コード例 #1
0
def concat_number(sentence):
    """
    concatenate numbers with '+'
    Input=sentence                                     Output=sentence
    """

    # init
    i = 0

    sentence = delete_and_from_number(sentence)

    while i < len(sentence):
        # There is a number
        if other_functions.number(sentence[i]) == 1:
            begin_pos = i

            while i < len(sentence) and other_functions.number(
                    sentence[i]) == 1:
                i += 1
            end_pos = i

            #We have to concatenate the last number if it is superlative
            if i < len(sentence) and other_functions.number(sentence[i]) == 2:
                end_pos += 1
            if i < len(sentence) and sentence[i] in ResourcePool(
            ).adjective_numbers:
                end_pos += 1

            sentence = sentence[:begin_pos] + [
                other_functions.convert_to_string(sentence[begin_pos:end_pos])
            ] + sentence[end_pos:]

        i += 1
    return sentence
コード例 #2
0
ファイル: preprocessing.py プロジェクト: Kuew/dialogs
def concat_number(sentence):
    """
    concatenate numbers with '+'
    Input=sentence                                     Output=sentence
    """

    #init
    i = 0

    sentence = delete_and_from_number(sentence)

    while i < len(sentence):
        #There is a number
        if other_functions.number(sentence[i]) == 1:
            begin_pos = i

            while i < len(sentence) and other_functions.number(sentence[i]) == 1:
                i += 1
            end_pos = i

            #We have to concatenate the last number if it is superlative
            if i < len(sentence) and other_functions.number(sentence[i]) == 2:
                end_pos += 1
            if i < len(sentence) and sentence[i] in ResourcePool().adjective_numbers:
                end_pos += 1

            sentence = sentence[:begin_pos] + [
                other_functions.convert_to_string(sentence[begin_pos:end_pos])] + sentence[end_pos:]

        i += 1
    return sentence
コード例 #3
0
ファイル: preprocessing.py プロジェクト: Kuew/dialogs
def delete_and_from_number(sentence):
    """
    delete 'and' between two numbers
    Input=sentence                                     Output=sentence
    """

    #init
    i = 0

    while i < len(sentence):
        if sentence[i] == 'and' and other_functions.number(sentence[i - 1]) == 1 and other_functions.number(
                sentence[i + 1]) == 1:
            sentence = sentence[:i] + sentence[i + 1:]
        i += 1
    return sentence
コード例 #4
0
ファイル: preprocessing.py プロジェクト: Kuew/dialogs
def reorganize_adj(sentence):
    """
    deletes ',' and 'and' if it is between adjectives
    Input=sentence                              Output=sentence
    """

    #init
    i = 0

    while i < len(sentence) - 1:
        if sentence[i] == ',' or sentence[i] == 'and':
            if analyse_nominal_group.is_an_adj(sentence[i + 1]) and analyse_nominal_group.is_an_adj(sentence[i - 1]):
                if other_functions.number(sentence[i + 1]) == 0 and other_functions.number(sentence[i - 1]) == 0:
                    sentence = sentence[:i] + sentence[i + 1:]
        i += 1
    return sentence
コード例 #5
0
def find_the_plural(sentence, position):
    """ Finds if there is a plural and add 'a'

    :param sentence and position of nominal group
    :return: the position of plural or -1
    """

    if len(sentence) - position - 1 < 0:
        return -1

    #It is a number the word is not a plural
    if other_functions.number(sentence[position]) == 1:
        return -1

    #If it is proposal we continue
    if sentence[position] in ResourcePool().proposals:
        return find_the_plural(sentence, position + 1)

    if sentence[position] in ResourcePool().nouns_end_s:
        return -1

    #If it is adjective we continue
    if is_an_adj(sentence[position]):
        if find_the_plural(sentence, position + 1) != -1:
            return position

    #If it is an  adjective ends with 's'
    if sentence[0].endswith("'s") or sentence[position].endswith("ous"):
        return -1

    #we have plural if the noun ends with 's'
    if find_sn_pos(sentence, position) == [] and sentence[position].endswith('s'):
        return position

    return -1
コード例 #6
0
def delete_and_from_number(sentence):
    """
    delete 'and' between two numbers
    Input=sentence                                     Output=sentence
    """

