"""엑셀 D~J vs verify_results path 비교.

분류:
- MATCH: D~J path가 추출된 path와 동일 (정규화 후)
- D_MATCH: D만 일치, 나머지 다름
- D_MISMATCH: D부터 다름
- EMPTY_PATH: path 추출 실패 (수동 확인 필요)
- HTTP_ERROR: 접속 실패
- EXTERNAL: 외부링크 (검증 제외)
"""
import sys, io, json, re
from collections import defaultdict, Counter
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')

with open(r'D:\01.프로젝트\DB수집\2주차\검토_리포트\verify_results.json', encoding='utf-8') as f:
    res = json.load(f)
with open(r'D:\01.프로젝트\DB수집\2주차\검토_리포트\excel_data.json', encoding='utf-8') as f:
    excel = json.load(f)

def norm(s):
    if not s: return ''
    s = re.sub(r'\s+', ' ', str(s)).strip()
    # 흔한 표기 차이 정리
    s = s.replace('·', '·').replace('∙', '·')
    return s

# 사이트별 비교
report = {}
for site, rows in excel.items():
    site_stats = Counter()
    site_details = []
    for r in rows:
        row = r['row']
        K = r['K']
        S = r['S']
        excel_path = [norm(v) for v in [r['D'], r['E'], r['F'], r['G'], r['H'], r['I'], r['J']] if v]
        # 외부링크
        if S == '외부링크':
            site_stats['EXTERNAL'] += 1
            continue
        if not K or not isinstance(K, str) or not K.startswith('http'):
            site_stats['NO_URL'] += 1
            continue
        v = res.get(str(row))
        if not v:
            site_stats['NOT_FETCHED'] += 1
            continue
        if v['http'] != 200:
            site_stats[f'HTTP_{v["http"]}'] += 1
            site_details.append({'row': row, 'tag': 'HTTP_ERROR', 'http': v['http'],
                                  'excel_path': excel_path, 'url': K})
            continue
        actual_path = [norm(x) for x in v['path']]
        if not actual_path:
            site_stats['EMPTY_PATH'] += 1
            site_details.append({'row': row, 'tag': 'EMPTY_PATH',
                                  'excel_path': excel_path, 'actual_path': [],
                                  'title': v.get('title', '')[:100], 'h2': v.get('h2', '')[:60],
                                  'url': K})
            continue
        # 비교 — 토큰 단위 정확 일치
        if excel_path == actual_path:
            site_stats['MATCH'] += 1
            continue
        # 부분 일치 — D(엑셀 첫 항목)이 actual_path의 어디든 일치하는지
        if excel_path and actual_path and excel_path[0] == actual_path[0]:
            # D 일치, 이후는 다름
            site_stats['D_MATCH'] += 1
            site_details.append({'row': row, 'tag': 'D_MATCH',
                                  'excel_path': excel_path, 'actual_path': actual_path,
                                  'url': K})
        elif excel_path and actual_path and excel_path[0] in actual_path:
            site_stats['D_IN_ACTUAL'] += 1
            site_details.append({'row': row, 'tag': 'D_IN_ACTUAL',
                                  'excel_path': excel_path, 'actual_path': actual_path,
                                  'url': K})
        else:
            site_stats['D_MISMATCH'] += 1
            site_details.append({'row': row, 'tag': 'D_MISMATCH',
                                  'excel_path': excel_path, 'actual_path': actual_path,
                                  'url': K})
    report[site] = {'stats': dict(site_stats), 'details': site_details}

# 통계 출력
print('=== 사이트별 비교 통계 ===')
for site, v in report.items():
    print(f'\n[{site}] 총 {sum(v["stats"].values())}건')
    for tag, n in v['stats'].items():
        print(f'  {tag}: {n}')

# 저장
with open(r'D:\01.프로젝트\DB수집\2주차\검토_리포트\compare_all.json', 'w', encoding='utf-8') as f:
    json.dump(report, f, ensure_ascii=False, indent=2)
print('\n[saved compare_all.json]')