import sys, io
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
import openpyxl
from openpyxl.utils import get_column_letter
ws = openpyxl.load_workbook(r'D:\01.프로젝트\DB수집\작업파일\_제출패키지_2026-06-03\프리\충청남도_부여군.xlsx'.replace('프리','프리랜서1')).active
merges=list(ws.merged_cells.ranges)
def merge_of(r,c):
    for m in merges:
        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=merge_of(r,c)
    if m: return ws.cell(m.min_row,m.min_col).value
    return ws.cell(r,c).value
def raw(r,c): return 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,'')]
dset=set(data)

# For each category column, find a standalone value cell followed by >=1 None-unmerged cell
# under the SAME parent (cols < c effective equal) -> an UNMERGED group (merge removed).
print('=== 병합이 풀린(또는 누락된) 카테고리 그룹 후보 ===')
found=[]
for c in (4,5,6,7):  # D,E,F,G
    for r in data:
        if merge_of(r,c): continue            # this cell is in a merge -> ok
        v=raw(r,c)
        if v in (None,''): continue           # need a value to be the group head
        # scan down for trailing None-unmerged under same parent (cols < c effective same)
        parent=tuple(eff(r,cc) for cc in range(4,c))
        run=[]
        r2=r+1
        while r2 in dset:
            # parent must match
            if tuple(eff(r2,cc) for cc in range(4,c))!=parent: break
            cell_in_merge=merge_of(r2,c)
            cv=raw(r2,c)
            if cv in (None,'') and not cell_in_merge:
                run.append(r2); r2+=1; continue
            break
        if run:
            found.append((c,r,run[-1],v,len(run)+1))

for c,a,z,v,n in found:
    col=get_column_letter(c)
    print(f'  {col}{a}:{col}{z}  값={v!r}  (행 {a}~{z}, {n}행이 병합 안 됨)  D={eff(a,4)!r} E={eff(a,5)!r} F={eff(a,6)!r}')
if not found:
    print('  없음')

# Also: list every None+unmerged cell in D/E/F/G that sits under a value (orphan blanks)
print('\n=== D/E/F/G 빈칸+비병합 셀(고아 빈칸) 전수 ===')
for c in (4,5,6,7):
    col=get_column_letter(c)
    orph=[r for r in data if raw(r,c) in (None,'') and not merge_of(r,c)]
    # filter: those that have a value somewhere above in same column within 50 rows (likely should inherit)
    print(f'  {col}: {len(orph)}개', orph[:30])