import pypdf, sys, os

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

reader = pypdf.PdfReader('resources/knowledge/original/원가계산/건설공사_표준품셈/2026년_건설공사_표준품셈.pdf')

targets = [
    ("2-8-7", "철골 안전망"),
    ("2-8-10", "수직형 추락방망"),
    ("2-11-5", "비산먼지"),
    ("2-11-7", "슬러지 제거"),
    ("4-4-4", "야자섬유매트"),
    ("5-1-3", "버팀보"),
    ("5-1-8", "복공판"),
    ("7-4-2", "앵커 긴결"),
    ("슬립폼", "설치 및 해체"),
    ("지수판", "소켓 연결"),
    ("정착구 설치", "쉬즈관"),
]

for sec, kw in targets:
    found = False
    for idx, page in enumerate(reader.pages[:300]): # common part is in first 300 pages
        txt = page.extract_text()
        if sec in txt or (kw in txt and ("품" in txt or "공" in txt)):
            print(f"=== Found '{sec}' on PDF page index {idx} ===")
            # print surrounding lines
            lines = txt.split('\n')
            for l_idx, l in enumerate(lines):
                if sec in l or kw in l:
                    start = max(0, l_idx - 3)
                    end = min(len(lines), l_idx + 15)
                    print("\n".join(lines[start:end]))
                    print("-" * 50)
                    found = True
                    break
        if found:
            break