import sys, io
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
import openpyxl, importlib.util
from openpyxl.utils import get_column_letter
from collections import Counter
P = r'D:\01.프로젝트\DB수집\작업파일\광역_사이트맵\충청남도\8.서산시\충청남도_서산시.xlsx'
wb=openpyxl.load_workbook(P); ws=wb.active
M=list(ws.merged_cells.ranges)
def mof(r,c):
    for m in M:
        if m.min_row<=r<=m.max_row and m.min_col<=c<=m.max_col: return m
    return None
def eff(r,c):
    m=mof(r,c); return ws.cell(m.min_row,m.min_col).value if m else ws.cell(r,c).value
data=[r for r in range(3,ws.max_row+1) if ws.cell(r,2).value not in (None,'')]
print(f'시트:{ws.title}  데이터행:{len(data)}  max_row:{ws.max_row}  총병합:{len(M)}')
print('헤더병합:', [h for h in ('B1:R1','S1:W1','Y1:AA1') if h in {str(m) for m in M}])

# 1) 구조검수 (검수체크 재사용)
spec=importlib.util.spec_from_file_location('chk', r'D:\01.프로젝트\DB수집\작업파일\광역_사이트맵\_검수체크.py')
chk=importlib.util.module_from_spec(spec); spec.loader.exec_module(chk)
n,issues=chk.inspect(P)
print(f'\n=== 구조검수: {"PASS" if not issues else "FAIL "+str(len(issues))+"건"} ===')
for x in issues: print('  ✗',x)

# 2) 정렬 분포 (가로/세로/wrap) per 주요 컬럼
print('\n=== 정렬 분포 (열: (horiz,vert,wrap) -> 행수) ===')
for c in (2,4,5,6,7,11,13,14,15):
    cnt=Counter()
    for r in data:
        a=ws.cell(r,c).alignment
        cnt[(a.horizontal, a.vertical, a.wrap_text)]+=1
    print(f'  {get_column_letter(c)}: {dict(cnt)}')

# 3) 행높이
hc=Counter(ws.row_dimensions[r].height for r in range(1,ws.max_row+1))
print('\n행높이 분포:', dict(hc))

# 4) 미병합 부모그룹 탐지 (E5,F6,G7)
def parent_groups(c):
    groups=[]; i=0
    while i<len(data):
        r=data[i]; v=ws.cell(r,c).value
        if v in (None,'') or (mof(r,c) and mof(r,c).min_row!=r): i+=1; continue
        parent=tuple(eff(r,cc) for cc in range(4,c)); rows=[r]; j=i+1
        while j<len(data):
            r2=data[j]
            if tuple(eff(r2,cc) for cc in range(4,c))!=parent: break
            cv=ws.cell(r2,c).value; m2=mof(r2,c)
            if (cv in (None,'')) or (m2 and m2.min_row<r2): rows.append(r2); j+=1
            else: break
        if len(rows)>1 and any(ws.cell(rr,c+1).value not in (None,'') for rr in rows):
            m=mof(r,c); groups.append((rows[0],rows[-1],v,bool(m and m.max_row>m.min_row))); i=j
        else: i+=1
    return groups
print('\n=== 부모그룹 병합상태 (★비병합=이상후보) ===')
for c,nm in [(5,'E'),(6,'F'),(7,'G')]:
    gs=parent_groups(c)
    if not gs: continue
    un=[g for g in gs if not g[3]]
    print(f'  {nm}: 총{len(gs)} 병합{len(gs)-len(un)} ★비병합{len(un)}')
    for a,z,v,mg in un:
        print(f'     ★ {nm}{a}:{z} {v!r} (E={eff(a,5)!r} F={eff(a,6)!r})')