Python 2.7: Format a list within a list using for loops

Question

Input:

data = [['A', 'B', 'C'], ['001', 'ddd', 'eee', '10'], ['002', 'fff', 'ggg', '20']]

Expected output:

data = ['A', 'B', 'C'], [1, 'ddd', 'eee', 10], [2, 'fff', 'ggg', 20]]

• Convert columns with numerical values from strings to integers (with quotes around the numbers removed

I have attempted with the below code but I am getting this error:

ValueError: could not convert string to float: A

Could anyone point out my mistake?

formatted = []
for row in data:
    new_row = []
    for i, col in enumerate(row):
        if i != [1,2]:
            new_row.append(col)
            new_row.append(float(col))
    formatted.append(new_row)

print formatted

Answer 1

The 'pythonic' way to do this is to try converting each element to an integer, and fall back to keeping the string in case of failure.

formatted = []
for row in data:
    new_row = []
    for elem in row:
        try:
            new_row.append(int(elem))
        except ValueError:
            new_row.append(elem)
    formatted.append(new_row)

print formatted

Answer 2

What about this:

def to_float(s):
    try:
        return float(s)
    except:
        return s

[[to_float(s) for s in row] for row in data]

Answer 3

Your variable i , an integer, will never be equal to [1,2], a list. What you meant to say/write is:

if i not in [1,2]:

Edit: I forgot the first row. Since your first row is not like the others, either handle it differently (not recommended), or use one of the other answers (recommended)

Answer 4

You can apply a lambda function with a condition using map:

for i,d in enumerate(data):
    data[i] = map(lambda x: float(x) if x.isdigit() else x, d)

map applies the lambda function on every subelement of d. If the element is a string representing a digit, it applies conversion using float , if it is not, it leaves the whole string as it is. It directly replaces the sublist in data.

In Python 3.X the result of map would need to be explicitly converted to a list, ie data[i] = list(map()) .

Answer 5

You can try by using json and regex as follows:

import json, re

data = [['A', 'B', 'C'], ['001', 'ddd', 'eee', '10'], ['002', 'fff', 'ggg', '20']]
data = json.dumps(data)
data = json.loads(re.sub(r'"0*(\d+)"', r'\1', data)

print (data)

Output:

[['A', 'B', 'C'], [1, 'ddd', 'eee', 10], [2, 'fff', 'ggg', 20]]

You also mentioned you wanted to remove commas after each list, well here is one way:

data = reduce(lambda oldList, newList: oldList + newList, data, [])

Output:

['A', 'B', 'C', 1, 'ddd', 'eee', 10, 2, 'fff', 'ggg', 20]