    # init
    i = 0

    while i < len(sentence):
        if sentence[i] == 'and' and other_functions.number(
                sentence[i - 1]) == 1 and other_functions.number(
                    sentence[i + 1]) == 1:
            sentence = sentence[:i] + sentence[i + 1:]
        i += 1
    return sentence
コード例 #7
0
def is_an_adj(word):
    """Determines if a word is an adjective

    :param string word: a word

    :return: True if the word is recognized as an adjective, False else. 
    """

    #It is a noun verb pronoun or determinant so we have to return False
    if word in ResourcePool().special_nouns + ResourcePool(
    ).special_verbs + ResourcePool().pronouns + ResourcePool().determinants:
        return False

    #For the regular adjectives
    for k in ResourcePool().adjective_rules:
        if word.endswith(k):
            return True

    #For adjectives created from numbers
    if word.endswith('th') and other_functions.number(word) == 2:
        return True

    #We use the irregular adjectives list to find it
    if word in list(ResourcePool().adjectives.keys()) + ResourcePool(
    ).adjective_numbers + ResourcePool().adj_quantifiers:
        return True

    return False
コード例 #8
0
def is_an_adj(word):
    """Determines if a word is an adjective

    :param string word: a word

    :return: True if the word is recognized as an adjective, False else. 
    """

    #It is a noun verb pronoun or determinant so we have to return False
    if word in ResourcePool().special_nouns + ResourcePool().special_verbs + ResourcePool().pronouns + ResourcePool().determinants:
        return False

    #For the regular adjectives
    for k in ResourcePool().adjective_rules:
        if word.endswith(k):
            return True

    #For adjectives created from numbers
    if word.endswith('th') and other_functions.number(word) == 2:
        return True

    #We use the irregular adjectives list to find it
    if word in list(
            ResourcePool().adjectives.keys()) + ResourcePool().adjective_numbers + ResourcePool().adj_quantifiers:
        return True

    return False
コード例 #9
0
def reorganize_adj(sentence):
    """
    deletes ',' and 'and' if it is between adjectives
    Input=sentence                              Output=sentence
    """

    # init
    i = 0

    while i < len(sentence) - 1:
        if sentence[i] == ',' or sentence[i] == 'and':
            if analyse_nominal_group.is_an_adj(
                    sentence[i + 1]) and analyse_nominal_group.is_an_adj(
                        sentence[i - 1]):
                if other_functions.number(
                        sentence[i + 1]) == 0 and other_functions.number(
                            sentence[i - 1]) == 0:
                    sentence = sentence[:i] + sentence[i + 1:]
        i += 1
    return sentence
コード例 #10
0
def convert_adj_to_digit(adj_list):
    """
    returns the list of adjectives after change number to digit                           
    :param the adjective                            
    :return: the adjective            
    """

    for i in adj_list:
        if i.endswith('th') and other_functions.number(i) == 2:
            adj_list[adj_list.index(i)] = other_functions.convert_to_digit(i) + 'th'

    return adj_list
コード例 #11
0
def convert_adj_to_digit(adj_list):
    """
    returns the list of adjectives after change number to digit                           
    :param the adjective                            
    :return: the adjective            
    """

    for i in adj_list:
        if i.endswith('th') and other_functions.number(i) == 2:
            adj_list[adj_list.index(
                i)] = other_functions.convert_to_digit(i) + 'th'

    return adj_list
コード例 #12
0
def find_sn(sentence):
    """
    Returns the first nominal group found in the sentence.                              
    
