I have a mixed list with strings and ints and I need to get to the sum of the numbers between each strings. Ideally the end result would be a list of tuples since each string and following numbers belong together (so order is important).

I can extract the number with iteration using isinstance but the actual list is very large and I sometimes I have 1 or 2 numbers for each string.

my_list = ['a', 2, 1, 'b', 3, 'h', 50, 4, 'd', 4, 'v', 20, 7]

ideal_output = [('a', 3) ('b', 3), ('h', 54), ('d', 4), (v, 27)]

3 Answers

2
DroidX86 On

Here's a solution using itertools.groupby:

my_list = ['a', 2, 1, 'b', 3, 'h', 50, 4, 'd', 4, 'v', 20, 7]

from itertools import groupby

groups = groupby(my_list, key=type) # group-by the type of the value

result = []
for key, group in groups:
    string = next(group) # get the string first, we'll skip over it otherwise
    if key is str:
        _, values = next(groups) # move the generator forward to get to the numbers
        result.append((string, sum(values))) # sum up the numbers
print(result)

Output:

[('a', 3), ('b', 3), ('h', 54), ('d', 4), ('v', 27)]

It does assume that there will be at least one number between the strings. If not you can check on the len of g and if that's more than 1, add a 0 for the value of the first value in g

0
Error - Syntactical Remorse On

You can just use simple iteration as well without groupby, this will be slightly faster because it only does one pass:

my_list = ['a', 2, 1, 'b', 3, 'h', 50, 4, 'd', 4, 'v', 20, 7]
new_list = []
new_element = []
for element in my_list:
    if isinstance(element, str):
        if new_element:
            new_list.append(tuple(new_element))
        new_element = [element, 0]
    else:
        new_element[1] += element
if new_element:
    new_list.append(new_element)
print(new_list)

Output:

[('a', 3), ('b', 3), ('h', 54), ('d', 4), ('v', 27)]
0
RomanPerekhrest On

With functools.reduce function:

from functools import reduce

def accum(prev, curr):    # prev - previous item, curr - current item
    if prev == [] or isinstance(curr, str):
        prev.append((curr, 0))
    elif isinstance(curr, int):
        prev[-1] = (prev[-1][0], prev[-1][1] + curr)
    return prev

my_list = ['a', 2, 1, 'b', 3, 'h', 50, 4, 'd', 4, 'v', 20, 7]
res = reduce(accum, my_list, [])   # [] is initializer
print(res)

The output:

[('a', 3), ('b', 3), ('h', 54), ('d', 4), ('v', 27)]