Python word.WordTokenizer类代码示例

    def test_tokenize_arabic_words(self):
        word_tokenizer = WordTokenizer('arabic')
        tests = ['اللُّغَةُ الْعَرَبِيَّةُ جَمِيلَةٌ.',
                 'انما الْمُؤْمِنُونَ اخوه فاصلحوا بَيْنَ اخويكم',
                 'الْعَجُزُ عَنِ الْإِدْرَاكِ إِدْرَاكٌ، وَالْبَحْثَ فِي ذاتِ اللَّه اشراك.',
                 'اللَّهُمُّ اُسْتُرْ عُيُوبَنَا وَأَحْسَنَ خَوَاتِيمَنَا الْكَاتِبِ: نَبِيلُ جلهوم',
                 'الرَّأْي قَبْلَ شَجَاعَة الشّجعَانِ',
                 'فَأَنْزَلْنَا مِنْ السَّمَاء مَاء فَأَسْقَيْنَاكُمُوهُ',
                 'سُئِلَ بَعْضُ الْكُتَّابِ عَنِ الْخَطّ، مَتَى يَسْتَحِقُّ أَنْ يُوصَفَ بِالْجَوْدَةِ ؟'
                ]

        results = []
        for test in tests:
            result = word_tokenizer.tokenize(test)
            results.append(result)

        target = [['اللُّغَةُ', 'الْعَرَبِيَّةُ', 'جَمِيلَةٌ', '.'],
                  ['انما', 'الْمُؤْمِنُونَ', 'اخوه', 'فاصلحوا', 'بَيْنَ', 'اخويكم'],
                  ['الْعَجُزُ', 'عَنِ', 'الْإِدْرَاكِ', 'إِدْرَاكٌ', '،', 'وَالْبَحْثَ', 'فِي', 'ذاتِ', 'اللَّه', 'اشراك', '.'],  # pylint: disable=line-too-long
                  ['اللَّهُمُّ', 'اُسْتُرْ', 'عُيُوبَنَا', 'وَأَحْسَنَ', 'خَوَاتِيمَنَا', 'الْكَاتِبِ', ':', 'نَبِيلُ', 'جلهوم'],  # pylint: disable=line-too-long
                  ['الرَّأْي', 'قَبْلَ', 'شَجَاعَة', 'الشّجعَانِ'],
                  ['فَأَنْزَلْنَا', 'مِنْ', 'السَّمَاء', 'مَاء', 'فَأَسْقَيْنَاكُمُوهُ'],
                  ['سُئِلَ', 'بَعْضُ', 'الْكُتَّابِ', 'عَنِ', 'الْخَطّ', '،', 'مَتَى', 'يَسْتَحِقُّ', 'أَنْ', 'يُوصَفَ', 'بِالْجَوْدَةِ', '؟']  # pylint: disable=line-too-long
                 ]
        self.assertEqual(results, target)

def stemmer_middle_high_german(text_l, rem_umlauts = True, exceptions = exc_dict):
	"""text_l: text in string format
	   rem_umlauts: choose whether to remove umlauts from string
	   exceptions: hard-coded dictionary for the cases the algorithm fails"""
	
	#Normalize text
	text_l = normalize_middle_high_german(text_l, to_lower_all = False, to_lower_beginning = True)
	
	#Tokenize text
	word_tokenizer = WordTokenizer("middle_high_german")
	text_l = word_tokenizer.tokenize(text_l)
	text = []

	
	for word in text_l:
		try:
			text.append(exceptions[word]) #test if word in exception dictionary
			
		except:
			if word[0].isupper():
				#MHG only uses upper case for locations, people, etc. So any word that starts with a capital
				#letter while not being at the start of a sentence will automatically be excluded.
				text.append(word)
				
			elif word in MHG_STOPS: 
				text.append(word) #Filter stop words
				
			else:
				text.append(stem_helper(word, rem_umlaut = rem_umlauts))
	return text

    def test_latin_word_tokenizer_base(self):
        """Test Latin-specific word tokenizer."""
        word_tokenizer = WordTokenizer('latin')

