How to count occurrence of strings within a dictionary made up of lists in Python?

Question

I have a satisfaction research dictionary that the evaluator can evaluate up to 6 products. The program stores the evaluator, the refrigerant and the note for the refrigerant.

And I have to show the names of the appraisers and the quantity of products assessed individually.

pesquisa = {'PEDRO': ['FANTA',5,'PEPSI',4],'ANA': ['SPRITE',5,'COCACOLA,4]}

def relatorio_avaliadores():
    print('Relatório de Avaliadores')

    for nome in sorted(pesquisa.keys()):
        
    print(f'{nome} avaliou o total de, , produto(s).')

How can I count the number of soft drinks that has been evaluated by each person?

Answer 1

If this is always the format of the counts you can also:

pesquisa = {'PEDRO': ['FANTA',5,'PEPSI',4],'ANA': ['SPRITE',5,'COCACOLA',4], 'DIEGO': ['INTEL', 7, 'PYTHON', 8, 'WINDOWS', 9, 'LINUX', 10]}
c = {i: sum((1 for j in pesquisa[i][::2])) for i in pesquisa}
print(c) # {'PEDRO': 2, 'DIEGO': 4, 'ANA': 2}

DEMONSTRATION

Answer 2

Specifically speaking of your answer, len already returns an integer, then do int(len(etc...)) is redundant and unnecessary. Do only count = len(pesquisa[nome]) // 2 would be enough (using the entire split operator // so that the result is not a float).

But actually dividing by 2 seems to me a little "gambiarra", to circumvent the fact that you are using a list, which does not seem to be the most appropriate structure for this case. If the idea is to have a note for each soda, I think it’s best to use a dictionary that maps the name of each soda with its respective note. It may seem a silly detail ("ah, but with lists it works"), but choosing the right data structure is halfway towards better code.

Then I’d do it this way:

def relatorio_avaliadores(pesquisa):
    print('Relatório de Avaliadores')
    for nome in sorted(pesquisa):
        print(f'{nome} avaliou o total de {len(pesquisa[nome])} produto(s).')

# as notas ficam em outro dicionário
pesquisa = {
  'PEDRO': { 'FANTA': 5, 'PEPSI': 4 }, # cada refrigerante é mapeado para sua nota
  'ANA': { 'SPRITE': 5, 'COCACOLA': 4 }
}

relatorio_avaliadores(pesquisa)

Note that to sort the keys, I can simply do sorted(pesquisa). And if I just want to print the quantity and I won’t use this value for anything else, nor would I need the variable cont.

And now I don’t have to divide by two anymore, because according to the documentation, the len of a dictionary returns the number of keys of this, and how each key corresponds to a soda evaluated by the person, when using len(pesquisa[nome]) I’ll have the right value.

Another detail is that the way you did it, the function relatorio_avaliadores only worked for the dictionary pesquisa raised out of it. But now I’ve changed the function to receive the dictionary as a parameter, so it works for any other dictionary that has the same structure. Ex:

from operator import itemgetter

def relatorio_avaliadores(pesquisa):
    print('Relatório de Avaliadores')
    # apenas para mostrar outra forma de iterar pelo dicionário
    for nome, notas in sorted(pesquisa.items(), key=itemgetter(0)):
        print(f'{nome} avaliou o total de {len(notas)} produto(s).')

# dados de uma pesquisa
pesquisa1 = {
  'PEDRO': { 'FANTA': 5, 'PEPSI': 4 },
  'ANA': { 'SPRITE': 5, 'COCACOLA': 4 }
}
relatorio_avaliadores(pesquisa1)

# dados de outra pesquisa
pesquisa2 = {
  'Fulano': { 'FANTA': 5, 'PEPSI': 4 },
  'Ciclano': { 'SPRITE': 5, 'COCACOLA': 4, 'TUBAÍNA': 5 },
  'Beltrano': {},
  'Trajano': { 'GUARANÁ': 3, 'FANTA': 2, 'PEPSI': 1, 'COCACOLA': 4 }
}
relatorio_avaliadores(pesquisa2)

# posso inclusive passar o dicionário diretamente
relatorio_avaliadores({
  'Maria': { 'FANTA': 3 },
  'José': { 'COCACOLA': 5, 'SPRITE': 2 },
  'Fulano': { 'GUARANÁ': 4, 'PEPSI': 2, 'COCACOLA': 4 }
})

Note that I can pass any dictionary to the function, and it works the same way (so it gets more generic, rather than depending on an external variable to it).

I also showed another way to iterate through the dictionary, returning the name and notes at the same time, and sorting through the key: items() returns keys and values in tuples, and in the sort I say to use the first element of the tuple, which is the dictionary key (use itemgetter(0) to indicate that I will use the first element - remember that the indexes start at zero, so I used itemgetter(0)).

---

And as recalled in the comments, if you don’t have control over the structure, you can convert its initial structure to the one I suggested above:

pesquisa = {'PEDRO': ['FANTA',5,'PEPSI',4],'ANA': ['SPRITE',5,'COCACOLA',4]}
pesquisa = { nome: dict(zip(notas[::2], notas[1::2])) for nome, notas in pesquisa.items() }

That is, for each reviewer, I take the list of notes and turn it into the dictionary. The trick is to use the Slices [::2] (takes the even indexes of the list) and [1::2] (picks up the odd indices). Read here to better understand how this syntax works.

Answer 3

I managed to solve it! I’m a beginner and still use basic things to solve my problems, if you know in a more pytonic way I accept suggestions!

def relatorio_avaliadores():
    print('Relatório de Avaliadores')

    for nome in sorted(pesquisa.keys()):
        count = int(len(pesquisa[nome])/2)
        
        print(f'{nome} avaliou o total de, {count}, produto(s).')

Answer 4

In case you don’t want to split it two ways, like you did here:

pesquisa = {'PEDRO': ['FANTA',5,'PEPSI',4],'ANA': ['SPRITE',5,'COCACOLA',4]}

def relatorio_avaliadores():
    print('Relatório de Avaliadores')

    for nome in sorted(pesquisa.keys()):
        count = int(len(pesquisa[nome])/2)
        
        print(f'{nome} avaliou o total de, {count}, produto(s).')

You can iterate using list comprehension and see which ones are str with isinstance:

pesquisa = {'PEDRO': ['FANTA',5,'PEPSI',4],'ANA': ['SPRITE',5,'COCACOLA',4]}

def relatorio_avaliadores():
    print('Relatório de Avaliadores')

    for nome, avaliacoes in sorted(pesquisa.items()):
        count = len([i for i in avaliacoes if isinstance(i, str)])
        print(f'{nome} avaliou o total de, {count} produto(s).')

You can do it with the filter too:

count = len(list(filter(lambda x : isinstance(x, str), avaliacoes)))

This way it will only count when you find something in the format 'string'.