CODEWITHSUNDEEP
pythonlistsuminteger
JachymDvorak
JachymDvorak

Reputation: 149

Sum all numbers in a list of strings

sorry if this is very noob question, but I have tried to solve this on my own for some time, gave it a few searches (used the "map" function, etc.) and I did not find a solution to this. Maybe it's a small mistake somewhere, but I am new to python and seem to have some sort of tunnel vision.

I have some text (see sample) that has numbers inbetween. I want to extract all numbers with regular expressions into a list and then sum them. I seem to be able to do the extraction, but struggle to convert them to integers and then sum them.

import re
df = ["test 4497 test 6702 test 8454 test",
 "7449 test"]

numlist = list()
for line in df:
    line = line.rstrip()
    numbers = re.findall("[0-9]+", line) # find numbers
    if len(numbers) < 1: continue # ignore lines with no numbers, none in this sample
    numlist.append(numbers) # create list of numbers

The sum(numlist) returns an error.

Upvotes: 0

Views: 2777

Answers (5)

Booboo
Booboo

Reputation: 44013

This is the source of your problem:

finadall returns a list which you are appending to numlist, a list. So you end up with a list of lists. You should instead do:

numlist.extend(numbers)

So that you end up with a single list of numbers (well, actually string representations of numbers). Then you can convert the strings to integers and sum:

the_sum = sum(int(n) for n in numlist)

Upvotes: 2

Wups
Wups

Reputation: 2569

Here is an option using map, filter and sum:
First splits the strings at the spaces, filters out the non-numbers, casts the number-strings to int and finally sums them.

# if you want the sum per string in the list
sums = [sum(map(int, filter(str.isnumeric, s.split()))) for s in df]
# [19653, 7449]

# if you simply want the sum of all numbers of all strings
sum(sum(map(int, filter(str.isnumeric, s.split()))) for s in df)
# 27102

Upvotes: 1

Maurice Meyer
Maurice Meyer

Reputation: 18106

Iterate twice over df and append each digit to numlist:

numlist = list()
for item in df:
  for word in item.split():
    if word.isnumeric():
      numlist.append(int(word))

print(numlist)
print(sum(numlist))

Out:

[4497, 6702, 8454, 7449]
27102

You could make a one-liner using list comprehension:

print(sum([int(word) for item in df for word in item.split() if word.isnumeric()]))
>>> 27102

Upvotes: 2

go2nirvana
go2nirvana

Reputation: 1638

It's as easy as

my_sum = sum(map(int, numbers_list))

Upvotes: 1

yatu
yatu

Reputation: 88226

You don't need a regex for this. Split the strings in the list, and sum those that are numeric in a comprehension:

sum(sum(int(i) for i in s.split() if i.isnumeric()) for s in df)
# 27102

Or similarly, flatten the resulting lists, and sum once:

from itertools imprt chain 
sum(chain.from_iterable((int(i) for i in s.split() if i.isnumeric()) for s in df))
# 27102

Upvotes: 3

Related Questions