app.py

from sumy.parsers.plaintext import PlaintextParser
from sumy.parsers.html import HtmlParser
from sumy.nlp.tokenizers import Tokenizer
from sumy.nlp.stemmers import Stemmer
from sumy.utils import get_stop_words

import gradio as gr

import nltk
nltk.download('punkt')


def summarize(algorithm, language, sentence_count, input_type, input_):
  if algorithm == 'LSA':
    from sumy.summarizers.lsa import LsaSummarizer as Summarizer
  if algorithm =='text-rank':
    from sumy.summarizers.text_rank import TextRankSummarizer as Summarizer
  if algorithm =='lex-rank':
    from sumy.summarizers.lex_rank import LexRankSummarizer as Summarizer
  if algorithm =='edmundson':
    from sumy.summarizers.edmundson import EdmundsonSummarizer as Summarizer
  if algorithm =='luhn':
    from sumy.summarizers.luhn import LuhnSummarizer as Summarizer
  if algorithm =='kl-sum':
    from sumy.summarizers.kl import KLSummarizer as Summarizer
  if algorithm =='random':
    from sumy.summarizers.random import RandomSummarizer as Summarizer
  if algorithm =='reduction':
    from sumy.summarizers.reduction import ReductionSummarizer as Summarizer

  if input_type=="URL":
    parser = HtmlParser.from_url(input_, Tokenizer(language))
  if input_type=="text":
    parser = PlaintextParser.from_string(input_, Tokenizer(language))
    
  stemmer = Stemmer(language)
  summarizer = Summarizer(stemmer)
  stop_words = get_stop_words(language)

  if algorithm=='edmundson':
    summarizer.null_words = stop_words
    summarizer.bonus_words = parser.significant_words
    summarizer.stigma_words = parser.stigma_words
  else:
    summarizer.stop_words = stop_words

  summary_sentences = summarizer(parser.document, sentence_count)
  summary = ' '.join([str(sentence) for sentence in summary_sentences])
  
  return summary


iface = gr.Interface(
    summarize,
    [
      gr.inputs.Dropdown(["LSA", "luhn", "edmundson", "text-rank", "lex-rank", "random", "reduction", "kl-sum"]),
      gr.inputs.Textbox(1, default="english"),
      gr.inputs.Number(default=5),
      gr.inputs.Radio(["URL", "text"], default="URL"),
      gr.inputs.Textbox(5),
    ],
    "text",
    examples=[
        ["LSA", 'english', 5, "URL", "https://en.wikipedia.org/wiki/Automatic_summarization"]
    ],
)

iface.launch()