    :param list sentence: the sentence as a list of words                                             
    :return: the nominal group                                                        
    """

    nb_position = 1

    #If sentence is empty
    if not sentence:
        return []

    for x in sentence:
        #If there is a pronoun
        if x in ResourcePool().pronouns:
            return [sentence[sentence.index(x)]]

        #If there is a nominal group with determinant
        if x in ResourcePool().determinants:
            nb_position += adjective_pos(sentence, sentence.index(x) + 1)
            return sentence[sentence.index(x):sentence.index(x) + nb_position]

        #If we have 'something'
        for k in ResourcePool().composed_nouns:
            if x.startswith(k):
                if x in ResourcePool().noun_not_composed:
                    return []
                return [sentence[sentence.index(x)]]

        #If there is a number, it will be the same with determinant
        if other_functions.number(x) == 1:
            nb_position += adjective_pos(sentence, sentence.index(x) + 1)
            return sentence[sentence.index(x):sentence.index(x) + nb_position]

        #If there is a proper name
        counter = sentence.index(x)
        while counter < len(sentence) and other_functions.find_cap_lettre(
                sentence[counter]) == 1:
            counter += 1
            #Not equal => there is a proper name
        if counter != sentence.index(x):
            return sentence[sentence.index(x):counter]

    #Default case
    return []
コード例 #13
0
def find_sn(sentence):
    """
    Returns the first nominal group found in the sentence.                              
    
    :param list sentence: the sentence as a list of words                                             
    :return: the nominal group                                                        
    """

    nb_position = 1

    #If sentence is empty
    if not sentence:
        return []

    for x in sentence:
        #If there is a pronoun
        if x in ResourcePool().pronouns:
            return [sentence[sentence.index(x)]]

        #If there is a nominal group with determinant
        if x in ResourcePool().determinants:
            nb_position += adjective_pos(sentence, sentence.index(x) + 1)
            return sentence[sentence.index(x): sentence.index(x) + nb_position]

        #If we have 'something'
        for k in ResourcePool().composed_nouns:
            if x.startswith(k):
                if x in ResourcePool().noun_not_composed:
                    return []
                return [sentence[sentence.index(x)]]

        #If there is a number, it will be the same with determinant
        if other_functions.number(x) == 1:
            nb_position += adjective_pos(sentence, sentence.index(x) + 1)
            return sentence[sentence.index(x): sentence.index(x) + nb_position]

        #If there is a proper name
        counter = sentence.index(x)
        while counter < len(sentence) and other_functions.find_cap_lettre(sentence[counter]) == 1:
            counter += 1
            #Not equal => there is a proper name
        if counter != sentence.index(x):
            return sentence[sentence.index(x): counter]

    #Default case
    return []
コード例 #14
0
ファイル: preprocessing.py プロジェクト: Kuew/dialogs
def am_pm(sentence):
    """
    This function separate 'am' or 'pm' to the digit
    Input=sentence                            Output=sentence
    """

    #init
    i = 0

    while i < len(sentence):
        if sentence[i].endswith('am') or sentence[i].endswith('pm'):
            if other_functions.number(sentence[i]) == 1:
                sentence = sentence[:i] + [sentence[i][:len(sentence[i]) - 2]] + [
                    sentence[i][len(sentence[i]) - 2:]] + sentence[i + 1:]
                i += 1
        i += 1

    return sentence
コード例 #15
0
def find_sn_pos(sentence, begin_pos):
    """
    We will find the nominal group which is in a known position                      
    We have to use adjective_pos to return the end position of nominal group         

    :param list sentence: the sentence (list of strings)
    :param begin_pos:the position of the nominal group       
    :return: the nominal group (as a list of words)
    """

    if begin_pos >= len(sentence):
        return []

    end_pos = 1

    #If it is a pronoun
    if sentence[begin_pos] in ResourcePool().pronouns:
        return [sentence[begin_pos]]

    #If there is a nominal group with determinant
    if sentence[begin_pos] in ResourcePool().determinants:
        end_pos += adjective_pos(sentence, begin_pos + 1)
        return sentence[begin_pos:end_pos + begin_pos]

    #If we have 'something'
    for k in ResourcePool().composed_nouns:
        if sentence[begin_pos].startswith(k):
            if sentence[begin_pos] in ResourcePool().noun_not_composed:
                return []
            return [sentence[begin_pos]]

            #If there is a number, it will be the same with determinant
    if other_functions.number(sentence[begin_pos]) == 1:
        end_pos += adjective_pos(sentence, begin_pos + 1)
        return sentence[begin_pos:end_pos + begin_pos]

    #If it is a proper name
    counter = begin_pos
    while counter < len(sentence) and other_functions.find_cap_lettre(
            sentence[counter]) == 1:
        counter += 1