        #Test sources:
        # - V. Aen. 1.1
        # - Prop. 2.5.1-2
        # - Ov. Am. 1.8.65-66
        # - Cic. Phillip. 13.14
        # - Plaut. Capt. 937
        # - Lucr. DRN. 5.1351-53
        # - Plaut. Bacch. 837-38
        # - Plaut. Amph. 823
        # - Caes. Bel. 6.29.2

        tests = ['Arma virumque cano, Troiae qui primus ab oris.',
                    'Hoc verumst, tota te ferri, Cynthia, Roma, et non ignota vivere nequitia?',
                    'Nec te decipiant veteres circum atria cerae. Tolle tuos tecum, pauper amator, avos!',
                    'Neque enim, quod quisque potest, id ei licet, nec, si non obstatur, propterea etiam permittitur.',
                    'Quid opust verbis? lingua nullast qua negem quidquid roges.',
                    'Textile post ferrumst, quia ferro tela paratur, nec ratione alia possunt tam levia gigni insilia ac fusi, radii, scapique sonantes.',  # pylint: disable=line-too-long
                    'Dic sodes mihi, bellan videtur specie mulier?',
                    'Cenavin ego heri in navi in portu Persico?',
                    'quae ripas Ubiorum contingebat in longitudinem pedum ducentorum rescindit']

        results = []

        for test in tests:
            result = word_tokenizer.tokenize(test)
            results.append(result)

        target = [['Arma', 'virumque', 'cano', ',', 'Troiae', 'qui', 'primus', 'ab', 'oris', '.'], ['Hoc', 'verumst', ',', 'tota', 'te', 'ferri', ',', 'Cynthia', ',', 'Roma', ',', 'et', 'non', 'ignota', 'vivere', 'nequitia', '?'], ['Nec', 'te', 'decipiant', 'veteres', 'circum', 'atria', 'cerae.', 'Tolle', 'tuos', 'tecum', ',', 'pauper', 'amator', ',', 'avos', '!'], ['Neque', 'enim', ',', 'quod', 'quisque', 'potest', ',', 'id', 'ei', 'licet', ',', 'nec', ',', 'si', 'non', 'obstatur', ',', 'propterea', 'etiam', 'permittitur', '.'], ['Quid', 'opust', 'verbis', '?', 'lingua', 'nullast', 'qua', 'negem', 'quidquid', 'roges', '.'], ['Textile', 'post', 'ferrumst', ',', 'quia', 'ferro', 'tela', 'paratur', ',', 'nec', 'ratione', 'alia', 'possunt', 'tam', 'levia', 'gigni', 'insilia', 'ac', 'fusi', ',', 'radii', ',', 'scapique', 'sonantes', '.'], ['Dic', 'sodes', 'mihi', ',', 'bellan', 'videtur', 'specie', 'mulier', '?'], ['Cenavin', 'ego', 'heri', 'in', 'navi', 'in', 'portu', 'Persico', '?'], ['quae', 'ripas', 'Ubiorum', 'contingebat', 'in', 'longitudinem', 'pedum', 'ducentorum', 'rescindit']]

        self.assertEqual(results, target)

 def test_middle_english_tokenizer(self):
     text = "    Fers am I ferd of oure fare;\n Fle we ful fast þer-fore. \n Can Y no cownsel bot care.\n\n"
     target = ['Fers', 'am', 'I', 'ferd', 'of', 'oure', 'fare', ';', 'Fle', 'we', 'ful', 'fast', 'þer', '-', 'fore', '.',
               'Can', 'Y', 'no', 'cownsel', 'bot', 'care', '.']
     tokenizer = WordTokenizer('middle_english')
     tokenized = tokenizer.tokenize(text)
     self.assertTrue(tokenized == target)

 def test_latin_word_tokenizer(self):
     """Test Latin-specific word tokenizer."""
     word_tokenizer = WordTokenizer('latin')
     text = 'atque haec abuterque nihil'
     tokens = word_tokenizer.tokenize(text)
     target = ['atque', 'haec', 'abuter', '-que', 'nihil']
     self.assertEqual(tokens, target)

 def test_latin_word_tokenizer(self):
     """Test Latin-specific word tokenizer."""
     word_tokenizer = WordTokenizer('latin')
     
