I've been trying to sort a top-5 leaderboard which includes the score and the username of the account that got the score. Here is my current code:
g = open("scores.txt", "r")
score1 = g.readline().strip()
score2 = g.readline().strip()
score3 = g.readline().strip()
score4 = g.readline().strip()
score5 = g.readline().strip()
print(score1,score2,score3,score4,score5)
print(sorted([score1,score2,score3,score4,score5], reverse = True))
g.close()
It seems to work, but it only sorts the leftmost digits of the scores, for example it sorts 100coolguy 50otherguy 10coolguy 2000coolguy 2otherguy as 50otherguy 2otherguy 2000coolguy 10coolguy 100coolguy.
A list of the scores, and how they should be formatted afterwards:
100otherexample 50example 10otherexample 2000example 2example
2000example 100otherexample 50example 10otherexample 2example
CodePudding user response：
from natsort import natsorted
print(natsorted([score1,score2,score3,score4,score5], reverse = True))
CodePudding user response：
You can first find digits then sort base digits like below:
st = ['100coolguy' , '50otherguy' , '10coolguy' , '2000coolguy' ,'2otherguy']
def find_int(text):
return int(''.join(t for t in text if t.isdigit()))
st.sort(key=find_int, reverse=True)
print(st)
Output:
['2000coolguy', '100coolguy', '50otherguy', '10coolguy', '2otherguy']
CodePudding user response：
You need give a custom key to sorted key is what is evaluated instead of the whole list item in the sorting process:
A regex implementation:
import re
l = ["100coolguy", "50otherguy", "10coolguy", "2000coolguy", "2otherguy"]
def my_key(item):
item_split = re.split("[a-z]", item, 1, flags=re.I)
return int(item_split[0])
l_sorted = [sorted(l, key=my_key, reverse=True)]
print(l_sorted)
Splits at any character ("[a-z]") for 1 time and ignores case with re.I.