Why does this return False in Python 3? And what is a way to make it return True?
e = "allé.png"
l = ["allé.png"]
print(e in l)
Asked
Active
Viewed 124 times
2
-
1It returns `True` for me – Cory Kramer Oct 25 '18 at 16:52
-
1it returns `True` for me – juanpa.arrivillaga Oct 25 '18 at 16:52
-
1It returns 'True' for me. – Mayank Porwal Oct 25 '18 at 16:52
-
2This is very strange! – Lucien S. Oct 25 '18 at 16:53
-
Would anyone know what could be the cause of this? – Lucien S. Oct 25 '18 at 16:57
-
Ran this is an online repl (3.7) and local machine (3.6) and it returned true. ...Dead sectors on an internal device? Really weird. – Quentin Oct 25 '18 at 17:01
-
2Really weird. can be related to encoding though; `print(sys.getdefaultencoding())` What output do you get – mad_ Oct 25 '18 at 17:07
-
How could I check your suggestions? – Lucien S. Oct 25 '18 at 17:08
-
@mad_ I'm getting utf-8 – Lucien S. Oct 25 '18 at 17:18
-
1I have ascii. Try your luck changing that.https://stackoverflow.com/questions/2276200/changing-default-encoding-of-python – mad_ Oct 25 '18 at 17:19
-
1What happens if you do: `import unicodedata e = "allé.png" e = unicodedata.normalize('NFC', e) l = [unicodedata.normalize('NFC', "allé.png")] print(e in l)` – Dani Mesejo Oct 25 '18 at 17:23
-
@DanielMesejo this works!!! why though? :) – Lucien S. Oct 25 '18 at 17:25
-
When comparing unicode strings you should always normalize, unicode has different representations for the same gliph which can cause this type of problems – Dani Mesejo Oct 25 '18 at 17:27
-
Thanks! so if I were to look in a large list of string to check if e in present, how could I normalize every entry? – Lucien S. Oct 25 '18 at 17:28
-
try with this # - * - coding: latin-1 - * - in your code's heading. – nsaura Oct 25 '18 at 17:30
-
@nsaura thanks, but no luck here :( – Lucien S. Oct 25 '18 at 17:32
-
1@Rodolphe Added an answer! – Dani Mesejo Oct 25 '18 at 17:35
1 Answers
2
When comparing unicode you should normalize your data using unicodedata. If you intend to search in a large list you could use map or list comprehension:
import unicodedata
from functools import partial
normalize = partial(unicodedata.normalize, 'NFC')
e = "allé.png"
e = normalize(e)
l = ["allé.png"]
print(e in map(normalize, l))
Output
True
Or as an alternative:
print(e in [normalize(s) for s in l])
Further
Dani Mesejo
- 61,499
- 6
- 49
- 76