# cleaning the texts

import pandas as pd
import re
import nltk
from nltk.corpus import stopwords
from nltk.stem.porter import PorterStemmer

nltk.download('stopwords')

def clean_text(training_data, testing_data):

  ps = PorterStemmer()
  def portstem(review):
    return [ps.stem(word) for word in review if not word in set(stopwords.words('english'))]

  new_data = pd.concat( [training_data, testing_data] )

  processed_reviews = new_data['Review'].apply(lambda x: re.sub('[^a-zA-Z]', ' ', x))
  processed_reviews = processed_reviews.map(lambda x: x.lower())
  processed_reviews = processed_reviews.map(lambda x: x.split())
  processed_reviews = processed_reviews.map(portstem)
  processed_reviews = processed_reviews.map(lambda x: ' '.join(x))

  training_corpus = processed_reviews[:len(training_data)]
  testing_corpus = processed_reviews[len(training_data):]

  return training_corpus, testing_corpus