    #Default case return [] => ok if counter=begin_pos
    return sentence[begin_pos:counter]
コード例 #16
0
def find_sn_pos(sentence, begin_pos):
    """
    We will find the nominal group which is in a known position                      
    We have to use adjective_pos to return the end position of nominal group         

    :param list sentence: the sentence (list of strings)
    :param begin_pos:the position of the nominal group       
    :return: the nominal group (as a list of words)
    """

    if begin_pos >= len(sentence):
        return []

    end_pos = 1

    #If it is a pronoun
    if sentence[begin_pos] in ResourcePool().pronouns:
        return [sentence[begin_pos]]

    #If there is a nominal group with determinant
    if sentence[begin_pos] in ResourcePool().determinants:
        end_pos += adjective_pos(sentence, begin_pos + 1)
        return sentence[begin_pos: end_pos + begin_pos]

    #If we have 'something'
    for k in ResourcePool().composed_nouns:
        if sentence[begin_pos].startswith(k):
            if sentence[begin_pos] in ResourcePool().noun_not_composed:
                return []
            return [sentence[begin_pos]]

            #If there is a number, it will be the same with determinant
    if other_functions.number(sentence[begin_pos]) == 1:
        end_pos += adjective_pos(sentence, begin_pos + 1)
        return sentence[begin_pos: end_pos + begin_pos]

    #If it is a proper name
    counter = begin_pos
    while counter < len(sentence) and other_functions.find_cap_lettre(sentence[counter]) == 1:
        counter += 1

    #Default case return [] => ok if counter=begin_pos
    return sentence[begin_pos: counter]
コード例 #17
0
def am_pm(sentence):
    """
    This function separate 'am' or 'pm' to the digit
    Input=sentence                            Output=sentence
    """

    # init
    i = 0

    while i < len(sentence):
        if sentence[i].endswith('am') or sentence[i].endswith('pm'):
            if other_functions.number(sentence[i]) == 1:
                sentence = sentence[:i] + [
                    sentence[i][:len(sentence[i]) - 2]
                ] + [sentence[i][len(sentence[i]) - 2:]] + sentence[i + 1:]
                i += 1
        i += 1

    return sentence
コード例 #18
0
def return_det(nominal_group):
    """
    This function returns the determinant of the nominal group                       
    :param nominal_group: the nominal group
    :return: the determinant          
    """

    #nominal_group is empty
    if not nominal_group:
        return []

    #We return the first element of the list
    if nominal_group[0] in ResourcePool().determinants:
        return [nominal_group[0]]

    #If there is a number
    if other_functions.number(nominal_group[0]) == 1:
        return [nominal_group[0]]

    #Default case
    return []
コード例 #19
0
def return_det(nominal_group):
    """
    This function returns the determinant of the nominal group                       
    :param nominal_group: the nominal group
    :return: the determinant          
    """

    #nominal_group is empty
    if not nominal_group:
        return []

    #We return the first element of the list
    if nominal_group[0] in ResourcePool().determinants:
        return [nominal_group[0]]

    #If there is a number
    if other_functions.number(nominal_group[0]) == 1:
        return [nominal_group[0]]

    #Default case
    return []
コード例 #20
0
def adjective_pos(sentence, word_pos):
    """
    Returns the position of the end of the nominal group                
    We have to use the list of irregular adjectives                                  
    
    :param list sentence: the sentence (list of strings)
    :param word_pos: the position of the first adjective    
    :return: the position of the last word of the nominal group                       
    """

    #If it is the end of the sentence
    if len(sentence) - 1 <= word_pos:
        return 1

    #The case of '2 of them'
    if sentence[word_pos] == 'of':
        return 0

    #It is a noun so we have to return 1
    if sentence[word_pos] in ResourcePool().special_nouns:
        return 1

    #For the regular adjectives
    for k in ResourcePool().adjective_rules:
        if sentence[word_pos].endswith(k):
            return 1 + adjective_pos(sentence, word_pos + 1)

    #For adjectives created from numbers
    if sentence[word_pos].endswith('th') and other_functions.number(
            sentence[word_pos]) == 2:
        return 1 + adjective_pos(sentence, word_pos + 1)

