Get first word of all strings in lists

Question

I have a CSV file which I'm reading in like below. I need to get the first word of all the strings. I know how to get first letter but I'm not sure how I can get words.

['diffuse systemic sclerosis', 'back', 'public on july 15 2008']
['diffuse systemic sclerosis', 'forearm', 'public on may 9 2014']

I want my output to be

diffuse
back
public
forearm

Answer 1

You can use a list comprehension , and split() function :

>>> l=['diffuse systemic sclerosis', 'back', 'public on july 15 2008']
>>> [i.split()[0] for i in l]
['diffuse', 'back', 'public']

Answer 2

You can use comprehension

>>> l = [['diffuse systemic sclerosis', 'back', 'public on july 15 2008']
,['diffuse systemic sclerosis', 'forearm', 'public on may 9 2014']]

>>> list({i.split()[0] for j in l for i in j})
['back', 'diffuse', 'forearm', 'public']

Answer 3

l = [
    ['diffuse systemic sclerosis', 'back', 'public on july 15 2008'],
    ['diffuse systemic sclerosis', 'forearm', 'public on may 9 2014']
    ]
d = lambda o: [a.split().pop(0) for a in o]
r = lambda a,b: d(a) + d(b)
print "\n".join(set(reduce(r, l)))
>>> 
public
forearm
diffuse
back

Answer 4

You can use str.split in a list comprehension, noting that you can specify maxsplit to reduce the number of operations:

L = ['diffuse systemic sclerosis', 'back', 'public on july 15 2008']

res = [i.split(maxsplit=1)[0] for i in L]
# ['diffuse', 'back', 'public']

You can also perform the same operations functionally:

from operator import itemgetter, methodcaller

splitter = methodcaller('split', maxsplit=1)
res = list(map(itemgetter(0), map(splitter, L)))

Across multiple lists, if you wish to maintain the order with which you observe unique first words, you can use the itertool unique_everseen recipe, also found in the more_itertools library:

from itertools import chain
from more_itertool import unique_everseen

L1 = ['diffuse systemic sclerosis', 'back', 'public on july 15 2008']
L2 = ['diffuse systemic sclerosis', 'forearm', 'public on may 9 2014']

res = list(unique_everseen(i.split(maxsplit=1)[0] for i in chain(L1, L2)))

# ['diffuse', 'back', 'public', 'forearm']