     #Test sources:
     # - V. Aen. 1.1
     # - Prop. 2.5.1-2
     # - Ov. Am. 1.8.65-66
     # - Cic. Phillip. 13.14
     
     tests = ['Arma virumque cano, Troiae qui primus ab oris.',
                 'Hoc verumst, tota te ferri, Cynthia, Roma, et non ignota vivere nequitia?',
                 'Nec te decipiant veteres circum atria cerae. Tolle tuos tecum, pauper amator, avos!',
                 'Neque enim, quod quisque potest, id ei licet, nec, si non obstatur, propterea etiam permittitur.']
     
     results = []
     
     for test in tests:
         result = word_tokenizer.tokenize(test)
         results.append(result)
                 
     target = [['Arma', 'que', 'virum', 'cano', ',', 'Troiae', 'qui', 'primus', 'ab', 'oris.'],
                 ['Hoc', 'verum', 'est', ',', 'tota', 'te', 'ferri', ',', 'Cynthia', ',', 'Roma', ',', 'et', 'non', 'ignota', 'vivere', 'nequitia', '?'],
                 ['Nec', 'te', 'decipiant', 'veteres', 'circum', 'atria', 'cerae.', 'Tolle', 'tuos', 'cum', 'te', ',', 'pauper', 'amator', ',', 'avos', '!'],
                 ['que', 'Ne', 'enim', ',', 'quod', 'quisque', 'potest', ',', 'id', 'ei', 'licet', ',', 'c', 'ne', ',', 'si', 'non', 'obstatur', ',', 'propterea', 'etiam', 'permittitur.']]
                 
     self.assertEqual(results, target)

 def test_old_norse_word_tokenizer(self):
     text = "Gylfi konungr var maðr vitr ok fjölkunnigr. " \
            "Hann undraðist þat mjök, er ásafólk var svá kunnigt, at allir hlutir gengu at vilja þeira."
     target = ['Gylfi', 'konungr', 'var', 'maðr', 'vitr', 'ok', 'fjölkunnigr', '.', 'Hann', 'undraðist', 'þat',
               'mjök', ',', 'er', 'ásafólk', 'var', 'svá', 'kunnigt', ',', 'at', 'allir', 'hlutir', 'gengu', 'at',
               'vilja', 'þeira', '.']
     word_tokenizer = WordTokenizer('old_norse')
     result = word_tokenizer.tokenize(text)
     self.assertTrue(result == target)

 def test_french_lemmatizer(self):
     text = "Li rois pense que par folie, Sire Tristran, vos aie amé ; Mais Dé plevis ma loiauté, Qui sor mon cors mete flaele, S'onques fors cil qui m’ot pucele Out m'amistié encor nul jor !"
     text = str.lower(text)
     tokenizer = WordTokenizer('french')
     lemmatizer = LemmaReplacer()
     tokens = tokenizer.tokenize(text)
     lemmas = lemmatizer.lemmatize(tokens)
     target = [('li', 'li'), ('rois', 'rois'), ('pense', 'pense'), ('que', 'que'), ('par', 'par'), ('folie', 'folie'), (',', ['PUNK']), ('sire', 'sire'), ('tristran', 'None'), (',', ['PUNK']), ('vos', 'vos'), ('aie', ['avoir']), ('amé', 'amer'), (';', ['PUNK']), ('mais', 'mais'), ('dé', 'dé'), ('plevis', 'plevir'), ('ma', 'ma'), ('loiauté', 'loiauté'), (',', ['PUNK']), ('qui', 'qui'), ('sor', 'sor'), ('mon', 'mon'), ('cors', 'cors'), ('mete', 'mete'), ('flaele', 'flaele'), (',', ['PUNK']), ("s'", "s'"), ('onques', 'onques'), ('fors', 'fors'), ('cil', 'cil'), ('qui', 'qui'), ("m'", "m'"), ('ot', 'ot'), ('pucele', 'pucele'), ('out', ['avoir']), ("m'", "m'"), ('amistié', 'amistié'), ('encor', 'encor'), ('nul', 'nul'), ('jor', 'jor'), ('!', ['PUNK'])]
     self.assertEqual(lemmas, target)