    #We use the irregular adjectives list to find it
    if sentence[word_pos] in list(ResourcePool().adjectives.keys(
    )) + ResourcePool().adjective_numbers + ResourcePool().adj_quantifiers:
        return 1 + adjective_pos(sentence, word_pos + 1)

    #Default case
    return 1
コード例 #21
0
def adjective_pos(sentence, word_pos):
    """
    Returns the position of the end of the nominal group                
    We have to use the list of irregular adjectives                                  
    
    :param list sentence: the sentence (list of strings)
    :param word_pos: the position of the first adjective    
    :return: the position of the last word of the nominal group                       
    """

    #If it is the end of the sentence
    if len(sentence) - 1 <= word_pos:
        return 1

    #The case of '2 of them'
    if sentence[word_pos] == 'of':
        return 0

    #It is a noun so we have to return 1
    if sentence[word_pos] in ResourcePool().special_nouns:
        return 1

    #For the regular adjectives
    for k in ResourcePool().adjective_rules:
        if sentence[word_pos].endswith(k):
            return 1 + adjective_pos(sentence, word_pos + 1)

    #For adjectives created from numbers
    if sentence[word_pos].endswith('th') and other_functions.number(sentence[word_pos]) == 2:
        return 1 + adjective_pos(sentence, word_pos + 1)

    #We use the irregular adjectives list to find it
    if sentence[word_pos] in list(
            ResourcePool().adjectives.keys()) + ResourcePool().adjective_numbers + ResourcePool().adj_quantifiers:
        return 1 + adjective_pos(sentence, word_pos + 1)

    #Default case
    return 1
コード例 #22
0
def find_the_plural(sentence, position):
    """ Finds if there is a plural and add 'a'

    :param sentence and position of nominal group
    :return: the position of plural or -1
    """

    if len(sentence) - position - 1 < 0:
        return -1

    #It is a number the word is not a plural
    if other_functions.number(sentence[position]) == 1:
        return -1

    #If it is proposal we continue
    if sentence[position] in ResourcePool().proposals:
        return find_the_plural(sentence, position + 1)

    if sentence[position] in ResourcePool().nouns_end_s:
        return -1

    #If it is adjective we continue
    if is_an_adj(sentence[position]):
        if find_the_plural(sentence, position + 1) != -1:
            return position

    #If it is an  adjective ends with 's'
    if sentence[0].endswith("'s") or sentence[position].endswith("ous"):
        return -1

    #we have plural if the noun ends with 's'
    if find_sn_pos(sentence,
                   position) == [] and sentence[position].endswith('s'):
        return position

    return -1
コード例 #23
0
def recover_quantifier(nom_gr):
    """
    recovers the quantifier and put the noun in singular form if it's in plural
    Input=nominal group class                     Output=nominal group class
    """

    #init
    flg = 0

    #The default case is 'ONE'
    if not nom_gr.det:
        #If the noun is 'anything' => SOME 
        if nom_gr.noun and nom_gr.noun[0] in ["anything"]:
            nom_gr._quantifier = 'SOME'
            #If the noun is 'everything' => ALL
        if nom_gr.noun and nom_gr.noun[0] in ["everything"]:
            nom_gr._quantifier = 'ALL'
            #If the noun starts with 'any' => we have 'all'
        if nom_gr.noun and nom_gr.noun[0].startswith('any'):
            nom_gr._quantifier = 'SOME'
            #If the noun starts with 'no' => we have 'none'
        if nom_gr.noun and nom_gr.noun[0].startswith('no'):
            nom_gr._quantifier = 'NONE'

    else:

        #If it is a number
        if other_functions.number(nom_gr.det[0]) == 1:
            nom_gr._quantifier = 'DIGIT'
            nom_gr.det = [other_functions.convert_to_digit(nom_gr.det[0])]

        #Here we will use the quantifier list
        for i in ResourcePool().det_quantifiers:
            if i[0] == nom_gr.det[0]:
                nom_gr._quantifier = i[1]

