# -*- coding: utf-8 -*-
import openpyxl, io, re
from collections import defaultdict
wb=openpyxl.load_workbook("한국지역난방공사.xlsx"); ws=wb.active
out=io.open("_menuno_out.txt","w",encoding="utf-8")
def w(*a): out.write(" ".join(str(x) for x in a)+"\n")
def sig(K):
    if not K: return None
    K=str(K)
    m=re.search(r'menuNo=(\d+)',K)
    b=re.search(r'/(B\d{6,})/',K)
    mc=re.search(r'menuCode=(M\d+)',K)
    parts=[]
    if b: parts.append("bbs="+b.group(1))
    if m: parts.append("menuNo="+m.group(1))
    if mc: parts.append("menuCode="+mc.group(1))
    return "|".join(parts) if parts else K[:50]
groups=defaultdict(list)
for r in range(3, ws.max_row+1):
    K=ws.cell(r,11).value
    s=sig(K)
    if s: groups[s].append(r)
w("=== same page signature (menuNo / bbs / menuCode) on >1 row ===")
cnt=0
for s,rs in groups.items():
    if len(rs)>1:
        cnt+=1
        for rr in rs:
            d=[ws.cell(rr,c).value for c in range(4,11)]
            path=" > ".join(str(x).strip() for x in d if x not in(None,""))
            w("  r%d B%s L=%s M=%s | %s"%(rr,ws.cell(rr,2).value,ws.cell(rr,12).value,ws.cell(rr,13).value,path))
        w("  SIG:",s,"K0=",str(ws.cell(rs[0],11).value),"\n")
w("groups with dup signature:",cnt)
out.close(); print("done")