将CSV数据转换为字典中的列表

Question

I have a CSV file in the following form:

Name_1,2,K,14
Name_1,3,T,14
Name_1,4,T,18
Name_2,2,G,12
Name_2,4,T,14
Name_2,6,K,15
Name_3,2,K,12
Name_3,3,T,15
Name_3,4,G,18

And I want to convert it into a dictionary where Name_x is the key and corresponding data is the value in list form. Something like this:

{'Name_1': [[2, 'K', 14], [3, 'T', 14], [4, 'T', 18]],
 'Name_2': [[4, 'T', 14], [4, 'T', 14], [6, 'K' ,15]],
...}

So far, I think I have to use use defaultdict :

from collections import defaultdict
d = defaultdict(list)

But how do I append the data to d ? I know defaultdict does not have an append method.

Answer 1

You need to use the name as the key and append the slice of the row as the value, there will be no order using a normal or defaultdict:

import csv
from collections import defaultdict

with open('in.csv') as f:
    r = csv.reader(f)
    d = defaultdict(list)
    for row in r:
        d[row[0]].append(row[1:])
print(d)

If you want to maintain order you will need an OrderedDict :

from collections import OrderedDict

with open('in.csv') as f:
    r = csv.reader(f)
    od = OrderedDict()
    for row in r:
        # get key/ first element in row
        key = row[0]
        # create key/list paring if it does not exist, else just append the value
        od.setdefault(key, []).append(row[1:])
print(od)

Output:

OrderedDict([('Name_1', [['2', 'K', '14'], ['3', 'T', '14'], ['4', 'T', '18']]), ('Name_2', [['2', 'G', '12'], ['4', 'T', '14'], ['6', 'K', '15']]), ('Name_3', [['2', 'K', '12'], ['3', 'T', '15'], ['4', 'G', '18']])])

You could also use groupby if the names are grouped which will group elements based on the first item/name in each row:

import csv
from collections import OrderedDict
from itertools import groupby
from operator import itemgetter

with open('in.csv') as f:
    r = csv.reader(f)
    od = OrderedDict()
    for k, v in groupby(r, key=itemgetter(0)):
        od[k] = [sub[1:] for sub in v]

If you are using python3 you can unpack using * :

with open("in.csv") as f:
    r = csv.reader(f)
    od = OrderedDict()
    for row in r:
        key, *rest = row
        od.setdefault(key, []).append(rest)


import csv
from collections import OrderedDict
from itertools import groupby
from operator import itemgetter

with open('in.csv') as f:
    r = csv.reader(f)
    od = OrderedDict()
    for k, v in groupby(r, key=itemgetter(0)):
        od[k] = [sub for _, *sub in v]
print(od)

Answer 2

txtcsv="""Name_1,2,K,14
Name_1,3,T,14
Name_1,4,T,18
Name_2,2,G,12
Name_2,4,T,14
Name_2,6,K,15
Name_3,2,K,12
Name_3,3,T,15
Name_3,4,G,18"""

def save():
    with open("test.csv","w") as f:
        f.write(txtcsv)


if __name__ == "__main__":
    save()
    with open("test.csv") as f:
        d = {}
        for l in f.readlines():
            name, val = l.rstrip().split(",", 1)
            d.setdefault(name, []).append(val.split(","))
        print (d)

Answer 3

Off the top of my head (because I'm not too familiar with defaultdict), this should do roughly what you want.

data is the CSV string

obj = {}

data = data.split('\n')
for row in data:
    row = row.split(',')
    if row[0] in obj:
        obj[row[0]].append(row[1:])
    else:
        obj[row[0]] = [row[1:]]

print obj