CODEWITHSUNDEEP
pythonarraysfunction
Vasia SowSow
Vasia SowSow

Reputation: 31

Python remove from list the same values

Please help. I have in input script the differing values and what need only remove the same values.

I do this but there is didn't check the type of value and only just didn't find it.

Example :

import sys
def clean_list(list_to_clean):
        c = []
        c = list(set(list_to_clean))
        return c
c= clean_list(sys.argv[1:])
print c

or 

import sys
def clean_list(list_to_clean):
        c = []
        for i in list_to_clean:
                if i not in c:
                        c.append(i)
        return c
c= clean_list(sys.argv[1:])
print c

so my problem, when i in input send values :

'asd', 'dsa', 1, '1', 1.0

it was back :

['asd', 'dsa', 1, '1']

but need :

['asd', 'dsa', 1, '1', 1.0]

Thanks

Upvotes: 2

Views: 166

Answers (2)

Andrew Carter
Andrew Carter

Reputation: 371

Since int 1 and float 1.0 hash to the same value, you need to retain some more information about the object -- it's 'type' is the best candidate here. You can do this in a one liner if you want: 

>>> seq = ['asd', 'dsa', 1, '1', 1.0, 'asd', 'dsa', 1]
>>> [pair[1] for pair in {(type(obj), obj,) for obj in seq}]
['asd', 'dsa', 1, 1.0, '1']

Breaking it down a bit more, the set comprehension: 

pairs = {(type(obj), obj,) for obj in seq}

Retains the type of obj in a tuple so it's not lost. Then the list comprehension just gets back the value (not the type): 

final_answer = [pair[1] for pair in pairs]

As an aside, you can use the builtin hash function to see what value python will use for dictionary/set hashing:

>>> hash(1.0)
1
>>> hash(1)
1

This is spelled out in the docs -- in fact, they mention the very case of 1 and 1.0.

Upvotes: 4

Thsise Faek
Thsise Faek

Reputation: 311

Python automatically shortens the integer and, when the cleaning code sees it, it thinks it is a duplicate. Try storing it as a string '1.0' and then converting it back to an integer for usage using the int function.

Upvotes: 0

Related Questions