        #If we have a plural
        if nom_gr.noun != [] and nom_gr.noun[0].endswith('s'):
            for x in ResourcePool().nouns_end_s:
                if x == nom_gr.noun[0]:
                    #It is a noun singular with 's' at the end
                    flg = 1
                    break
            if flg == 0:
                #We delete determinant added in processing with his quantifier
                if nom_gr.det[0] == 'a':
                    nom_gr.det = []
                    nom_gr._quantifier = 'ONE'
                elif nom_gr.det[0] == 'no':
                    nom_gr._quantifier = 'ANY'

                #We have to put the noun in singular form
                for y in ResourcePool().plural_nouns:
                    #If it is an irregular noun
                    if y[0] == nom_gr.noun[0]:
                        nom_gr.noun[0] = y[1]
                        if nom_gr._quantifier == 'ONE':
                            nom_gr._quantifier = 'ALL'
                        return nom_gr
                    #Else
                nom_gr.noun[0] = nom_gr.noun[0][:len(nom_gr.noun[0]) - 1]
                if nom_gr._quantifier == 'ONE':
                    nom_gr._quantifier = 'ALL'
                return nom_gr

        return nom_gr
コード例 #24
0
def upper_to_lower(sentence):
    """
    converts the upper case to lower case
    Input=sentence, beginning sentence list                  Output=sentence
    """

    # If the sentence begins with upper case
    if other_functions.find_cap_lettre(sentence[0]):

        # We convert upper case to lower case if it is not 'I'
        if sentence[0] == 'I':
            sentence = expand_contractions(sentence)
            return sentence
        else:
            sentence[0] = sentence[0][0].lower() + sentence[0][1:]

        #We make changes here because we need lower case and not upper case
        sentence = expand_contractions(sentence)
        sentence = adverbial_or_subsentence(sentence)
        stc = process_and_beginning_sentence(sentence)

        #If sentence is modified we can return it
        if stc != sentence:
            return stc

        #We find an action verb => it is an imperative sentence
        if sentence[0] in ThematicRolesDict().get_all_verbs():
            return sentence

        #If we find the word in the Beginning_sentence list
        for v in ResourcePool().sentence_starts:
            if sentence[0] == v[0]:
                return sentence

        #We find a number
        if other_functions.number(sentence[0]) == 1:
            return sentence

        #If there is plural
        sentence = analyse_nominal_group.find_plural(sentence)
        #If it still start with adjective
        if analyse_nominal_group.is_an_adj(sentence[0]) == 1:
            sentence = ['the'] + sentence

        #If there is a nominal group
        if analyse_nominal_group.find_sn_pos(sentence, 0):
            return sentence

        #Default case: we assume a proper name, we convert lowercase to uppercase
        sentence[0] = sentence[0][0].upper() + sentence[0][1:]

    # If the sentence begins with lower case
    else:
        #We make changes here because we need lower case and not upper case
        sentence = expand_contractions(sentence)
        sentence = adverbial_or_subsentence(sentence)
        sentence = process_and_beginning_sentence(sentence)

        #If we find the word in the Beginning_sentence list so we can return it
        for v in ResourcePool().sentence_starts:
            if sentence[0] == v[0]:
                return sentence

        #If there is plural
        sentence = analyse_nominal_group.find_plural(sentence)
        #If it still start with adjective
        if analyse_nominal_group.is_an_adj(sentence[0]) == 1:
            sentence = ['the'] + sentence

    return sentence
コード例 #25
0
ファイル: preprocessing.py プロジェクト: Kuew/dialogs
def upper_to_lower(sentence):
    """
    converts the upper case to lower case
    Input=sentence, beginning sentence list                  Output=sentence
    """

    #If the sentence begins with upper case
    if other_functions.find_cap_lettre(sentence[0]):

        #We convert upper case to lower case if it is not 'I'
        if sentence[0] == 'I':
            sentence = expand_contractions(sentence)
            return sentence
        else:
            sentence[0] = sentence[0][0].lower() + sentence[0][1:]

        #We make changes here because we need lower case and not upper case
        sentence = expand_contractions(sentence)
        sentence = adverbial_or_subsentence(sentence)
        stc = process_and_beginning_sentence(sentence)

