from flask import Flask, request, render_template
import pickle5 as pkl
from preprocessing import clean_text, clean_stopword, preprocess_text

import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
import re
import string

import pandas as pd
import networkx as nx
import matplotlib.pyplot as plt
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

nltk.download('stopwords')
nltk.download('punkt_tab')

# Initialize Flask app
app = Flask(__name__)

def process_input(text):
    prep_result = preprocess_text(text)
    kalimat = nltk.sent_tokenize(prep_result)
    tfidf_vectorizer = pkl.load(open('./tfidf_vectorizer.pkl', 'rb'))
    tfidf_preprocessing = pkl.load(open('./tfidf_preprocessing.pkl', 'rb'))
    terms = tfidf_vectorizer.get_feature_names_out()
    tfidf = pd.DataFrame(data=tfidf_preprocessing.toarray(), columns=terms)

    cossim = cosine_similarity(tfidf, tfidf)
    similarity_matrix = pd.DataFrame(cossim, 
                                     index=range(len(kalimat)), 
                                     columns=range(len(kalimat)))
    
    G = nx.DiGraph()
    for i in range(len(cossim)):
        G.add_node(i)

    for i in range(len(cossim)):
        for j in range(len(cossim)):
            similarity = cossim[i][j]
            if similarity > 0.1 and i != j:
                G.add_edge(i, j)

    closeness = nx.closeness_centrality(G)
    
    sorted_closeness = sorted(closeness.items(), key=lambda x: x[1], reverse=True)

    print(sorted_closeness)

    ringkasan_closeness = ""
    for node, closeness_preprocessing in sorted_closeness[:3]:
        top_sentence = kalimat[node]
        ringkasan_closeness += top_sentence + " "
        print(ringkasan_closeness)

    return top_sentence


@app.route('/', methods=['GET', 'POST'])
def summarize():
    result = None
    if request.method == 'POST':
        text = request.form['text']
        if text:
            result = process_input(text)
    return render_template('index.html', result=result)


if __name__ == '__main__':
    app.run(debug=True)