总结元组列表中相同项目的值，而它们是字符串

Question

If I have list of tuples like this:

my_list = [('books', '$5'), ('books', '$10'), ('ink', '$20'), ('paper', '$15'), ('paper', '$20'), ('paper', '$15')] 

how can I turn the list to this:

[('books', '$15'), ('ink', '$20'), ('paper', '$50')]

ie to add the expense of same item while both the items are string in the tuples. I have problem with the price items being string. Any hint would be greatly appreciated. Thanks a lot!

I am getting the first list in this way:

my_list=[]
for line in data:
        item, price  = line.strip('\n').split(',') 
        cost = ["{:s}".format(item.strip()), "${:.2f}".format(float(price))]
        my_list.append(tuple(cost))

Now my_list should look like given above.

Answer 1

You can use defaultdict to do this:

>>> from collections import defaultdict
>>> my_list = [('books', '$5'), ('books', '$10'), ('ink', '$20'), ('paper', '$15'), ('paper', '$20'), ('paper', '$15')] 
>>> res = defaultdict(list)
>>> for item, price in my_list:
...     res[item].append(int(price.strip('$')))
... 
>>> total = [(k, "${}".format(sum(v))) for k, v in res.items()]
>>> total
[('ink', '$20'), ('books', '$15'), ('paper', '$50')]

Answer 2

This is fairly easy using a dictionary:

result={} #initialize an empty dictionary
for (type,cost) in my_list:
    if type not in result.keys():
        result[type]=int(cost[1:]) #add entry
    else:
        result[type]=result[type]+int(cost[1:]) #increment cost

#make dictionary a list again
dictlist=[]
for key, value in result.iteritems():
    temp = [key,"$"+str(value)] #add dollar sign
    dictlist.append(temp)

print dictlist

Edit: forgot two lines

Answer 3

You can simply use a dictionary to solve this problem.

my_list = [('books', '$5'), ('books', '$10'), ('ink', '$20'),
       ('paper', '$15'), ('paper', '$20'), ('paper', '$15')]

sums = {}
for item, price in my_list:
    sums[item] = sums.get(item, 0) + int(price[1:])

print sums

And if you need list of tuples, just do

print sums.items()

And to get your the output you expect,

print [(item, '$' + str(price)) for item, price in sums.items()]

Update

If you have costs with floating point value, you can simply change int to float like this and rest of the code remains the same,

sums[item] = sums.get(item, 0) + float(price[1:])

Answer 4

If your data is already grouped like you sample input, you can use itertools.groupby , grouping by each first element of the tuples and summing the prices in each group:

from itertools import groupby
from operator import itemgetter

my_list = [('books', '$5'), ('books', '$10'), ('ink', '$20'), ('paper', '$15'), ('paper', '$20'), ('paper', '$15')]

grouped = [(k, "${}".format(sum(int(s[1][1:]) for s in v)))    
               for k ,v in groupby(my_list, itemgetter(0))]

Output:

[('books', '$15'), ('ink', '$20'), ('paper', '$50')]

If it were not ordered you could call sorted groupby(sorted(my_list), itemgetter(0)) but changes the complexity to n log n so a dict option may be better in that case.

You can also do the groupby on data and forget building an intermediary list, csv.reader will also split the data for you:

from itertools import groupby
from operator import itemgetter

from csv import reader
grouped = [(k, "${}".format(sum(int(s[1]) for s in v))) 
              for k, v in groupby(reader(data), itemgetter(0))]

You might be better also leaving the values as ints and formatting when you want to output. Also to handle floats cast to float not int:

from csv import reader
grouped = [(k, sum(float(s[1]) for s in v))
              for k, v in groupby(reader(data), itemgetter(0))]

When you want to output you can add the dollar sign and also pad so you get nicely formatted output:

In [10]: "${:.2f}".format(1.0)
Out[10]: '$1.00'

Answer 5

Just for fun, a one-liner:

[(k,'$'+str(sum(int(e[1][1:]) for e in my_list if e[0]==k))) for k in set(map(lambda x:x[0], my_list))]

Don't actually do this.