Incorrect removal of accents when preceded by a capital letter

Question

ViliamVadocz opened this issue 3 years ago · 3 comments

>>> import cleantext
>>> cleantext.clean("všetko")
'vsetko'
>>> cleantext.clean("Všetko") 
'va!etko'

>>> cleantext.__version__
'0.4.0'

Answer 1 · 2021-08-12T13:50:39.000Z

unidecode does not have this issue, so something must be happening before.

>>> import unidecode
>>> unidecode.unidecode("všetko")
'vsetko'
>>> unidecode.unidecode("Všetko") 
'Vsetko'

Answer 2 · 2021-08-12T13:56:52.000Z

Issue seems to be with the fix_bad_unicode function.

>>> from cleantext import fix_bad_unicode
>>> fix_bad_unicode("všetko") 
'všetko'
>>> fix_bad_unicode("Všetko") 
'VÅ¡etko'

Answer 3 · 2021-08-26T00:09:07.000Z

Thanks for reporting! This is fixed in the upcoming release.