    def test_middle_high_german_stopwords(self):
        """Test filtering  Middle High German stopwords."""

        sentence = "Swer was ze Bêârosche komn, doch hete Gâwân dâ genomn den prîs ze bêder sît al ein wan daz dervor ein ritter schein, bî rôtem wâpen unrekant, des prîs man in die hœhe bant."
        lowered = sentence.lower()
        tokenizer = WordTokenizer('middle_high_german')
        tokens = tokenizer.tokenize(lowered)
        no_stops = [w for w in tokens if w not in MHG_STOPS]
        target_list = ['swer', 'bêârosche', 'komn', ',', 'gâwân', 'genomn', 'prîs', 'bêder', 'sît', 'dervor', 'ritter', 'schein', ',', 'rôtem', 'wâpen', 'unrekant', ',', 'prîs', 'hœhe', 'bant', '.']
        self.assertEqual(no_stops,target_list)

 def test_akkadian_sign_tokenizer(self):
     """
     Tests sign_tokenizer.
     """
     tokenizer = WordTokenizer('akkadian')
     word = ("{gisz}isz-pur-ram", "akkadian")
     output = tokenizer.tokenize_sign(word)
     goal = [("gisz", "determinative"), ("isz", "akkadian"),
             ("pur", "akkadian"), ("ram", "akkadian")]
     self.assertEqual(output, goal)

 def test_akkadian_word_tokenizer(self):
     """
     Tests word_tokenizer.
     """
     tokenizer = WordTokenizer('akkadian')
     line = 'u2-wa-a-ru at-ta e2-kal2-la-ka _e2_-ka wu-e-er'
     output = tokenizer.tokenize(line)
     goal = [('u2-wa-a-ru', 'akkadian'), ('at-ta', 'akkadian'),
             ('e2-kal2-la-ka', 'akkadian'),
             ('_e2_-ka', 'sumerian'), ('wu-e-er', 'akkadian')]
     self.assertEqual(output, goal)

    def test_middle_high_german_tokenize(self):
        """
        Test tokenizing Middle High German
        """
        word_tokenizer = WordTokenizer('middle_high_german')
        text = "Mīn ougen   wurden liebes alsō vol, \n\n\ndō ich die minneclīchen ērst gesach,\ndaȥ eȥ mir hiute und   iemer mē tuot wol."

        tokenized = word_tokenizer.tokenize(text)
        target = ['Mīn', 'ougen', 'wurden', 'liebes', 'alsō', 'vol', ',', 'dō', 'ich', 'die', 'minneclīchen', 'ērst', 'gesach', ',', 'daȥ', 'eȥ', 'mir', 'hiute', 'und', 'iemer', 'mē', 'tuot', 'wol', '.']

