import streamlit as st
import pandas as pd
import numpy as np
import re
import pickle
import pdfminer
from pdfminer.high_level import extract_text
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Embedding, Conv1D, MaxPooling1D, LSTM, Dense, GlobalMaxPooling1D
from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.utils import to_categorical
from sklearn.preprocessing import LabelEncoder


def cleanResume(resumeText):
    # Your existing cleanResume function remains unchanged
    resumeText = re.sub('http\S+\s*', ' ', resumeText)
    resumeText = re.sub('RT|cc', ' ', resumeText)
    resumeText = re.sub('#\S+', '', resumeText)
    resumeText = re.sub('@\S+', '  ', resumeText)
    resumeText = re.sub('[%s]' % re.escape("""!"#$%&'()*+,-./:;<=>?@[\]^_`{|}~"""), ' ', resumeText)
    resumeText = re.sub(r'[^\x00-\x7f]',r' ', resumeText)
    resumeText = re.sub('\s+', ' ', resumeText)
    return resumeText

def pdf_to_text(file):
    # Use pdfminer.six to extract text from the PDF file
    text = extract_text(file)
    return text

def predict_category(resumes_data, selected_category):
    # Load the trained DeepRank model
    model = load_deeprank_model()

    # Process the resumes data
    resumes_df = pd.DataFrame(resumes_data)
    resumes_text = resumes_df['ResumeText'].values

    # Tokenize the text and convert to sequences
    tokenized_text = tokenizer.texts_to_sequences(resumes_text)

    # Pad sequences to have the same length
    max_sequence_length = 500  # Assuming maximum sequence length of 500 words
    padded_text = pad_sequences(tokenized_text, maxlen=max_sequence_length)

    # Make predictions
    predicted_probs = model.predict(padded_text)

    # Assign probabilities to respective job categories
    for i, category in enumerate(label.classes_):
        resumes_df[category] = predicted_probs[:, i]

    resumes_df_sorted = resumes_df.sort_values(by=selected_category, ascending=False)

    # Get the ranks for the selected category
    ranks = []
    for rank, (idx, row) in enumerate(resumes_df_sorted.iterrows()):
        rank = rank + 1
        file_name = row['FileName']
        ranks.append({'Rank': rank, 'FileName': file_name})

    return ranks

def load_deeprank_model():
    # Load the saved DeepRank model
    model = Sequential()
    # Add layers to the model (example architecture, adjust as needed)
    model.add(Embedding(input_dim=vocab_size, output_dim=128, input_length=max_sequence_length))
    model.add(Conv1D(filters=128, kernel_size=5, activation='relu'))
    model.add(MaxPooling1D(pool_size=2))
    model.add(LSTM(64))
    model.add(Dense(num_classes, activation='softmax'))
    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
    model.load_weights('deeprank_model.h5')  # Replace 'deeprank_model.h5' with your saved model file
    return model

def main():
    st.title("Resume Ranking App")
    st.text("Upload resumes and select a category to rank them.")

    resumes_data = []
    selected_category = ""

    # Handle multiple file uploads
    files = st.file_uploader("Upload resumes", type=["pdf"], accept_multiple_files=True)
    if files:
        for file in files:
            text = cleanResume(pdf_to_text(file))
            resumes_data.append({'ResumeText': text, 'FileName': file.name})
        selected_category = st.selectbox("Select a category to rank by", label.classes_)

    if st.button("Rank Resumes"):
        if not resumes_data or not selected_category:
            st.warning("Please upload resumes and select a category to continue.")
        else:
            ranks = predict_category(resumes_data, selected_category)
            st.write(pd.DataFrame(ranks))

if __name__ == '__main__':
    # Load label encoder and tokenizer
    df = pd.read_csv('UpdatedResumeDataSet.csv')
    df['cleaned'] = df['Resume'].apply(lambda x: cleanResume(x))
    label = LabelEncoder()
    df['Category'] = label.fit_transform(df['Category'])

    # Tokenize text and get vocabulary size and number of classes
    text = df['cleaned'].values
    tokenizer = Tokenizer()
    tokenizer.fit_on_texts(text)
    vocab_size = len(tokenizer.word_index) + 1
    num_classes = len(label.classes_)

    main()