# -*- coding: utf-8 -*- import openpyxl, io wb=openpyxl.load_workbook("한국지역난방공사.xlsx"); ws=wb.active out=io.open("_mergecand_out.txt","w",encoding="utf-8") def w(*a): out.write(" ".join(str(x) for x in a)+"\n") COLS={4:'D',5:'E',6:'F',7:'G',8:'H',9:'I',10:'J'} mx=ws.max_row # existing merged ranges by (col)->list of (r1,r2) merged_cols={} for mr in ws.merged_cells.ranges: if mr.min_col==mr.max_col: merged_cols.setdefault(mr.min_col,[]).append((mr.min_row,mr.max_row)) def already_merged(c,r1,r2): for (a,b) in merged_cols.get(c,[]): if a<=r1 and b>=r2: return True return False # literal value def lit(r,c): v=ws.cell(r,c).value return str(v).strip() if v not in(None,"") else "" # effective parent path D..c-1 (fill merged) eff={} for r in range(3,mx+1): for c in range(4,11): eff[(r,c)]=ws.cell(r,c).value for mr in ws.merged_cells.ranges: if mr.min_col>=4 and mr.max_col<=10: v=ws.cell(mr.min_row,mr.min_col).value for r in range(mr.min_row,mr.max_row+1): for c in range(mr.min_col,mr.max_col+1): eff[(r,c)]=v def effv(r,c): v=eff[(r,c)]; return str(v).strip() if v not in(None,"") else "" def parentpath(r,c): # D..c-1 effective return tuple(effv(r,cc) for cc in range(4,c)) total=0 for c in range(4,11): cand=[] r=3 while r<=mx: v=lit(r,c) if not v: r+=1; continue # extend run while next row literal == v AND same parent path r2=r while r2+1<=mx and lit(r2+1,c)==v and parentpath(r2+1,c)==parentpath(r,c): r2+=1 if r2>r and not already_merged(c,r,r2): cand.append((r,r2,v,parentpath(r,c))) r=r2+1 if cand: w(f"\n=== column {COLS[c]} — {len(cand)} unmerged repeated-value runs ===") for (r1,r2,v,pp) in cand: total+=1 pps=" > ".join(x for x in pp if x) w(f" {COLS[c]}{r1}:{COLS[c]}{r2} '{v}' [부모: {pps if pps else '(대분류)'}] (행 {r1}~{r2}, {r2-r1+1}개)") w(f"\nTOTAL merge-candidate runs: {total}") out.close(); print("done total",total)