app.py

import gradio as gr

import nltk
import simplemma
from nltk.tokenize import word_tokenize
from nltk.tokenize import sent_tokenize
from nltk.probability import FreqDist
from simplemma import text_lemmatizer
nltk.download('punkt')

file = "text.txt"

def get_lists(file):
  with open(file, 'r', encoding='utf-8') as f:
    text = f.read()

  sent_tokenized_text = sent_tokenize(text, language='italian')
  sent_tokenized_text_lower = [sent.lower() for sent in sent_tokenized_text]

  return sent_tokenized_text, sent_tokenized_text_lower

sentences, sentences_lower = get_lists(file)

def search_engine(target):

  result = []
  for i,sent in enumerate(sentences_lower):
    if target.lower() in sent:
      result.append(sentences[i])

  if len(result) == 0:
    return f"Non ho trovato la parola '{target}' nei testi.\n"
  
  else:
    return f"""Ho trovato {len(result)} {"frasi" if len(result) > 1 else "frase"} in cui è presente la parola {target}.\n {s for s in result}"""

demo = gr.Interface(fn=search_engine, inputs='text', outputs='text')
demo.launch()

'''
def numerical(number):
    return int(number)
    
def show_results(result, number):
    display = []
    try:
      for num,sent in enumerate(result[1][:int(number)]):
        display.append(f"{num+1}: {sent}\n")
      return display
    except:
      for num,sent in enumerate(result):
        display.append(f"{num+1}: {sent}")
      return display
    

demo = gr.Blocks()

with demo:
    target = gr.Textbox("Inserisci la parola da cercare")
    b1 = gr.Button()
    b2 = gr.Button()
    result = b2.click(bot, inputs=result, outputs='text')
    
    b1.click(search_engine, inputs=target, outputs='text')

    
    number = numerical(gr.Textbox())
    b2 = gr.Button("Quanti esempi vuoi vedere?")
    b2.click(upper, inputs=[result, number], outputs=f"""{s for s in display}""")

demo.launch()
'''
'''
demo = gr.Interface([fn=sentence, inputs='text', outputs='text'], [n=upper, inputs=cerca_una_parola, outputs='text'])
demo.launch()


demo = gr.Interface(
    sentence_builder,
    [
        gr.Textbox(),
        gr.Radio(["park", "zoo", "road"]),
        gr.CheckboxGroup(["ran", "swam", "ate", "slept"]),
        gr.Checkbox(label="Is it the morning?"),
    ],
    "text",
    examples=[
        ["cats", "park", ["ran", "swam"], True],
        ["dog", "zoo", ["ate", "swam"], False],
        ["bird", "road", ["ran"], False],
        ["cat", "zoo", ["ate"], True],
    ],
)


'''


'''
def update(name):
    return f"Welcome to Gradio, {name}!"

def search_engine_bot(target):

  result = []
  for i,sent in enumerate(sentences_lower):
    if target.lower() in sent:
      result.append(sentences[i])

  if len(result) == 0:
    return (f"Non ho trovato la parola '{target}' nei testi.\n", result)
  
  else:
    return (f"""Ho trovato {len(result)} {"frasi" if len(result) > 1 else "frase"} in cui è presente la parola {target}.\n""", result)


with gr.Blocks() as demo:
    gr.Markdown("Inerisci la parola da cercare.")
    with gr.Row():
        target = gr.Textbox(placeholder="What is your name?")
        out = gr.Textbox()
    btn = gr.Button("Run")
    btn.click(fn=search_engine_bot, inputs=target, outputs=out)

demo.launch()


file = "text.txt"

def get_lists(file):
  with open(file, 'r', encoding='utf-8') as f:
    text = f.read()

  word_tokenized_text = word_tokenize(text, language='italian')
  word_tokenized_text_lower = [word.lower() for word in word_tokenized_text]

  sent_tokenized_text = sent_tokenize(text, language='italian')
  sent_tokenized_text_lower = [sent.lower() for sent in sent_tokenized_text]

  return word_tokenized_text, word_tokenized_text_lower, sent_tokenized_text, sent_tokenized_text_lower

words, words_lower, SENTENCES, SENTENCES_LOWER = get_lists(file)


 
def num_sentences(show=''):
  try:
    number = int(show)
    return number
  except:
    number = ''
    return number

def show_results(result, number):
    display = []
    try:
      for num,sent in enumerate(result[1][:int(number)]):
        display.append(f"{num+1}: {sent}\n")
      return display
    except:
      for num,sent in enumerate(result):
        display.append(f"{num+1}: {sent}")
      return display

def search_engine(parola_da_cercare):
  sentences_lower = sentences_lower
  sentences = sentences
  target= parola_da_cercare #input("Inserisci una o più parole da cercare.\n")
  #target= input("Inserisci un'altra parola da cercare.") #input("Inserisci un'altra parola da cercare.\n")


  result = []
  for i,sent in enumerate(sentences_lower):
    if target.lower() in sent:
      result.append(sentences[i])

  if len(result) == 0:
    return(f"Non ho trovato la parola '{target}' nei testi.\n")
    #search_engine(round=True)
  
  else:
    return(f"""Ho trovato {len(result)} {"frasi" if len(result) > 1 else "frase"} in cui è presente la parola {target}.\n""")
    show = gr.Number(label="Quante frasi vuoi vedere? Scrivi un numero oppure 'tutte', se vuoi vederle tutte :-) \n", show_label=True)

    try:
      for num,sent in enumerate(result[:int(show)]):
        print(f"{num+1}: {sent}")
    except:
      for num,sent in enumerate(result):
        print(f"{num+1}: {sent}")

def sentence_builder(cerca_una_parola, place, activity_list, morning):
    return f"""The {cerca_una_parola}s went to the {place} where they {" and ".join(activity_list)} until the {"morning" if morning else "night"}"""

demo = gr.Interface(search_engine, inputs="text", outputs="text")

demo = gr.Interface(
    sentence_builder,
    [
        gr.Textbox(),
        gr.Radio(["park", "zoo", "road"]),
        gr.CheckboxGroup(["ran", "swam", "ate", "slept"]),
        gr.Checkbox(label="Is it the morning?"),
    ],
    "text",
    examples=[
        ["cats", "park", ["ran", "swam"], True],
        ["dog", "zoo", ["ate", "swam"], False],
        ["bird", "road", ["ran"], False],
        ["cat", "zoo", ["ate"], True],
    ],
)


if __name__ == "__main__":
    demo.launch()
'''