Python/other/detecting_english_programmatically.py

61 lines
1.4 KiB
Python
Raw Normal View History

2016-08-02 17:46:55 +00:00
import os
2019-10-05 05:14:13 +00:00
UPPERLETTERS = "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
LETTERS_AND_SPACE = UPPERLETTERS + UPPERLETTERS.lower() + " \t\n"
2016-08-02 15:33:29 +00:00
def loadDictionary():
2016-08-02 17:46:55 +00:00
path = os.path.split(os.path.realpath(__file__))
2016-08-02 15:33:29 +00:00
englishWords = {}
2019-10-05 05:14:13 +00:00
with open(path[0] + "/dictionary.txt") as dictionaryFile:
for word in dictionaryFile.read().split("\n"):
englishWords[word] = None
2016-08-02 15:33:29 +00:00
return englishWords
2019-10-05 05:14:13 +00:00
2016-08-02 15:33:29 +00:00
ENGLISH_WORDS = loadDictionary()
2019-10-05 05:14:13 +00:00
2016-08-02 15:33:29 +00:00
def getEnglishCount(message):
message = message.upper()
message = removeNonLetters(message)
possibleWords = message.split()
if possibleWords == []:
return 0.0
matches = 0
for word in possibleWords:
if word in ENGLISH_WORDS:
matches += 1
return float(matches) / len(possibleWords)
2019-10-05 05:14:13 +00:00
2016-08-02 15:33:29 +00:00
def removeNonLetters(message):
lettersOnly = []
for symbol in message:
if symbol in LETTERS_AND_SPACE:
lettersOnly.append(symbol)
2019-10-05 05:14:13 +00:00
return "".join(lettersOnly)
2016-08-02 15:33:29 +00:00
2019-10-05 05:14:13 +00:00
def isEnglish(message, wordPercentage=20, letterPercentage=85):
2016-08-02 17:46:55 +00:00
"""
>>> isEnglish('Hello World')
True
>>> isEnglish('llold HorWd')
False
"""
2016-08-02 15:33:29 +00:00
wordsMatch = getEnglishCount(message) * 100 >= wordPercentage
numLetters = len(removeNonLetters(message))
messageLettersPercentage = (float(numLetters) / len(message)) * 100
lettersMatch = messageLettersPercentage >= letterPercentage
return wordsMatch and lettersMatch
2016-08-02 17:46:55 +00:00
import doctest
2019-10-05 05:14:13 +00:00
2016-08-02 17:46:55 +00:00
doctest.testmod()