CODEWITHSUNDEEP
pythonlist
heyitsmev
heyitsmev

Reputation: 13

Splitting string with multiple delimiters

I am reading from a file that has the following in it.

87965164,Paris,Yu,6/27/1997
87965219,Heath,Moss,10/13/1996
87965187,Cale,Blankenship,10/22/1995
87965220,Terrence,Watkins,12/7/1996
87965172,Ansley,Padilla,3/30/1997

i need to split the the lines at the "," and the "/" and also removing the "\n" from the end.

i want my output to look like this when put into a list:

[['87965164', 'Paris', 'Yu', 6, 27, 1997], ['87965219', 'Heath', 'Moss', 10, 13, 1996], ['87965187', 'Cale', 'Blankenship', 10, 22, 1995], ['87965220', 'Terrence', 'Watkins', 12, 7, 1996], ['87965172', 'Ansley', 'Padilla', 3, 30, 1997]]

Upvotes: 0

Views: 210

Answers (4)

Elazar
Elazar

Reputation: 21585

Simpler than regex:

[line.replace('/', ',').split(',') for line in text.split('\n')]

You can transform numbers into ints afterwards.

However, I believe that you are looking for the wrong way to do it. The right way is to split by commas, then give special fields a dedicated treatment.

from datetime import datetime
from collections import namedtuple

Person = namedtuple('Row', ['idn', 'first', 'last', 'birth'])

def make_person(idn, first, last, birth):
    return Person(idn, first, last,
                  datetime.strptime(birth, "%m/%d/%Y"))

records = [make_person(*line.split(',')) for line in text.split('\n')]

Upvotes: 1

TigerhawkT3
TigerhawkT3

Reputation: 49320

Rather than storing heterogeneous data in a homogeneous data type, I'd recommend using dictionaries or creating a class.

With dictionaries:

results = {}
with open('in.txt') as f:
    for line in f:
        id, first, last, day = line.split(',')
        month, day, year = map(int, day.split('/'))
        results[id] = {'id':id, 'first':first, 'last':last,
                       'month':month, 'day':day, 'year':year}

With a class:

class Person:
    def __init__(self, id, first, last, day):
        self.id = id
        self.first = first
        self.last = last
        self.month, self.day, self.year = map(int, day.split('/'))

results = {}
with open('in.txt') as f:
    for line in f:
        id, first, last, day = line.split(',')
        results[id] = Person(id, first, last, day)

Note that in each case I am storing each person's info as an entry in a dictionary, with a key of what looks like their ID number.

Upvotes: 1

Alex Hall
Alex Hall

Reputation: 36013

For each line:

parts = line.split(',')
parts[-1:] = map(int, parts[-1].split('/'))

This will correctly handle input that has any slashes in the non-date parts, and easily handles the conversion to integers at the same time.

Upvotes: 0

cwallenpoole
cwallenpoole

Reputation: 81988

You're going to want regular expressions.

import re

results = []
for line in fl:
  # [,/] means "match if either a , or a / is present"
  results.append(re.split('[,/]',line.strip()))

If you have a particularly big file, you can wrap it in a generator:

import re
def splitter(fl):
   for line in fl:
     # By using a generator, you are only accessing one line of the file at a time.
     yield re.split('[,/]',line.strip())

Upvotes: 2

Related Questions