        self.assertEqual(tokenized, target)

def normalize_fr(string):
    string = string.lower()
    word_tokenizer = WordTokenizer('french')
    tokens = word_tokenizer.tokenize(string)
    normalized_text = []
    for token in tokens:
        for matches_rule, apply_rule in rules:
            if matches_rule(token):
                normalized = apply_rule(token)
                normalized_text.append(normalized)
    return normalized_text

 def test_bigram_pos_lemmatizer(self):
     train = [[('dixissem', 'dico', 'v')], [('de', 'de', 'r'), ('te', 'tu', 'p'), ('autem', 'autem', 'c'), (',', 'punc', 'u'), ('catilina', 'catilina', 'n'), (',', 'punc', 'u'), ('cum', 'cum2', 'c'), ('quiescunt', 'quiesco', 'v'), (',', 'punc', 'u'), ('probant', 'probo', 'v'), (',', 'punc', 'u'), ('cum', 'cum2', 'c'), ('patiuntur', 'patior', 'v'), (',', 'punc', 'u'), ('decernunt', 'decerno', 'v'), (',', 'punc', 'u'), ('cum', 'cum2', 'c'), ('tacent', 'taceo', 'v'), (',', 'punc', 'u'), ('clamant', 'clamo', 'v'), (',', 'punc', 'u'), ('neque', 'neque', 'c'), ('hi', 'hic', 'p'), ('solum', 'solus', 'd'), ('quorum', 'qui', 'p'), ('tibi', 'tu', 'p'), ('auctoritas', 'auctoritas', 'n'), ('est', 'sum', 'v'), ('uidelicet', 'uidelicet', 'd'), ('cara', 'carus', 'a'), (',', 'punc', 'u'), ('uita', 'uita', 'n'), ('uilissima', 'uilis', 'a'), (',', 'punc', 'u'), ('sed', 'sed', 'c'), ('etiam', 'etiam', 'c'), ('illi', 'ille', 'p'), ('equites', 'eques', 'n'), ('romani', 'romanus', 'a'), (',', 'punc', 'u'), ('honestissimi', 'honestus', 'a'), ('atque', 'atque', 'c'), ('optimi', 'bonus', 'a'), ('uiri', 'uir', 'n'), (',', 'punc', 'u'), ('ceteri', 'ceterus', 'a'), ('-que', '-que', 'c'), ('fortissimi', 'fortis', 'a'), ('ciues', 'ciuis', 'n'), ('qui', 'qui', 'p'), ('circumstant', 'circumsto', 'v'), ('senatum', 'senatus', 'n'), (',', 'punc', 'u'), ('quorum', 'qui', 'p'), ('tu', 'tu', 'p'), ('et', 'et', 'c'), ('frequentiam', 'frequentia', 'n'), ('uidere', 'uideo', 'v'), ('et', 'et', 'c'), ('studia', 'studium', 'n'), ('perspicere', 'perspicio', 'v'), ('et', 'et', 'c'), ('uoces', 'uox', 'n'), ('paulo', 'paulus', 'd'), ('ante', 'ante', 'd'), ('exaudire', 'exaudio', 'v'), ('potuisti', 'possum', 'v'), ('.', 'punc', 'u')]]
     lemmatizer = BigramPOSLemmatizer(train=train, include=['cum'])
     test_str = """Quod cum esset intellectum et animadversum fecit animo libentissimo populus Romanus"""
     target = [('quod', None), ('cum', 'cum2'), ('esset', None), ('intellectum', None), ('et', None), ('animaduersum', None), ('fecit', None), ('animo', None), ('libentissimo', None), ('populus', None), ('romanus', None)]  # pylint: disable=line-too-long
     jv_replacer = JVReplacer()
     tokenizer = WordTokenizer('latin')
     test_str = test_str.lower()
     test_str = jv_replacer.replace(test_str)
     tokens = tokenizer.tokenize(test_str)
     lemmas = lemmatizer.lemmatize(tokens)
     self.assertEqual(lemmas, target)

def stopwords_filter(string):

    text = string
    # strip tashkeel because the stop words list contains voweled words
    text = araby.strip_tashkeel(text)
    word_tokenizer = WordTokenizer("arabic")
    tokens = word_tokenizer.tokenize(text)

    # filter stop words
    no_stops = [w for w in tokens if w not in ARABIC_STOPS]

    return no_stops