from sys import stdin, stderr
from collections import deque
from copy import copy



def beste_sti(nm, prob):

    cProb = [0]*len(nm)   #Current probability at each node
    done = []             # Done nodes
    pi=[-1]*len(nm)     #Predecessor array

    cProb[0] = prob[0]


    Q = [0]  #Available nodes.


    while Q:
      print "Current Probability: ",cProb
      print "Parent: ",pi
      print "Q: ",Q
      print "Done: ", done

      #Finds the node with max probability
      # Picks highest node from Q that is not finished.

      maxNode = -1
      for i in range(len(Q)):
        u = Q[i]
        if cProb[u] > cProb[maxNode]:
          maxNode = u

      if maxNode == -1:
        return

      print "maxNode: ", maxNode
      print ""

      Q.remove(maxNode)
      done.append(maxNode)

      for nabo in range(len(nm)):
        isNabo = nm[maxNode][nabo]

        #Just adds the nabo to Q if it isnt done
        if isNabo == 1:
          if not nabo in Q and not nabo in done:
            Q.append(nabo)
        #Gives new probability if possible
          if cProb[maxNode]*prob[nabo] > cProb[nabo]:
            cProb[nabo] = cProb[maxNode]*prob[nabo]
            pi[nabo] = maxNode

    streng = ""

    #Find the path from start to stop.

    curr = len(nm)-1
    streng+=str(curr)
    #print "Starting backtracking"
    while not pi[curr] == -1:
      #print "Path: ",streng
      # print "Current: ",curr
      if prob[curr] == 0:
        return 0
      streng+="-"+str(pi[curr])
      curr = pi[curr]


    #print streng
    streng = streng[::-1]
    #print streng


    return streng















n = int(stdin.readline())
sansynligheter = [float(x) for x in stdin.readline().split()]
nabomatrise = []
for linje in stdin:
    naborad = [0] * n
    naboer = [int(nabo) for nabo in linje.split()]
    for nabo in naboer:
        naborad[nabo] = 1
    nabomatrise.append(naborad)
print beste_sti(nabomatrise, sansynligheter)