up/plagiat_1_full.v2.py

# Проверка одной со всеми из readme (для диф.зачетов)

import os
from difflib import SequenceMatcher
from tqdm import tqdm
import datetime
import requests

# download stopwords corpus, you need to run it once
import nltk
#nltk.download("stopwords")
from nltk.corpus import stopwords
import pymorphy2
from string import punctuation

from tqdm import tqdm

# ------------------------------- НАСТРОЙКИ ------------
# директория файла (на уровень выше, для структуры репозиториев 2 сем. 2022-23)
BASE_DIR = os.path.abspath(os.path.dirname(os.path.dirname(__file__)))
# проверяемая директория
# LECTION_DIR = os.path.join("ISRPO", "Лекции")
LECTION_DIR = os.path.join("EASvZI", "Лекции")
# LECTION_DIR = os.path.join("TZI", "Лекции", "ПМ3.2")

# ссылка для проверки
url = "http://213.155.192.79:3001/u20lebed/ISRPO/raw/88f913f3d1af9ee9bed0468b183ec9d4f6e5f2b4/2022-23/%d0%ad%d0%ba%d0%b7%d0%b0%d0%bc%d0%b5%d0%bd/Lebedkin.md"

# ------------------------------- / НАСТРОЙКИ ------------













#Create lemmatizer and stopwords list
morph = pymorphy2.MorphAnalyzer()
russian_stopwords = stopwords.words("russian")

#Preprocess function
def preprocess_text(text):
    translator = str.maketrans(punctuation, ' '*len(punctuation))
    words = text.translate(translator)
    words = words.lower().split()
    
    # очистка от прилегающего к слову мусора (слово, "или так")
    clear_words = []
    for word in words:
        clear_word = ""
        for s in word:
            if not s in punctuation:
                clear_word = clear_word + s
        clear_words.append(clear_word)
    tokens = []
    tokens = [morph.parse(token)[0].normal_form for token in clear_words if token not in russian_stopwords\
            and token != " " \
            and token.strip() not in punctuation \
            ]

    text = " ".join(tokens)    
    return tokens, text






print()

now = datetime.datetime.now().strftime('%d-%m-%Y %H:%M')
out_str = f"Время проверки: {now} \n"
# print(out_str)

response = requests.get(url)
post_html = response.text

post_tokens, post_uniq_text = preprocess_text(post_html)
print(f"количество уникальных слов: {len(set(post_tokens))}")
print()


post_list = post_html.split("\n")
# ищем другие лекции
readme_path = os.path.join(BASE_DIR, LECTION_DIR, "README.md")
try:
    with open(readme_path, encoding="utf-8") as f:
        readme_html = f.read()
except:
    with open(readme_path, encoding="cp1251") as f:
        readme_html = f.read()

min_ratio = 100
min_ratio_file = ""

readme_list = readme_html.split("\n")
for readme_str in tqdm(readme_list):
    if '[' in readme_str:
        # print(f"проверяю {readme_str}")
        variant_name, t = readme_str.split("]")
        variant_name = variant_name.strip("[")
        t, variant_uri = readme_str.split("(")
        variant_uri = variant_uri.replace("),", "")
        variant_uri = variant_uri.replace(")", "")
        variant_uri = variant_uri.strip()
        
        variant_path = os.path.join(BASE_DIR, LECTION_DIR, variant_uri)
        try:
            with open(variant_path, encoding="utf-8") as f:
                variant_html = f.read()
        except:
            with open(variant_path, encoding="cp1251") as f:
                variant_html = f.read()

        variant_tokens, variant_uniq_text = preprocess_text(variant_html)
        # print(f"количество уникальных слов варианта: {len(set(variant_tokens))}")

        # пересечение множеств 
        min_tokens_len = min([len(set(post_tokens)), len(set(variant_tokens))])
        c = list(set(post_tokens) & set(variant_tokens))
        ratio = (1 - (len(c) / min_tokens_len)) * 100
        if ratio < min_ratio:
            min_ratio = ratio
            min_ratio_file = variant_path

        # print(f"количество совпадающих слов: {len(c)} / {ratio:.2f}%")
        # print()


print()
print(f"min_ratio: {min_ratio}%")
print(f"min_ratio_file: {min_ratio_file}")
print("success")