        #If sentence is modified we can return it
        if stc != sentence:
            return stc

        #We find an action verb => it is an imperative sentence
        if sentence[0] in ThematicRolesDict().get_all_verbs():
            return sentence

        #If we find the word in the Beginning_sentence list
        for v in ResourcePool().sentence_starts:
            if sentence[0] == v[0]:
                return sentence

        #We find a number
        if other_functions.number(sentence[0]) == 1:
            return sentence

        #If there is plural
        sentence = analyse_nominal_group.find_plural(sentence)
        #If it still start with adjective
        if analyse_nominal_group.is_an_adj(sentence[0]) == 1:
            sentence = ['the'] + sentence

        #If there is a nominal group
        if analyse_nominal_group.find_sn_pos(sentence, 0):
            return sentence

        #Default case: we assume a proper name, we convert lowercase to uppercase
        sentence[0] = sentence[0][0].upper() + sentence[0][1:]

    #If the sentence begins with lower case
    else:
        #We make changes here because we need lower case and not upper case
        sentence = expand_contractions(sentence)
        sentence = adverbial_or_subsentence(sentence)
        sentence = process_and_beginning_sentence(sentence)

        #If we find the word in the Beginning_sentence list so we can return it
        for v in ResourcePool().sentence_starts:
            if sentence[0] == v[0]:
                return sentence

        #If there is plural
        sentence = analyse_nominal_group.find_plural(sentence)
        #If it still start with adjective
        if analyse_nominal_group.is_an_adj(sentence[0]) == 1:
            sentence = ['the'] + sentence

    return sentence
コード例 #26
0
def recover_quantifier(nom_gr):
    """
    recovers the quantifier and put the noun in singular form if it's in plural
    Input=nominal group class                     Output=nominal group class
    """

    #init
    flg = 0

    #The default case is 'ONE'
    if not nom_gr.det:
        #If the noun is 'anything' => SOME 
        if nom_gr.noun and nom_gr.noun[0] in ["anything"]:
            nom_gr._quantifier = 'SOME'
            #If the noun is 'everything' => ALL
        if nom_gr.noun and nom_gr.noun[0] in ["everything"]:
            nom_gr._quantifier = 'ALL'
            #If the noun starts with 'any' => we have 'all'
        if nom_gr.noun and nom_gr.noun[0].startswith('any'):
            nom_gr._quantifier = 'SOME'
            #If the noun starts with 'no' => we have 'none'
        if nom_gr.noun and nom_gr.noun[0].startswith('no'):
            nom_gr._quantifier = 'NONE'

    else:

        #If it is a number
        if other_functions.number(nom_gr.det[0]) == 1:
            nom_gr._quantifier = 'DIGIT'
            nom_gr.det = [other_functions.convert_to_digit(nom_gr.det[0])]

        #Here we will use the quantifier list
        for i in ResourcePool().det_quantifiers:
            if i[0] == nom_gr.det[0]:
                nom_gr._quantifier = i[1]

        #If we have a plural
        if nom_gr.noun != [] and nom_gr.noun[0].endswith('s'):
            for x in ResourcePool().nouns_end_s:
                if x == nom_gr.noun[0]:
                    #It is a noun singular with 's' at the end
                    flg = 1
                    break
            if flg == 0:
                #We delete determinant added in processing with his quantifier
                if nom_gr.det[0] == 'a':
                    nom_gr.det = []
                    nom_gr._quantifier = 'ONE'
                elif nom_gr.det[0] == 'no':
                    nom_gr._quantifier = 'ANY'

                #We have to put the noun in singular form
                for y in ResourcePool().plural_nouns:
                    #If it is an irregular noun
                    if y[0] == nom_gr.noun[0]:
                        nom_gr.noun[0] = y[1]
                        if nom_gr._quantifier == 'ONE':
                            nom_gr._quantifier = 'ALL'
                        return nom_gr
                    #Else
                nom_gr.noun[0] = nom_gr.noun[0][:len(nom_gr.noun[0]) - 1]
                if nom_gr._quantifier == 'ONE':
                    nom_gr._quantifier = 'ALL'
                return nom_gr

        return nom_gr