将数据结构：CSV 转换为字典

Question

I have a text file like this:

Tomm Sietsema,Rooster and Owl,2.5,El Sapo Cuban Social Club,3.5,St. Anselm,3.0,Shibumi,3.5,Mama Chang,2.5,Punjab Grill,3.0    

Jonathan Golder,Rooster and Owl,3.0,El Sapo Cuban Social Club,3.5,St. Anselm,1.5,Shibumi,5.0,Punjab Grill,3.0,Mama Chang,3.5     

Brette Anderson,Rooster and Owl,2.5,El Sapo Cuban Social Club,3.0,Shibumi,3.5,Punjab Grill,4.0     

Michael Baumer,El Sapo Cuban Social Club,3.5,St. Anselm,3.0,Punjab Grill,4.5,Shibumi,4.0,Mama Chang,2.5    

Corby Kumar,Rooster and Owl,3.0,El Sapo Cuban Social Club,4.0,St. Anselm,2.0,Shibumi,3.0,Punjab Grill,3.0,Mama Chang,2.0    

Pete Wellsworth,Rooster and Owl,3.0,El Sapo Cuban Social Club,4.0,Punjab Grill,3.0,Shibumi,5.0,Mama Chang,3.5    

Jay Samuel,El Sapo Cuban Social Club,4.5,Mama Chang,1.0,Shibumi,4.0

I want output like this:

{    'Tomm Sietsema': {'Rooster and Owl': 2.5, 'El Sapo Cuban Social Club': 3.5,
'The Godfather': 3.0, 'Shibumi': 3.5, 'Mama Chang': 2.5, 'Punjab Grill':
3.0},
    'Jonathan Golder': {'Rooster and Owl': 3.0, 'El Sapo Cuban Social Club':
3.5, 'The Godfather': 1.5, 'Shibumi': 5.0, 'Punjab Grill': 3.0, 'Mama Chang':
3.5}}     

I am new to Python. I tried using list and appending the lists. I don't know how to proceed after this.

lists=list()    
with open(file_name) as b:       
for line in b:         
    lists.append(line.split(":")[0])   
    lists.append(line.split(":")[1])

Answer 1

Quite a bit shorter, and probably faster, than other solutions. Doing the loop as a dict comprehension should save some time.

with open(file_name) as fhin:
    for line in fhin:
        line = line.strip()
        if not line:
            continue

        entry = line.split(',')
        d[entry[0]] = {entry[i]: float(entry[i+1]) for i in range(1, len(entry)-1, 2)}

Answer 2

You may split your line using , then read the name at first, then read the elements by pair, and put all of this in nested dicts

# import json
result = {}  # outer dict

for line in lines:
    content = line.split(",")   # ['Tomm Sietsema', 'Rooster and Owl', '2.5', 'El Sapo Cuban Social Club', '3.5', 'St. Anselm', '3.0', 'Shibumi', '3.5', 'Mama Chang', '2.5', 'Punjab Grill', '3.0']
    name = content[0]           # 'Tomm Sietsema'
    name_values = {}            # inner dict

    for i in range(1, len(content), 2):
        val_name = content[i]              # read a element name
        val_float = float(content[i + 1])  # read the float value next
        name_values[val_name] = val_float  # save to inner dict

    result[name] = name_values             # save to outer dict

print(json.dumps(result, indent=4))

For the begnning it gives

{
    "Tomm Sietsema": {
        "Rooster and Owl": 2.5,
        "El Sapo Cuban Social Club": 3.5,
        "St. Anselm": 3.0,
        "Shibumi": 3.5,
        "Mama Chang": 2.5,
        "Punjab Grill": 3.0
    },
    "Jonathan Golder": {
        " Rooster and Owl": 3.0,
        " El Sapo Cuban Social Club": 3.5,
        " St.Anselm": 1.5,
        " Shibumi": 5.0,
        " Punjab Grill": 3.0,
        " Mama Chang": 3.5
    }
}

Answer 3

You can do something like this.

my_dict = {} 

with open("my_file.txt", "r") f:
    for line in f:
        key, *values = line.split(",") # key is user name, values are the rest
        tmp = {} # nested dict by user 
        val = iter(values) # smart way to create an iterator from the list
        for k in val:
             tmp[k] = float(next(val)) # add the key to the right value in the nested dict
        my_dict[key] = tmp # add the nested dict to the user

That produce the results you want.

{ "Tomm Sietsema": { "Rooster and Owl": 2.5, "El Sapo Cuban Social Club": 3.5, "St. Anselm": 3.0, "Shibumi": 3.5, "Mama Chang": 2.5, "Punjab Grill": 3.0 }, "Jonathan Golder": { " Rooster and Owl": 3.0, " El Sapo Cuban Social Club": 3.5, " St.Anselm": 1.5, " Shibumi": 5.0, " Punjab Grill": 3.0, " Mama Chang": 3.5 } }

Answer 4

So your syntax for each line is

name, key1, value1, key2, value2, ...

And you want to get

{name: {key1: value1, key2: value2, ...}}

This will give aa corresponding dict:

result = {}
for line in s.split("\n"):
    elements = line.split(",")
    name = elements[0].strip()    
    result[name] = {}

    # now iterate over key, value pairs across the rest of the elements
    for key, value in zip(elements[1::2], elements[2::2]):
        result[name][key.strip()] = float(value)

{'Tomm Sietsema': 
    {'Rooster and Owl': '2.5', 'El Sapo Cuban Social Club': '3.5', 'St. Anselm': '3.0', 'Shibumi': '3.5', 'Mama Chang': '2.5', 'Punjab Grill': '3.0'},
'Jonathan Golder':
    {'Rooster and Owl': '3.0', 'El Sapo Cuban Social Club': '3.5', 'St. Anselm': '1.5', 'Shibumi': '5.0', 'Punjab Grill': '3.0', 'Mama Chang': '3.5'},
'Brette Anderson':
    {'Rooster and Owl': '2.5', 'El Sapo Cuban Social Club': '3.0', 'Shibumi': '3.5', 'Punjab Grill': '4.0'},
'Michael Baumer':
    {'El Sapo Cuban Social Club': '3.5', 'St. Anselm': '3.0', 'Punjab Grill': '4.5', 'Shibumi': '4.0', 'Mama Chang': '2.5'},
'Corby Kumar':
    {'Rooster and Owl': '3.0', 'El Sapo Cuban Social Club': '4.0', 'St. Anselm': '2.0', 'Shibumi': '3.0', 'Punjab Grill': '3.0', 'Mama Chang': '2.0'},
'Pete Wellsworth':
    {'Rooster and Owl': '3.0', 'El Sapo Cuban Social Club': '4.0', 'Punjab Grill': '3.0', 'Shibumi': '5.0', 'Mama Chang': '3.5'},
'Jay Samuel':
    {'El Sapo Cuban Social Club': '4.5', 'Mama Chang': '1.0', 'Shibumi': '4.0'}
}

Answer 5

With the CSV import it is quite simple

import csv

dict = {}

with open('input.csv') as csv_file:
    csv_reader = csv.reader(csv_file, delimiter=',')
    line_count = 0
    for row in csv_reader:
        try:
            dict[row[0]] = {}
            for i in range(0,len(row),2):
                dict[row[0]].update({row[i+1]:row[i+2]})
        except:
            pass
        line_count += 1