# -*- coding: utf-8 -*-

import os

import codecs

from tagger import rfttag

import feature_extractors

import csv

import metrics

# the directory with text for training of classifier

input_root = "C:\\Users\\Askinkaty\\PycharmProjects\\TextReadability\\out_dir"

# function for tagging input txt file, parsing features and counting

def parse_text_by_tagger(input_text):

return values, grades

# the current file is run only if we build a matrix of training corpus saved in input directory

# the matrix of feature frequencies will be saved to out.csv

# you should choose which heading vector to write, it depends of the number of chosen features

# in the function parse_text_by_tagger(input_text)

if __name__ == '__main__':

with open('out.csv', 'w', newline='\n') as csvfile:

vectorwriter = csv.writer(csvfile)

# vectorwriter.writerow(

# ['id', 'first_person_pronoun', 'second_person_pronoun', 'third_person_pronoun',

# 'all_pronouns', 'finite_verbs', 'modifiers', 'past_tense', 'perf_aspect', 'present_tense',

# 'total_adverb', 'nominalization', 'nouns', 'genitive', 'neuter', 'passive', 'infin',

# 'speech_verb', 'mental_verb', 'that_compl', 'wh_relative',

# 'total_PP', 'word_length', 'word_complexity', 'text_spans', 'sentence_length', 'type_token_ratio',

# 'verbal_adverbs', 'passive_participial_clauses', 'active_participial_clauses',

# 'imperative_mood', 'predicative_adjectives', 'attributive_adjective',

# 'causative_subordinate', 'concessive_subordinate', 'conditional_subordinate',

# 'purpose_subordinate', 'conditional_mood', 'modal_possibility', 'modal_necessity',

# 'evaluative_vocabulary', 'academic vocabulary', 'parenthesis_attitude_evaluation', 'animate_nouns',

# 'parenthesis_accentuation', 'parenthesis_relation', 'degree_adverb', 'particles',

# 'numeral', 'top_100_nouns', 'non_top_100_nouns', 'top_1000_nouns_minus_head',

# 'non_top_1000_nouns_minus_head', 'top_100_verbs', 'non_top_100_verbs', 'top_1000_verbs_minus_head',

# 'non_top_1000_verbs_minus_head',

# 'top_100', 'top_300', 'top_500', 'top_10000', 'top_5000', 'complex_endings',

# 'first_second_person_verb', 'fk_grade',

# 'fk_grade_flex', 'cl_grade', 'smog_grade', 'dale_grade', 'ari_index', 'complexity_grade'])

vectorwriter.writerow (["first_person_pronoun","second_person_pronoun","all_pronouns","finite_verbs",

"modifiers","total_adverb","nominalization","genitive","passive","speech_verb",

"mental_verb","text_spans","sentence_length","type_token_ratio","verbal_adverbs",

"passive_participial_clauses","active_participial_clauses","imperative_mood",

"causative_subordinate","academic.vocabulary","parenthesis_attitude_evaluation",

"parenthesis_accentuation","degree_adverb","particles","top_1000_nouns_minus_head",

"top_100_verbs","top_5000","complex_endings","first_second_person_verb"])

for name in os.listdir(input_root):

filename = os.path.join(input_root, name)

with codecs.open(filename, mode='r', encoding='utf-8') as f:

text = f.read()

vector = parse_text_by_tagger(text)

name = os.path.basename(filename)

vector.insert(0, name.split(".")[0])

vectorwriter.writerow(vector)

csvfile.flush()