detecting_english_programmatically.py

importos
fromstringimportascii_letters

LETTERS_AND_SPACE=ascii_letters+" \t\n"


defload_dictionary() ->dict[str, None]:
path=os.path.split(os.path.realpath(__file__))
english_words: dict[str, None] = {}
withopen(path[0] +"/dictionary.txt") asdictionary_file:
forwordindictionary_file.read().split("\n"):
english_words[word] =None
returnenglish_words


ENGLISH_WORDS=load_dictionary()


defget_english_count(message: str) ->float:
message=message.upper()
message=remove_non_letters(message)
possible_words=message.split()
matches=len([wordforwordinpossible_wordsifwordinENGLISH_WORDS])
returnfloat(matches) /len(possible_words)


defremove_non_letters(message: str) ->str:
"""
 >>> remove_non_letters("Hi! how are you?")
 'Hi how are you'
 >>> remove_non_letters("P^y%t)h@o*n")
 'Python'
 >>> remove_non_letters("1+1=2")
 ''
 >>> remove_non_letters("www.google.com/")
 'wwwgooglecom'
 >>> remove_non_letters("")
 ''
 """
return"".join(symbolforsymbolinmessageifsymbolinLETTERS_AND_SPACE)


defis_english(
message: str, word_percentage: int=20, letter_percentage: int=85
) ->bool:
"""
 >>> is_english('Hello World')
 True
 >>> is_english('llold HorWd')
 False
 """
words_match=get_english_count(message) *100>=word_percentage
num_letters=len(remove_non_letters(message))
message_letters_percentage= (float(num_letters) /len(message)) *100
letters_match=message_letters_percentage>=letter_percentage
returnwords_matchandletters_match


if__name__=="__main__":
importdoctest

doctest.testmod()