import re
import os
import codecs
import csv
from os import walk

f = []
for (dirpath, dirnames, filenames) in walk(os.path.dirname(__file__)+"/news"):
    f.extend(filenames)
    break

se_pattern = re.compile(r"(?<=<se>)[\s\S\w\W]*?(?=</se>)")


file_toWrite = open("Sentence_number.txt", 'w', encoding='utf-8')
for path in f:
    file = codecs.open(os.path.dirname(__file__)+"/news/"+path, 'r', encoding='windows-1251')
    text = file.read()

    words_number = se_pattern.findall(text)
    file_toWrite.write(path + "\t" + str(len(words_number)) + "\n")

file_toWrite.close()


#Second task
with open('authors.csv', 'w',  encoding='utf-8') as csvfile:
    fieldnames = ['Название файла', 'Автор', 'Тематика текста']

    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
    writer.writeheader()

    author_pattern = re.compile(r"(?<=meta content=\").*?(?=\" name=\"author\")")
    topic_pattern = re.compile(r"(?<=meta content=\").*?(?=\" name=\"topic\")")

    for path in f:
        file = open(os.path.dirname(__file__) + "/news/" + path, 'r', encoding='windows-1251')
        text = file.read()

        author = author_pattern.search(text).group()
        topic = topic_pattern.search(text).group()
        writer.writerow({'Название файла' : path, 'Автор':author, 'Тематика текста':topic})