Thursday, September 4, 2025
HomeLanguagesPython program to sort strings by substring range

Python program to sort strings by substring range

Given a String List, sort by a range of substrings.

Input : test_list = [“neveropen”, “best”, “neveropen”, “preparation”, “interview”], i, j = 1, 3 
Output : [‘neveropen’, ‘neveropen’, ‘best’, ‘interview’, ‘preparation’]

Explanation : “eek” < “eek” < “est” < “nte” < “rep”.

Input : test_list = [“apple”, “orange”, “banana”], i, j = 2, 4 
Output : [‘orange’, ‘banana’, ‘apple’] 

Explanation : “ang” < “nan” < “ple”. 

Method #1 : Using sort() + slicing

In this, we perform the task of sorting using sort() and the task of extracting a range is done using slicing.

Python3




# Python3 code to demonstrate working of
# Sort Strings By Substring Range
# Using sort() + slicing
 
# helper function
def get_substr(test_str):
     
    # getting range
    return test_str[i : j]
 
# initializing list
test_list = ["neveropen", "best", "neveropen", "preparation"]
 
# printing original list
print("The original list is : " + str(test_list))
 
# initializing range
i, j = 1, 3
 
# calling func.
test_list.sort(key=get_substr)
 
# printing result
print("Strings after sorting : " + str(test_list))


Output

The original list is : ['neveropen', 'best', 'neveropen', 'preparation']
Strings after sorting : ['neveropen', 'neveropen', 'best', 'preparation']

Time Complexity: O(nlogn)

Auxiliary Space: O(n)

Method #2 : Using lambda function + sort() + slicing

In this, we perform slicing using lambda function rather than calling external function.

Python3




# Python3 code to demonstrate working of
# Sort Strings By Substring Range
# Using lambda function + sort() + slicing
 
# initializing list
test_list = ["neveropen", "best", "neveropen", "preparation"]
 
# printing original list
print("The original list is : " + str(test_list))
 
# initializing range
i, j = 1, 3
 
# lambda function providing sort fnc.
test_list.sort(key=lambda test_str : test_str[i : j])
 
# printing result
print("Strings after sorting : " + str(test_list))


Output

The original list is : ['neveropen', 'best', 'neveropen', 'preparation']
Strings after sorting : ['neveropen', 'neveropen', 'best', 'preparation']

Time Complexity: O(nlogn)

Auxiliary Space: O(n)

Method #3:  Using the functools.cmp_to_key() function and sorted()

Step-by-step algorithm:

  1. Import the cmp_to_key function from the functools module.
  2. Create a list of strings test_list.
  3. Initialize i and j to 1 and 3 respectively.
  4. Define a comparison function cmp that takes two strings a and b as input and compares the substring of a and b from index i to index j. If the substring of a is less than the substring of b, return -1. If the substring of a is greater than the substring of b, return 1. Otherwise, return 0.
  5. Sort the list test_list using the sorted() function and the key argument set to cmp_to_key(cmp).
  6. Print the sorted list.

Python3




from functools import cmp_to_key
# initializing list
test_list = ["neveropen", "best", "neveropen", "preparation"]
i, j = 1, 3
# printing original list
print("The original list is : " + str(test_list))
 
def cmp(a, b):
    if a[i:j] < b[i:j]:
        return -1
    elif a[i:j] > b[i:j]:
        return 1
    else:
        return 0
 
#initializing sorted
sorted_list = sorted(test_list, key=cmp_to_key(cmp))
 
# printing result
print(sorted_list)


Output

The original list is : ['neveropen', 'best', 'neveropen', 'preparation']
['neveropen', 'neveropen', 'best', 'preparation']

Time complexity: O(n log n), where n is the length of the list test_list. This is because sorting using the sorted() function takes O(n log n) time.

Auxiliary Space: O(n), where n is the length of the list test_list. This is because we are creating a new list to store the sorted version of the input list.

Method #4:Using itemgetter() from operator module

  • Importing the operator module to use itemgetter() function for sorting by substring range.
  • Initializing a list test_list containing strings.
  • Printing the original list.
  • Initializing the range (start, end) for the substring.
  • Sorting the list using the sorted() function and itemgetter() function from the operator module.
  • The key argument takes operator.itemgetter(i, j) as a parameter. It sorts the list based on the substring from index i to index j.
  • Printing the sorted list.

Python3




# importing operator module
import operator
 
# initializing list
test_list = ["neveropen", "best", "neveropen", "preparation"]
 
# printing original list
print("The original list is : " + str(test_list))
 
# initializing range
i, j = 1, 3
 
# sorting by substring range
test_list = sorted(test_list, key=operator.itemgetter(i, j))
 
# printing result
print("Strings after sorting : " + str(test_list))


Output

The original list is : ['neveropen', 'best', 'neveropen', 'preparation']
Strings after sorting : ['neveropen', 'neveropen', 'best', 'preparation']

Time Complexity: O(n log n), where n is the length of the list test_list.

Auxiliary Space: O(n), where n is the length of the list test_list. 

Method #6: Using list comprehension and sorted()

steps : 

We initialize a list test_list with some strings.
We print the original list test_list.
We initialize two variables i and j to represent the start and end indices of the substring range we want to sort on.
We sort the list test_list using the sorted() function and a lambda function as the key. The lambda function extracts the substring range of each string in the list using the variables i and j. The sorted() function returns a new sorted list.
We update the original list test_list with the sorted list.
We print the updated list test_list.

Python3




# initializing list
test_list = ["neveropen", "best", "neveropen", "preparation"]
 
# printing original list
print("The original list is : " + str(test_list))
 
# initializing range
i, j = 1, 3
 
# sorting by substring range
test_list = sorted(test_list, key=lambda x: x[i:j+1])
 
# printing result
print("Strings after sorting : " + str(test_list))


Output

The original list is : ['neveropen', 'best', 'neveropen', 'preparation']
Strings after sorting : ['neveropen', 'neveropen', 'best', 'preparation']

Time complexity: O(n log n), where n is the length of the list. 

Auxiliary space: O(n), where n is the length of the list. 

Dominic
Dominichttp://wardslaus.com
infosec,malicious & dos attacks generator, boot rom exploit philanthropist , wild hacker , game developer,
RELATED ARTICLES

Most Popular

Dominic
32264 POSTS0 COMMENTS
Milvus
81 POSTS0 COMMENTS
Nango Kala
6632 POSTS0 COMMENTS
Nicole Veronica
11800 POSTS0 COMMENTS
Nokonwaba Nkukhwana
11859 POSTS0 COMMENTS
Shaida Kate Naidoo
6749 POSTS0 COMMENTS
Ted Musemwa
7025 POSTS0 COMMENTS
Thapelo Manthata
6698 POSTS0 COMMENTS
Umr Jansen
6718 POSTS0 COMMENTS