc.)
process_tables = []
for t in plant_tables:
    # check title or groups
    title = t['title']
    groups = t['groups']
    if any(k in title for k in ['공정별', '작업구분', '공정', '설치', '조립', '소각', '보일러', '터빈', '발전기', '수차']) or \
       any(any(w in g for w in ['조립', '설치', '해체', '관리', '검사', 'Chipping', '손질']) for g in groups):
        process_tables.append(t)

print(f"Process / Work classification tables in Plant: {len(process_tables)}")
for pt in process_tables:
    print(f"\n=======================================================")
    print(f"L{pt['line']}: {pt['title']}")
    print(f"Groups: {pt['groups'][:5]}")
    start = max(0, pt['line'] - 1)
    end = min(len(lines), pt['line'] + pt['total_rows'] + 2)
    for i in range(start, min(end, start + 25)):
        print(f"{i+1:4d}: {lines[i].rstrip()}")