# -*- coding: utf-8 -*-
"""KCOPA 공공누리(O) 부착 전수 라이브 재스캔 (2026-06-24)
페이지=본문 opentype0N / 게시판=상세 view.do 샘플 union. 시트 O 대조."""
import openpyxl, requests, re, sys
from urllib.parse import urljoin
from concurrent.futures import ThreadPoolExecutor
requests.packages.urllib3.disable_warnings()
H={'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64)'}
S=requests.Session(); S.headers.update(H); S.verify=False

def get(u):
    try:
        return S.get(u,timeout=25).text
    except Exception:
        return ''

def marks(html):
    return set(int(d) for d in re.findall(r'opentype0?([1-4])', html, re.I))

def view_links(html, base):
    h=html.replace('&','&')
    raw=re.findall(r'(view\.do\?article_seq=\d+)', h)+re.findall(r'(view\.do\?ntt_id=\d+)', h)
    seen=[]
    for v in raw:
        if v not in seen: seen.append(v)
    return [urljoin(base, v) for v in seen[:12]]

def scan_row(args):
    r,label,L,url=args
    if not url: return (r,label,L,set(),'NO_URL')
    html=get(url)
    if L=='게시판' and 'list.do' in url:
        vl=view_links(html, url)
        agg=set()
        if vl:
            for d in vl:
                agg|=marks(get(d))
            return (r,label,L,agg,f'bbs {len(vl)}상세')
        # program/gallery board: mark may be on list/contents itself
        agg=marks(html)
        return (r,label,L,agg,'list직접')
    else:
        return (r,label,L,marks(html),'page본문')

def main():
    wb=openpyxl.load_workbook('한국저작권보호원.xlsx'); ws=wb.active
    rows=[]
    for r in range(3,98):
        b=ws.cell(r,2).value
        L=ws.cell(r,12).value
        O=ws.cell(r,15).value or '미부착'
        hl=ws.cell(r,11).hyperlink
        url=hl.target if hl else ws.cell(r,11).value
        f=ws.cell(r,6).value or ws.cell(r,5).value
        rows.append((r,f,L,url,O))
    args=[(r,f,L,url) for (r,f,L,url,O) in rows]
    res={}
    with ThreadPoolExecutor(max_workers=8) as ex:
        for out in ex.map(scan_row, args):
            res[out[0]]=out
    def parse_O(o):
        return set(int(d) for d in re.findall(r'([1-4])유형', o or ''))
    print('ROW | 라벨 | 시트O | 검출 | 판정 | 방법')
    mismatch=[]
    for (r,f,L,url,O) in rows:
        _,_,_,det,how=res[r]
        sheet=parse_O(O)
        ok = (det==sheet)
        tag='OK' if ok else 'MISMATCH'
        if not ok: mismatch.append((r,f,O,sorted(det),how))
        if (det or sheet or not ok):
            print(f'{r:>3} | {str(f)[:28]:28s} | {O:12s} | {sorted(det)} | {tag} | {how}')
    print('\n=== MISMATCH', len(mismatch),'건 ===')
    for m in mismatch: print(m)

main()