What is wrong with python below? Any suggestions for another solution?

Question

Problem: Having a 5 5 matrix filled with random (real) values between 0 and 99, show which is the second highest value existing in the matrix.

Can’t use numpy or similar.

My solution:

n=2
def constroi_matriz(n):
    import random
    matriz =[[]]
    linha =[]
    for i in range(n):
        for j in range(n):
            num = random.randint(0,99)
            linha.append(num)
        matriz.append(linha)
        linha =[]
    matriz.pop(0)#excluir a posicao que contem []
    return matriz


def segundo_maior(matriz):
 maior = max(max(matriz))
 for i in range(len(matriz)):
     for j in range(len(matriz)):
         if matriz[i][j] == maior:
             matriz[i].remove(maior)
             print(matriz)




 return max(max(matriz))

matriz = constroi_matriz(2)
print(f"A matriz é {matriz}")
print(f"O segundo maior é: {segundo_maior(matriz)}")

I don’t know why I sometimes get error from "builtins.Indexerror: list index out of range"

Answer 1

TL;DR

Knowing a little the modules that Python provides and some language functionality you can do it in a simpler way:

from random import randint
from itertools import chain


def constroi_matriz(m, n, min_num=0, max_num=99):
    return [ 
        [randint(min_num, max_num) for _ in range(m)]
        for _ in range(n)
    ]


def segundo_maior(matriz):
    return sorted(chain.from_iterable(matriz))[-2]


matriz = constroi_matriz(5, 5)

print(matriz)
print(segundo_maior(matriz))

^{See rotating on Repl.it}

---

Explanation

In function constroi_matriz used list comprehensions to create the matrix, where:

matriz = [ 
    [randint(min_num, max_num) for _ in range(m)]
    for _ in range(n)
]

Would be equivalent to:

matriz = []
for i in range(m):
    matriz.append([])
    for j in range(n):
        matriz[i].append(j)

^{PS: In the documentation of list comprehensions you will see that they can repeat themselves N for to create as many matrix dimensions as you want.}

Already in office segundo_maior I follow a simpler rule:

1. Turn the matrix into a list using itertools.chain.from_iterable. Ex.:

   from itertools import chain
   
   matriz = [[10, 20, 30], [5, 50, 500]]
   lista = chain.from_iterable(matriz)
   print(lista)
   # [10, 20, 30, 5, 50, 500]
   

2. I sort this list using the function sorted. Ex.:

   lista = [10, 20, 30, 5, 50, 500]
   ordenada = sorted(lista)
   print(ordenada)
   # [5, 10, 20, 30, 50, 500]
   

   ^{OBS.: sorted returns a new list with the sorted data, keeping the original list. If you want to overwrite the original list you should use list.sort(). Related questions: ""Sort" method returning only "None", what is the reason?" and "Method Reverse returns None".}

3. I take the penultimate item on the orderly list, that is, the second largest. Ex.:

   ordenada = [5, 10, 20, 30, 50, 500]
   print(ordenada[-2])
   # 50
   

Answer 2

your problem this on this line, it may change the length of the list but as this inside a codition is not always it will make this change that causes the problem.

matriz[i].remove(maior)  

I’ve made some changes, as a suggestion

n=2
def constroi_matriz(n):
    import random
    matriz =[] 
    for i in range(n):
        linha =[] 
        for j in range(n):
            num = random.randint(0,99)
            linha.append(num)
        matriz.append(linha)
    return matriz


def segundo_maior(matriz):
    maior = []
    for i in matriz:
        maior.extend(i)
    maior.sort()
    return maior[-2]

matriz = constroi_matriz(n)
print('A matriz é', *matriz, sep='\n')
print(f"O segundo maior é: {segundo_maior(matriz)}") 

in that second code would be a more lazy approach

from random import randint
from itertools import chain

n=2

def constroi_matriz(n):
    matriz =[[randint(0, 99) for x in range(n)]
            for y in range(n)]
    return matriz


def segundo_maior(matriz):
    lista = [x for x in chain(*matriz)]
    lista = sorted(set(lista))

    if len(lista) > 1:
        return lista[-2]
    return lista[0]

matriz = constroi_matriz(n)
print('A matriz é', *matriz, sep='\n')
print(f"O segundo maior é: {segundo_maior(matriz)}")  insira o código aqui

in this case I do some conversions, of list for set which removes all equality, and back to list to organize and take the position