app.py

import streamlit as st
import sqlite3
import hashlib
import spacy
from nltk.corpus import stopwords
import re
import docx2txt
import PyPDF2
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

DB_PATH = 'resume_assistant.db'

# Initialize NLP and stop words
nlp = spacy.load("en_core_web_sm")
stop_words = set(stopwords.words("english"))

def initialize_db():
    with sqlite3.connect(DB_PATH) as conn:
        cur = conn.cursor()
        cur.execute('''
            CREATE TABLE IF NOT EXISTS users (
                id INTEGER PRIMARY KEY AUTOINCREMENT,
                username TEXT UNIQUE,
                email TEXT UNIQUE,
                password_hash TEXT
            )
        ''')
        conn.commit()

def hash_password(password):
    return hashlib.sha256(password.encode('utf-8')).hexdigest()

def signup_user(username, email, password):
    pwd_hash = hash_password(password)
    try:
        with sqlite3.connect(DB_PATH) as conn:
            cur = conn.cursor()
            cur.execute(
                "INSERT INTO users (username, email, password_hash) VALUES (?, ?, ?)",
                (username, email, pwd_hash)
            )
            conn.commit()
        return True, "Signup successful!"
    except sqlite3.IntegrityError:
        return False, "Username or email already exists."

def verify_login(username, password):
    pwd_hash = hash_password(password)
    with sqlite3.connect(DB_PATH) as conn:
        cur = conn.cursor()
        cur.execute("SELECT id FROM users WHERE username=? AND password_hash=?", (username, pwd_hash))
        data = cur.fetchone()
        return (True, data[0]) if data else (False, None)

def extract_text(file):
    if not file:
        return ""
    if file.type == "application/pdf":
        reader = PyPDF2.PdfReader(file)
        text = ""
        for page in reader.pages:
            text += page.extract_text() or ""
        return text
    elif file.type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
        return docx2txt.process(file)
    elif file.type == "text/plain":
        return file.getvalue().decode("utf-8")
    return ""

def clean_text(text):
    text = re.sub(r'[^a-zA-Z\s]', '', text.lower())
    doc = nlp(text)
    return set([token.lemma_ for token in doc if token.is_alpha and token.text not in stop_words and len(token.text)>2])

def ats_score(resume_text, jd_text):
    vectorizer = TfidfVectorizer(stop_words='english')
    tfidf = vectorizer.fit_transform([resume_text, jd_text])
    return cosine_similarity(tfidf[0:1], tfidf[1:2])[0]*100

def ml_score_dummy(resume_text, jd_text):
    # Dummy relevance score (just scaled ATS score)
    return ats_score(resume_text, jd_text) * 0.9

def main():
    initialize_db()
    st.title("AI-Powered Resume Assistant")

    menu = ["Login", "Signup"]
    choice = st.sidebar.selectbox("Choose Action", menu)

    if choice == "Signup":
        st.subheader("Create New Account")
        username = st.text_input("Username")
        email = st.text_input("Email")
        password = st.text_input("Password", type="password")
        if st.button("Sign Up"):
            if username and email and password:
                success, msg = signup_user(username, email, password)
                if success:
                    st.success(msg)
                else:
                    st.error(msg)
            else:
                st.warning("Please enter all fields")

    elif choice == "Login":
        st.subheader("Login")
        username = st.text_input("Username")
        password = st.text_input("Password", type="password")
        if st.button("Login"):
            if username and password:
                valid, user_id = verify_login(username, password)
                if valid:
                    st.success(f"Welcome, {username}!")
                    
                    resume_file = st.file_uploader("Upload Resume (.pdf, .docx, .txt)", type=['pdf','docx','txt'])
                    jd_text = st.text_area("Paste Job Description here", height=150)

                    if st.button("Analyze Resume"):
                        resume_text = extract_text(resume_file)
                        if not resume_text:
                            st.error("Failed to extract resume text.")
                        elif not jd_text.strip():
                            st.warning("Please provide job description.")
                        else:
                            ats = ats_score(resume_text, jd_text)
                            ml = ml_score_dummy(resume_text, jd_text)
                            resume_skills = clean_text(resume_text)
                            jd_skills = clean_text(jd_text)
                            matched = sorted(resume_skills & jd_skills)
                            missing = sorted(jd_skills - resume_skills)

                            st.markdown(f"**ATS Score:** {ats:.2f}%")
                            st.markdown(f"**ML Relevance Score:** {ml:.2f}%")
                            st.markdown(f"**Matched Skills:** {', '.join(matched) if matched else 'None'}")
                            st.markdown(f"**Missing Skills:** {', '.join(missing) if missing else 'None'}")
                else:
                    st.error("Invalid username or password")
            else:
                st.warning("Please enter username and password")

if __name__ == "__main__":
    main()