import pandas as pd
import networkx as nx
import re
from collections import defaultdict
import math
from pyvis.network import Network

[..................]

# Функция для извлечения пар по заданным правилам
def extract_term_pairs(text):
    pairs = []
    words = [word for word in re.split(r'\s+', str(text)) if word and word not in stop_words]
    
    for i in range(len(words)-1):
        w1, w2 = words[i], words[i+1]
        
        if len(w1) < 3 or len(w2) < 3:
            continue
            
        pattern1 = w1 in adjectives and w2 in nouns     # прил + сущ
        pattern2 = w1 in nouns and w2 in nouns          # сущ + сущ
        pattern3 = w1 in verbs and w2 in nouns          # гл + сущ
        pattern4 = w1 in verbs and w2 in adverbs        # гл + нар
        
        if pattern1 or pattern2 or pattern3 or pattern4:
            pairs.append((w1, w2))
    
    return pairs

# Создание графа
G = nx.Graph()

# Добавление центрального узла "Гербовый щит"
G.add_node('Гербовый щит', size=50, color='#FF5733', group='central', title='Гербовый щит', shape='star')

from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics.pairwise import cosine_similarity

from typing import List
from nltk.stem import SnowballStemmer
import re

# Функция стемминга
snowball = SnowballStemmer(language="russian")

def apply_stemming_ru(text: str) -> str:
    text = text.lower()
    text = re.sub(r"[^а-яё ]", " ", text)
    x = list(set([snowball.stem(t) for t in text.split()]))
    return ' '.join(x)

# Загрузка стоп-слов из файла
def load_stopwords_from_file(filepath: str) -> List[str]:
    with open(filepath, 'r', encoding='utf-8') as f:
        stopwords = [line.strip().lower() for line in f if line.strip()]
    return stopwords

# Вычисление косинусного расстояния
def compute_stem_cosine_distance_ru(reference: str, candidate: str, stop_words: List[str]) -> float:
    vectorizer = CountVectorizer(
        stop_words=stop_words,
        analyzer='word',
    )
    stem_matrix = vectorizer.fit_transform([reference, candidate])
    sim = cosine_similarity(stem_matrix[0:1], stem_matrix[1:2])[0][0]
    return sim