CODEWITHSUNDEEP
python
theuserid01
theuserid01

Reputation: 59

Python removing specific duplicates from a list

I want to remove specific duplicates from a list. With Perl I would do the task with this code:

my @list = ( 'a1', 'a1', 'b1', 'b1' );

my %seen;
@list = grep( !/a\d/ || !$seen{ $_ }++, @list );

and the wanted result would be this:

@list = ( 'a1', 'b1', 'b1' );

How could I do this in Python 3 using regular expression and list comprehension. Thanks.

Upvotes: 3

Views: 161

Answers (3)

Yuri Kovalev
Yuri Kovalev

Reputation: 659

import re
from functools import reduce  # this import is not needed in python 2.*

l = ['a1', 'a1', 'b1', 'b1']

print reduce(lambda acc, el: acc if re.match(r'a\d', el) and el in acc else acc + [el], l, [])

Sorry, this is solution without list comprehensions. Is it strictly required?

Upvotes: 1

user554546
user554546

Reputation:

Here's another solution, using list(set(stuff)) to generate a list of unique things from stuff (since sets automatically deduplicate things):

In [1]: import re

In [2]: l = ["a1", "a1", "b1", "b1"]

In [3]: items_to_dedupe = [x for x in l if re.match(r"a\d", x)]

In [4]: leave_alone = [x for x in l if x not in items_to_dedupe]

In [5]: list(set(items_to_dedupe)) + leave_alone
Out[5]: ['a1', 'b1', 'b1']

Upvotes: 0

Kasravnd
Kasravnd

Reputation: 107357

You can use itertools.chain and groupby :

>>> list(chain(*[[i[0]] if 'a1' in i else i for i in [list(g) for _,g in groupby(sorted(l))]]))
['a1', 'b1', 'b1']

and if you just want to use regex you can concatenate the elements the n use re.sub , but note that it works for this special case ! that , is the delimiter ! :

>>> l =['a1', 'a1', 'b1', 'b1']
>>> re.sub(r'(a1,)+','a1,',','.join(sorted(l))).split(',')
['a1', 'b1', 'b1']

Upvotes: 1

Related Questions